Injecting prior to turbo comp' impellers

[Mart]

Hi there,

I've recently installed a charge-cooler system to my car, running ERL's 1s water injection system. I used to have an fmic with the water jet located on the outlet end-tank, which worked well, but because of the short pipe-run with the C.C (approx 8" from turbo to C.C inlet, approx 12" from C.C outlet to plenum), I'm undecided where to locate the jet now?

As I don't really want to insert the jet on the C.C outlet pipe, as I think this would be too near to the plenum (agree/disagree?), and it isn't really possible to have the jet on the C.C inlet pipe/turbo compressor outlet, I was wondering if it would be ok to have the jet/mist firing into the turbo compressor intake air-stream; ie, between the air-filter & compressor housing.

Do you think this idea is feasible? Has it been tried & tested before? (with good results?)

If so, which jet size would be a good starting point, and at what boost threshold level should the jet/pump be activating?

Apologies for all the questions/long-winded post.

Just for reference, the turbo being used is of the T28 variety running @ 25psi, and I have 0.4, 0.5, 0.6 and 0.7mm water jets to hand.

Thanks.

Mart

[Charged Performance]

Depending on your plenum and intake runner lengths 8-12" may be sufficient to get a good uniform water injection/air charge mixture.

There are a few posts around about injecting before the turbo compressor/impeller. Historically it has not been good for the blades. Some think improved materials along with improved jet atomization may have negated some of the prior experience. I think in general though you will not find many people willing to risk injecting in front of their turbo.

[bill Shurvinton]

Sorry to reawaken this, but I just want to put the record straight on this.

Pre-compressor injection offers a host of benefits. In a nutshell, injecting water (and ideally a few other miscible fluids of high specific heat capacity) pushes the compression from adiabatic to near isothermal. This is way more efficient (up to 30%). So you can reach the same boost for considerably less exhaust flow.

Yes atomisation is critical, but the technology is there.

[DuMaurier 7]

I have been spraying into the turbo for more than a year WITHOUT any problems . I am using an 0.9 nozzle and a system pressure of 100 psi , this produses a spray with very fine atomisation.

[JohnA]

... In a nutshell, injecting water (and ideally a few other miscible fluids of high specific heat capacity) pushes the compression from adiabatic to near isothermal. ....
Can you expand more on this mate?
I find it interesting... :)

[bill Shurvinton]

Quickly. Normal compression is adiabatic, which to a first order means that as the gas is compressed, it gets hot. Now this heat is one of the biggest problems with forced induction, for 2 reasons. Firstly you have to get rid of the heat, and secondly you need to take power out the turbine shaft to perform the heating.( heat is work and work is heat).

Now with the right level of water injection, the heat is removed before it builds up, pushing the compression closer to isothermal (not all the way, but closer). In round terms this is about 30% more efficient (less exhaust gas required for the same boost, or more boost at the same exhaust flow).

problem 1 is that water on its own, whilst having a very high latent heat of vapourisation, doesn't have that good a saturation partial pressure. You can only put so much in before the air is at 100%RH. Above this the water will not cool the air until you compress it in the engine. However, if you add a second fluid, say methanol, Dalton proved with his law of partial pressures that the methanol doesn't know that the air is saturated with water vapour and goes on to vapourise as well. Add a 3rd, such as acetone (all available chemicals) and you get a 3rd tranch of cooling.

The net result is that you may be able to cool the air to slightly below ambient with the right mix. So your compressor becomes more efficient, you can throw away the intercooler, increasing your flow, and in some cases significantly improve the flow from turbo to inlet.

If you leave the intercooler in, you are generally warming the air back up again, so you have to take the plunge and remove it to gain the best benefit.

Its good, very very good.

Bill

[JohnA]

Bill,

your theory sounds very exciting (to me, at least)

So to recoup, you're saying that injecting before the compressor actually increases the efficiency of the compressor - leading to less heat in the charge as it leaves the compressor.

You're also saying that injecting a water/methanol mix at that stage would be far better than just pure water, right?

I always thought that such a setup would decrease the intercooler's efficiency, compensating most of the benefits. You agree with this, to the point where you see the intercooler as a thermal liability

So what sort of injection volumes are we talking here?

[hotrod]

Yes spraying before the compressor impeller has its place.

As stated above, it improves the effeciency of the compressor. Many people tend ot look at WI from an engine centric point of view. If they cannot infer a direct benefit to the engine they assume it is a bad idea.

You are working with a complex system of mechanical devices that interact with each other in many ways. Even though on first blush injecting infront of the compressor or between the compressor and the intercooler might appear to be less effecient you need to account for ALL the interactions. In many cases we simply don't have enough information to predict the results so frequently experimentation will give you better data in a matter of minutes, than all the incomplete computer simulations you can afford.

Injection in front of the compressor accomplishes several things. A turbocharger is a constant pressure variable volume "DYNAMIC" compressor.

A turbocharger only knows 2 important properties of the gas it is compressing. The density of the gas at the compressor inlet and the pressure ratio it is operating at, which is determined by the rotor rpm and the gas density. If you increase the pressure or reduce the temperature at the inlet you will modify both of those parameters. In both cases (increased inlet pressure, or lower inlet temperature) you increase the apparent density of the gas passing through the compressor. At a given rotor rpm with a given gas density you will flow a very specific volume of gas and it will be compressed to a specific pressure ratio on exit. That is what the compressor map is based on. If you change the inlet conditions (gas density) you in effect slide the compressor map left and right. This is the "corrected flow" of the turbocharger.

By injecting water/alcohol ahead of the compressor two things happen. You cool the inlet air substantially, this in effect moves your true operating point to the left on the compressor map. (in most cases for max performance this is a good thing, although on some turbocharger conditions it can cause compressor surge.)

You also change the pressure temperature profile inside the compressor wheel itself. You probably actually change the shape of the compressor map. As the gas moves outward and is compressed, heat that would have gone into heat and increased pressure is absorbed by the WI mist and so the compressor has less work to do since it is no longer fighting this temperature driven pressure increase, it can achieve more mass flow at that pressure ratio. The cooling should also modify the speed of sound in the gas and the mach number of the compressor blade tips should also change. This should change the choke flow characteristics of the compressor but I don't have the information to comment in detail on that.

Net result is, you increase the mass flow through the compressor --- in effect you make it act like it is bigger than under normal conditions.

During WWII this was the way ADI (anti detonation injection --- the common term in the aircraft world for WI ) was set up on military aircraft in most cases. The water and the fuel was injected into the eye of the centrifugal supercharger.

Errosion of the compressor blades is not a problem if steps are taken to ensure the mist is very fine when it arrives at the compressor inlet so that it follows the airflow and does not imping on the blades with a high differential speed.

The ideal is to get drop size down as close to 10 microns as possible but due to the brief periods of use and intermittent nature of most WI systems, in reality you can live with larger drop sizes in real world systems.

If you have ever ridden a bicycle or motor cycle in a rain storm you know how sharp the impact of a large water dropplet can be, but a fog or drizzle will not cause the same painful experience because the droplets are small enough they are strongly influenced by the direction of flow of the air stream around you and impact with much less velocity and obviously also have lower momentum.

For people in hot dry climates that lose a lot of turbocharger performance in hot weather, pre-turbo injection should be looked at.

In regard to the effects on the compressor mass flow, maximum results appear to occur with mist flow of about 3% - 10% of the air mass, so you will likely need to inject additional WI near the throttle body to reach maximum detonation suppression and best power.

Larry

[JohnA]

...By injecting water/alcohol ahead of the compressor two things happen. You cool the inlet air substantially

If you mean that the ambient temp air flow is affected, I wouldn't expect that to be much. Say ambient is 20C, how much lower can it get by injecting mist of water that's 30C+ (the bottle is in the engine bay!). I suspect that after the pump the water is even warmer.
this in effect moves your true operating point to the left on the compressor map. (in most cases for max performance this is a good thing, although on some turbocharger conditions it can cause compressor surge.)
I would have thought that this 'efficiency gain' is the effect of water droplets inside the compressor blades, as they try to squeeze the air.
Am I right?
.You also change the pressure temperature profile inside the compressor wheel itself. You probably actually change the shape of the compressor map.
This one actually
... it can achieve more mass flow at that pressure ratio. The cooling should also modify the speed of sound in the gas and the mach number of the compressor blade tips should also change.
fascinating :!: ...

[hotrod]

If you mean that the ambient temp air flow is affected, I wouldn't expect that to be much. Say ambient is 20C, how much lower can it get by injecting mist of water that's 30C+ (the bottle is in the engine bay!). I suspect that after the pump the water is even warmer.
Quote:

Actually it can be very substantial. The cooling is by evaporation, and typically will exceed 10 - 20 deg C . The initial temp of the water is of little consequence. The cooling due to evaporation (latent heat of evaporation) is very much larger than the latent heat of the liquid water.
As you can see below the evaporation of a gram of water will absorb 542 x the heat energy required to lower that same quantity of liquid water one degree in temperature. That means that even if the water were nearly at boiling temps when injected, evaporative cooling would reduce the air flow temp below ambient temperatures long before all the water changed to vapor.

Specific heat capacity of liquid water - 4.187 kJ/kgK
Latent heat of evaporation - 2,270 kJ/kg

Since the evaporation of the water and the alcohol will essentially stop when the air becomes saturated (ie. 100% humidity) there is a practical limit to the cooling or around 20-30 deg C for alcohol water mixes, and in real systems you seldom get much more than about 15-20 deg C.

For every 5 deg C you drop the inlet air temp you will increase mass flow by about 1% due to density increases.


I would have thought that this 'efficiency gain' is the effect of water droplets inside the compressor blades, as they try to squeeze the air.
Am I right?

Only in a very crude sense. Your thinking in a mechanical piston pushes on air sort of way, but what happens inside the compressor impeller occurs at a molecular level.

Think of it this way If you could freeze frame time, and stop what was happening inside the impeller while its spinning at 120,000 rpm. Each impeller passage between adjacent pairs of compressor blades contains a wedge shaped parcel of air. When spinning at 120,000 rpm the air is subject to huge centrifugal forces as it moves away from the hub of the impeller and toward the rim of the compressor. The trapped air would like very much to be slung out of the impeller but like a crowd at a stadium after a match it simply cannot all get out as fast as it would like. As a result it stacks up (compresses) as it gets near the exit. In this process a lot of internal friction occurs. The air near the tips of the compressor might be moveing near the speed of sound at maximum flow, this heating makes the air try to expand. This increases the pressure which fights the outward movement of the air. Eventually a balance is achieved between the centrifugal forces trying to throw the air out of the impeller and the pressure build up due to the compression and the pressure build up due to the heating. The addition of the water mist removes a very large fraction of the pressure gain due to heating. As a result more air can exit the impeller over a given period of time, and more of the pressure gain is real compression rather than waste heat. The net result is a more isothermic compression which is always more effecient than an adiabatic compression.

Larry

[JohnA]

...In regard to the effects on the compressor mass flow, maximum results appear to occur with mist flow of about 3% - 10% of the air mass, ...
another 2 minor points:
1. a 2 litre engine at 1 bar boost and 5Krpm will be consuming around
10.000 litres of air per minute, or 10cubic metres of air, which is about 12.2 kg air/minute.
10% of that equates to 1.2kg water /min, that's quite a lot isn't it? With the aquamist pump and nozzles, it will have to be closer to 3% in practice.
Would this be enough?

2. Wouldn't the extreme centrifugal forces in the compressor force the tiny water droplets to become liquid again?

[SaabTuner]

...In regard to the effects on the compressor mass flow, maximum results appear to occur with mist flow of about 3% - 10% of the air mass, ...
another 2 minor points:
1. a 2 litre engine at 1 bar boost and 5Krpm will be consuming around
10.000 litres of air per minute, or 10cubic metres of air, which is about 12.2 kg air/minute.
10% of that equates to 1.2kg water /min, that's quite a lot isn't it? With the aquamist pump and nozzles, it will have to be closer to 3% in practice.
Would this be enough?

2. Wouldn't the extreme centrifugal forces in the compressor force the tiny water droplets to become liquid again?

A 2 liter engine consumes 2 liters every TWO revolutions (720 degrees) so assuming 100%VE, 100% intercooling, it would consume 5,000 liters/minute.

Adrian~

[JohnA]

..A 2 liter engine consumes 2 liters every TWO revolutions (720 degrees) so assuming 100%VE, 100% intercooling, it would consume 5,000 liters/minute.

Adrian~
you think you got me there on the two revolutions thing? :twisted:
Well you didn't, because I mention 1 bar boost pressure, which multiplies it by two.

Nice try though. :wink:

[hotrod]

Four cycle engines only intake stroke every other revolution for each cylinder ;)

JohnA -- actually its not all that much for two reasons. First your not on full boost for more than just a few seconds at a time unless your making a landspeed record run on the salt flats.

Also 1.2 kg/min is only about 20 cc/sec or about 2/3's of a shot glass of fluid, divided among the engines cylinders. On a 4 cylinder 2 liter engine with 500 cc/min injector running static will inject 2000 cc/min of fuel

The turbo and the air flow currents and heating through the rotor actually tend to shred and tear the dropplets apart and cause nearly explosive evaporation, very little liquid water exits the rotor.

If you want to have a base reference on how much water an engine can ingest during WWII Pratt and Whitney ran what they called flood tests on their aircraft engines where they turned up the water injection to the point that liquid water was pouring out the exhaust ports and the engines still ran with no problems. That level of injectant reduced the engine power from a max of 2000-2800 hp down to about 600 hp but there were no problems.

As far as the limits of the pump and injectors you are probably correct. That is why I am running a 100 psi Shurflo pump with a capacity of 1.4 gal/min ( 5.3 L/min) and a 5 Gal / hr nozzle ( 315 cc/min) on a rather small 13G turbocharger. The 13G only flows about 25 lb/min of air (11.45 kg/min) which comes out very close to 3 % of air flow for the spray flow.

I will probably be adding nozzles here soon.

Larry

[JohnA]

Four cycle engines only intake stroke every other revolution for each cylinder ;)
...and 4-cylinder, 4-stroke engines (inline) have one cylinder in intake stroke every half revolution...the end result is the same: every 2 rpm 2litres of air have been inhaled (assuming 100% VE of course)

But I'd like to believe that the level of discussion here is a bit higher :cool:


JohnA -- actually its not all that much for two reasons. First your not on full boost for more than just a few seconds at a time unless your making a landspeed record run on the salt flats.
Yeah, but when calculating (roughly ofcourse) the max water flow, we have to assume sustained full throttle.
And Volumetric Efficiency won't be 100% of course, especially at full-power revs.

The turbo and the air flow currents and heating through the rotor actually tend to shred and tear the dropplets apart and cause nearly explosive evaporation, very little liquid water exits the rotor.
explosive evaporation ... interesting. Didn't know it even exists! :lol:

It's OK for people with traditional superchargers to say that injecting all this water before the blower is fine. But I had reservations about centrifugal compressors. If the centrifugal forces don't mess up the mist, then that's one less headache.

So such a setup wouldn't be boost-driven like std water injection, would it?
To realise the gains of near iso-thermal compression we'd want to inject water from zero boost, yes?

Do you think that's safe then?
Unfortunately my aquamist setup is not boost-sensitive, it's rather on/off.
:cry:

[hotrod]

If the centrifugal forces don't mess up the mist, then that's one less headache.

Very little of the mist survives the path through the compressor in liquid form. In the discharge you have much cooler air at very high relative humidity but almost no water droplets.

Literally 10's of thousands of military aircraft engines in the period 1939 - through the early 1950's used before the compressor injection with no problems at all. It is also widely used in the tractor pulls, and on diesel trucks, CART race cars I understand, used a pre-compressor injection of some of their fuel for the same purpose, a handful of production cars and trucks have also used the system.

All you have to do is avoid large water droplets or a solid stream of water from impacting the impeller.

So such a setup wouldn't be boost-driven like std water injection, would it?
To realise the gains of near iso-thermal compression we'd want to inject water from zero boost, yes?

Depends on how you are defining boost driven --- some systems use a pressurized water reservoir so that boost pressure provides the motive force for injection. These systems are best for trucks pulling heavy loads up mountain passes where they stay on boost for extended periods of time. A presurize reservoir system will not react quick enough in my opinion for a high performance application.

Other systems use a pump supplied pressurized system with a valve of some sort that opens to allow spray at a certain boost pressure.

You generally do not want a continous on system (except for an rv or truck as mentioned above). On my system I have a pressure switch which activates at about 8 psi to turn on the spray. It is a simple on/off WI system. It will spray anytime boost is over the 8 psi threshold I have currently set.

As far as computing max water flow, your correct to establish a limiting case you would figure at max air flow which would typically be at max power rpm and a true engine VE of about .85-.88 for most engines.

However for detonation control your most critical flow rate is in the midrange max torque rpm range, say 3000-5500 rpm. That is where the engine has max VE and by definition you have max torque because you have max cylinder pressures. This is the RPM range where a sudden loss of WI can be fatal to the engine.

If you look at a manifold pressure log of a typical turbocharged engine when the turbo begins to make serious boost you go from near normal atmospheric pressure to a significant boost pressure >6 psi for example in just a fraction of a second ( ie only a 1000 rpm change in engine speed at WOT).

On my WRX on a 3rd gear WOT blast, I go from -1.2 psig manifold pressure at cruise, to 13.5 psig manifold pressure in 0.922 seconds when I just stab the throttle wide open, on the shift to 4th I go from 10.4 psig to -8.4 psig manifold pressure in 0.234 seconds as I close the throttle for the shift, and then jump to +5.9 psig .235 seconds later as I go back to WOT and another 235 milliseconds later (the time resolution of this log), I am backup to a manifold pressure of 11.9 psig. (this is all at 5800 ft altitude).

Total time under load in Low gear is about 1.5 seconds, 2rd gear was 3.828 seconds, 3rd gear 2.359 seconds, and 4th gear only lasts a couple of seconds before I'm over 100 mph.

So in a highway acceleration type situation, you would seldom see more than 7 - 8 seconds of flow, with brief interruptions of about .5 second every 2 - 3 seconds, so your actual duty cycle for the spray activation will be something like 80% or so. In my case I have an accumulator in line with the solenoid so the system pressure changes very little over this short interval.

If you are looking for detonation supression you need it most in the midrange rpms near the engines torque peak. That is when the engine is most likely to detonate under high engine load. You want the WI to kick in just below the rpm range and boost pressure that you can first experience detonation under heavy load. If you turn it on early you cool the exhaust gasses just as the turbo needs hot EGT's to help it spool up. So in that case you delay the WI spray as long as you safely can.

If your looking for maximum efficiency of the compressor, you want the spray to come on at a boost pressure when the compressor is just moving out of its maximum effeciency island on the compressor map. What you are doing is artificially stretching the compressor map to the right so the compressor has a wider island of efficiency.

In my case I will eventually be using a compound system with some spray pre-turbo for the purpose of maximising the compressors efficiency, and if I need it, a secondary jet spraying pre-throttle body to control any detonation that is not supressed by the pre-turbo injection.

Larry

[SaabTuner]

On advantage of injecting pre-compressor is that, should the WI fail, it would take several seconds for all the water to evaporate off the walls of all the tubing before the engine. This was noted by the folks at Linkoping University when doing experiments with WI and an Ionization Gap Sensor feedback ignition system.

If you have a very very good ECU on the engine a WI failure would be not much worse than running into a bad tank of fuel and the ECU should eliminate the knocking within one or two engine rotations. Most engines are designed to be able to handle rapid changes in octane even if they like higher better.

EGT might be another problem, but if you feel the car suddenly getting slower it doesn't take a rocket scientist to know something might be amiss and worth investigating.

Adrian~

[JohnA]

..In my case I will eventually be using a compound system with some spray pre-turbo for the purpose of maximising the compressors efficiency, and if I need it, a secondary jet spraying pre-throttle body to control any detonation that is not supressed by the pre-turbo injection.
That's easier to implement actually, I'll probably give it a try. Just Tee the existing WI line to another nozzle just before the compressor. The aquamist pump should keep atomisation at decent levels, eh?

[SaabTuner]

..A 2 liter engine consumes 2 liters every TWO revolutions (720 degrees) so assuming 100%VE, 100% intercooling, it would consume 5,000 liters/minute.

Adrian~
you think you got me there on the two revolutions thing? :twisted:
Well you didn't, because I mention 1 bar boost pressure, which multiplies it by two.

Nice try though. :wink:

Technically the engine consumes 5,000 L/min but at 29.4 psi instead of 14.7. :wink: But fair enough, I didn't notice the 1 bar bit.

Adrian~

[JohnA]

..Technically the engine consumes 5,000 L/min but at 29.4 psi instead of 14.7. :wink:
The exact psi figure depends on atmospheric pressure conditions, doesn't it? And assuming sea-level is a bit presumptious. :smile:

Tell you what - why don't we end this futile excercise in bandwidth wastage? :wink:

[wrx+]

You guys make my head hurt!

Observations from a mechanic that drag races are that injecting prior to the turbo produces positive effects in every instance. The ones I've seen that say injecting prior to the turbo creats pitting, are also the ones that run their cars with volocity stacks and NO air filter.

Kent

[b_boy]

I'm into trying this out. I've now read what I can on the pre-compressor experiments and it looks like if the injection is not continuous and droplet size is as small as possible, you'll have minimal wear on the impeller wheel.

I'll have the Element Hyday in my STI in about a week or two. I can program it to open a valve at 5000 or 6000 RPM at 15 psi of boost and let the Aquamist take over from there.

From my reading on "wet compression" or "fogging" as decribed in the turbine engine lingo it looks like 2-3% water(alcohol etc):air is about all you need or want to shift toward isothermal compression. Any thoughts out there on how much to inject?

I'm primarily interested in try to "stretch" out the compressor map at high RPM/boost when my stock turbo is starting to wheeze.

How about droplet size from an Aquamist jet? Hotrod mentioned 10 uM. That is pretty small on the absolute edge of visual perception of a single dot. The smallest cells are 10 uM and you can only see them if you know what to look for.

Some sort of atomizer like is used in "cold" humidifiers might be a better solution. They use a small apeture combined with some sort of high frequency osillation to break up the water. Or maybe an ultra sonic tip of some kind.

[hotrod]

The 10 micron drop size would be ideal, but appears to not really be necessary in a practical real world system.

Most common spray mist nozzles can achieve sprays with drop sizes down near 50 microns, plus you have evaporation that takes place as the mist moves down the induction path, which reduces the maximum drop size.

I just finished a turbocharger swap, and I took a good look at the compressor impeller on my old turbo which I had pre-compressor injection on for several months at about the 3% rate. (4 GPH nozzle max rated air flow for the compressor at sea level is about 25 lbs/min. Here at altitude of 5800 ft it is probably about 20 lbs/min.)

If the nozzle was spraying at its rated 4 GPH that would be about 252 cc/min or about .55 lb/min which works out to 2.7% of air mass at max flow. My turn on point is 8 psi so at the turn-on air flow, the mass fraction would be higher, probably near 5%.

On casual examination you could see no evidence of compressor blade errosion, just the normal discoloration you see after the turbo has been in use for some 37 thousand miles. On very careful examination under high magnification (about 10x), you can just see a small bit of roughness on the outermost 1 - 2 mm of the compressor impeller blades at the very leading edge.

I will need to do a similar examination of another turbo of the same design that was never exposed to water injection to determine if this is normal wear. This engine (like most Subaru WRX engines) has a crankcase breather inlet in the inlet tract a few inches ahead of the turbo inlet. Under high boost, these engines can blow noticiable amounts of oil mist into the intake. This over time results in a build up of a very thin layer of oil and dust "crud" on the inside of the inlet pipe and I'm sure that from time to time bits of this crud, and small oil dropplets gets suspended into the intake air stream under high air flow conditions.

I also on 2 occasions ran this engine with no air filter for a couple of drag strip passes to determine how much influence air filter resistance had on performance, so there was most likely a small amount of dust ingestion from those experiments.

Lastly when I first fabricated the system I assembled the prototype with a 35 psi pump which would not have given best misting behavior by the nozzles. The last couple months was with the system running at 100 psi max pump pressure which should have given a much finer mist.

Larry

[b_boy]

Thanks for the data.

I'm thinking I will inject about 6 inches after the MAF sensor, the water will travel about another foot to the turbo. I'll look into getting a couple of thermocouples to measure temp post turbo and post intercooler. I'll have to rely on data logging to look at pressure change rates and increases +/- pre-turbo WI.

I shoot for 2-3% injection rate and see how it goes. I'll have the pre throttle body injection going too. By the end of month I should have some experience and data.

[Gelf]

Excellent thread, the most compelling read on a forum ive had in ages.

[JohnA]

This is my pre-turbo injection nozzle, half way between air mass meter and compressor.

One thing I can say for certain is that the (aquamist) pump is louder now that it runs both nozzles.


http://homepage.ntlworld.com/johnnya/max-boost/images/WI/WI_preturbo_nozzle_lg.jpg

[Gelf]

Aquamist 2d installed on 2 ltr engine tuned to 250 - 260 bhp @ flywheel.

Two questions,

Ive decided to inject at the throttle body and before the turbo, what size jets will i need ?

Intercooler becomes interheater ! Ive had the stock intercooler replaced with a larger one, im a bit nervous of having no intercooler, what about when pressure is below the injection trigger level (9 psi) or if the water injection fails ?

I still have the stock intercooler, which i believe has a lower pressure drop.

[b_boy]

I would not worry about the intercooler becoming a heater just yet. Prove it's heating and then worry about it.

2-3% of air mass injection pre-turbo is just an educated guess. I read some turbine engine papers that indicated 2% was best and hotrod figured he was injecting 2.7%. Turbine engines are not the best model for our little turbos, much higher compression, much larger surface area. I'd love more input on what percentage to inject and droplet size if anyone has these data. Maybe Richard knows some mavericks.

2-3% is alot to inject. Generally WI injection is expressed as percentage of fuel injection 2.5% of air mass would correspond to 25% at 10:1 AFR, or 33% at an AFR of 12.5:1. That is a lot of water to move through the turbo, more than my pump and reservoir can flow.

I will start with something more modest. I have two 0.5 mm jets at the throttle body, shunting to a third 0.5, 0.4, or 0.3 mm will give me a range of about 125 cc/min, 90 cc/min, and 60 cc/min, respectively. At its highest, that corresponds to about 0.5% water:air mass with 1% water:air mass left over for injection at the throttle body. So, I would be injecting only about 20% of the amount that hotrod injects.

That may be too little, but I'm not sure anyone knows. I can monitor pressure and temperature post-turbo and see if temp continues to drop and pressure increase with added water.

Commentary on this set up is most welcome.

[SaabTuner]

Another up-shot to injecting prior to the compressor is for those people who run greater than 30 psi of boost. Above Pressure Ratios of 3:1 the air-charge is heated enough at the aluminum compressor begins to weaken from the heat.

Holset has been working on their cast Titanium compressors to solve that problem, but water injection might also help to keep the compressor wheel sufficiently cool.

Just a thought. It's probably already been discussed anyway.

Here's Holset's page on "high pressure ratio compressors": http://www.holset.co.uk/files/2_5_1_8-high%20pressure%20ratio%20compressor.php

Adrian~

[Gelf]

I found this link to a simulator from JohnA's web site, it has some features that simulate water injection before the TC or IC and after the IC.

I thought it would be interesting if i used the data i had from a RR to see the effects of WI. I entered the variables to the best of my ability, the models behavour is quite realistic. RR 230 bhp / 222 lbft @ 5371 on a cold March morning. I can only guess at VE.

The link below has all the data saved to it, which means you can see what i mean when you change the WI settings:

[admin - have to make the link's font smaller to enable the page wrap to function]


http://not2fast.wryday.com/turbo/glossary/turbo_calc.shtml?FeetASL=250&Tamb=14&Bore=86&Stroke=86&nCyl=4&RPM=5371&VE=88&Boost=11&Ec=70&Eic=70&PdropIC=2.5&TambIC=14&wiPercentMethanol=50&wiRate=250&wiTemp=14&SFC=.49&AFR=11&maxInjectorDutyCycle=85



Is interesting to note that a the model show that injecting 400cc before the TC, which is about 3.5% of the air liquid ratio, shows that the IC now only cools the charge by 1 C. The IC is at the threshold of becoming an Inter heater.

Injecting 320 cc, which is about 2.9% of the air liquid ratio, (and the limit of the pump with 1 injector) shows the IC still being useful cooling from 31 C to 19 C.

Probably shouldnt read to much in to this, but it would indicate that to inject at the manifold, a seperate pump would be desirable, in which case will the FIA 2 run two pumps and HSV's ? I remember that the MF2 can !

Your views....

[hotrod]

That is an interesting work sheet on the effects of WI, but you need to read the cautions about the model. I did a lot of playing with it and at reasonable amounts of injectant, it seems to give very good information.

It does not do any "bounds checking". If you put absurdly high amounts of water in, it while happily compute the cooling from total evaporation even though the amount of water is far beyond what is necessary to reach 100% humiditiy. Sometimes giving unrealistic air charge temps as a result.

So like any tool its no better than the base assumptions of the model. It does however, clearly indicate the basis for the assumption that post intercooler injection is the best "from the perspective of intake air cooling per gram of water injected".

It has no way of accounting for things like the change in timing of the pressure peak in the cylinder with different amounts of water and how that would effect the power output of a specific engine design.

The pre-turbo injection totally discounts any changes in the turbocharger compressor effeciency and any secondary feedbacks like, higher mass flow, lower turbo discharge temp, or lower exhaust gas back pressure that might result from that increase in effeciency.

Of course all those changes would be turbo type specific and would require a ton of code to cover all the popular turbochargers.

Larry

[b_boy]

I too have played with that calculator. And for me the best thing about it is getting a handle on how the different components of intake, IC, turbo, AFR, and WI cooperate to make hp. I agree with hotrod the model used in the calculator is too simple for our stuff and does not account for the saturation limit of dissolved water. The 100% "humidity" is an upper bound of WI, but both ambient air saturation and temperature affect this bound. Ideally we'd like to have a system that never reached the 100% saturation limit at any point in the air path (the bound is going to change, being highest just after compression, and probably the lowest while inside the intercooler". We dont' want precipitation at any point (e.g rain fall, fog formation), that will lead to water build up and all sort of "unintended" effects.

One calculation that I think would be worth performing is to figure the amount of heat produced at the highest operating RPM of your turbocharger and then calculate using the heat of evaporation for water +/- % methanol the amount of fluid needed to bring the compressor to isothermal compression. Assume that you are getting "explosive evaporation" as hotrod suggested.

Instead of using a compressor map with "predicted efficiencies" and a PR, if one were to actually measure pressure and temperature pre and post turbo, the energy going into heat should fall out of the ideal gas law. Now we divide this energy loss by the energy gain of evaporation, and we'll get a percentage of water/methanol injection. It's a ball park ideal figure, but it could be informative. Anyone have thermocouples in said places to take the measurements, and a boost gauge?

[b_boy]

Ok, I'm out on a limb here, but I did some thermodynamic calculations on water absorbtion of heat during compression at the turbo impeller. The calculation is in 4 parts and relates to my WRX STI 2.5 L turbo car.

Part 1 Assumptions and definitions

At 15-16 psi the engine is consuming ~250-275 g air/sec
Temperature of compression at 60% turbo efficiency, produces ~100 degrees C change at 7000 RPM
Heat of evaportation 540 calories/ gram of water
1 calorie = 1 degree C/ gram/sec at one atmosphere
1 calorie = 4.184 joules
Specific heat of nitrogen gas 0.25 calories/degree C (air is mostly nitrogen)


Part 2 Heat change under compression and amount of water to restore starting temperature


For 1 degree change in water temp we get 1 calorie*grams/sec.

For 100 C change (increase) of air (nitrogen) of 1 gram = 25 calories to heat 1g of air 1 degree C

540 calories/gram of water evaporated, heat of evaporation

Too cool 250 g air / sec heated 100 C costs 250*25 or 6250 calories / sec

6250 calories/ 540 calories/g water is 11.6 g water / sec = 11.6 ml water/ sec or 695 ml of water / min

Thus, 695 ml/min of water brings the compressor to isothermal compression with 100 C change in temp
during compression.


Part 3 Engine's consumption of air at 7000 RPM in a 2.5L engine


For my 2.5 L engine at one atmosphere and 7000 RPM, that's 3500 cylinder fillings per min,
or 2.5L * 3500, or 8750 L / min * 1000 ml/L =

8.8 million ml of air / min

At 15 psi of boost the amount of air is double or 17.5 million ml / min


695 ml/min cools 17.5 million ml/min air 100C or

0.004% (by volume) water to air ratio = isothermal compression with 100C temperature change


Part 4 Aquamist pre-turbo injection effects with my set up


100% stated "efficiency" of a turbo compressor is not isothermal, but adiabatic compression.

With Aquamist and a 0.5 mm jet I can inject about 125 ml / min, that is about 18% of 695 ml / min.

The turbo is operating at ~60% adiabatic efficiency at 7000 RPM, the 695 ml/sec water to brings it
to isothermal compression which corresponds to about to 118% efficiency at a pressure ratio of 2,
a difference of 58%.

18% (my 125 ml/min) of 58% is 10.5%

60% efficiency of turbo + 10.5% cooling from water = 70.5% efficiency of compression adiabatic.

250 ml / min water injection gives twice that or 81% compressor efficiency.

So with just 125 ml/min WI, compressor efficiency is near 70% at redline, quite a feat!

It's like having a 20% larger turbo.

I don't know the actual change in temperature during compression at 7000 RPM on my '04 Subaru STI with
the stock turbo. 100C is a reasonble guess, but it could be as high as 150C if the compressor efficiency
is lower than the predicted 60%. Thus, these numbers are only a rough estimate. While isothermal
compression would be nice to achieve, I can be happy with just extending the efficiency of the turbo
beyond it's normal bounds, and make up for the adiabatic heat by injecting more water after
the intercooler. Under this scenario maximum total water injection is 15% of fuel mass, very reasonable.

[hotrod]

That fits well with my experience.

The cliffs notes version is:

As you make the compression more isothermal, the true operating point of the compressor moves to the left on the compressor map. Since most performance applications push the compressor off the compressor map on the right side as they try to squeeze the last bit of performance out of the turbo this is in most cases a good thing.

Here at high altitude, the already small TD04L-13G which comes stock on the WRX is not just operating off the compressor map, it is on the next page.

When I started with pre-compressor WI, it made the turbo much more responsive in the mid range, and improved the max airflow a small but noticable amount. The mid range rpm onset of boost became so agressive I had to modify my boost controller settings to keep the throttle from becoming an ON-OFF switch.

Larry

[b_boy]

Thanks Larry for the "real world" experience.

It's just as I would like, move the compressor effectiveness back (left) into a more efficient island of the compressor map. I'll compress not necessarily more air, but compress it more easily with less intoduction of heat.

What effect the water will have on impeller speed questionable. While theoretically more viscous, the water mist may evaporate so quickly at the impeller to essentially behave as a gas, albeit a more dense gas. I believe it will slow the impeller to some degree, another benefit that I think Saabtuner mentioned.

I'm going to try to get WI volume up to 20% of fuel, such that when shunting 1/3 to the turbo we sending a bit more water/methanol through the turbo, around 0.7% by air mass. Or I'll shunt a larger percentage than 1/3. I'll have to see what works best.

We'll see how it works around mid-Oct.

[JohnA]

...With Aquamist and a 0.5 mm jet I can inject about 125 ml / min, that is about 18% of 695 ml / min.....
I've measured this jet at well over 200cc/min, almost 240cc actually with the engine running.
That was static, with no boost (or vacuum) in front of the nozzle.
If it's pre-compressor, then maybe the flow could be even higher (I'm not sure about this without testing further)

[Richard L]

I wish I can contribute more but it is getting beyond my level.

[SaabTuner]

What effect the water will have on impeller speed questionable. While theoretically more viscous, the water mist may evaporate so quickly at the impeller to essentially behave as a gas, albeit a more dense gas. I believe it will slow the impeller to some degree, another benefit that I think Saabtuner mentioned.

It might slow the impeller slightly ... or it might speed it up. The reason I think either is possible is that whether it slows or speeds up will depend on exactly where the water is atomized. If atomized near the tips of the impeller blades (as I suspect it would be) the over-all pressure drop across the turbo would be reduced at a given mass-airflow, and that reduction could allow the compressor to speed up much more quickly. I think Larry said something about it turning the throttle into an "on/off switch", which would go along with those lines perhaps?

Also I mentioned earlier that it would cool the impeller wheel slightly, which when boosting past 2 bar can be quite beneficial.

Adrian~

[b_boy]

It's interesting that you would say that Saabtuner. I've reading about air density and humidity. Surprisingly humid air is often stated as less dense. Let's take a different look at this, not from a throttle/turbo response view, but an efficiency point of view.

At the same temperature, water+air should be more dense than dry air alone--a simple more molecules/cubic area argument. I think the reason for the humid air being seen as less less dense than dry air is a misunderstanding of temperature, hot air being significantly less dense than cold air--hence the hot air ballon--and hot air holding more water--higher saturation point (the mid-west is summer).

If we picture a droplet entering the center of the impeller, it is immediately whisked centrifugally outward and "backward," evaporating along the way, perhaps explosively. Locally, liquid water is being transformed into gas, along with the air being compressed. On a purely local level there are more molecules in a gaseous state at the end of the compression, but since the liquid evaporated heat was absorbed.

If we imagine a model were droplet size reduces exponentially as it traverses the impeller to expeller length, at the beginning the density of the air with or without the droplet is nearly the same, however, at the end of the traverse (notably the point of most leverage) the air + droplet gas will be much more dense since the gaseous contribution of water will be huge.

I think what we will see by injecting water is that there is a slowing of the impeller wheel that results in the same compression of air at that slower speed plus the added benefit of reduced temperature of that compressed air. It my case, turning water at 5000 RPM, we may shift the impeller speed back to where it was at 4000 (and 70% efficiency), as we increase RPM on up to 6000, 7000 while increasing the water injection correponding, we can try to stay at the same 4000 RPM, 70% efficiency range.

This is one way to look at: view it as creating a static impeller speed of high adiabatic efficiency (e.g. the 4000 RPM equivalent impeller speed, say 120,000 impeller RPM) by injecting more water to slow the impeller down. But we don't want to do that, we want the impeller to continue it's upward climb in speed so that it can compress more air. The reality is most likely in middle, slowing the impeller some (while cooling the air that is compressed), but allowing for more compression with increasing speed (but still cooling the air compression). There is no way density of water + air is going to completely overcome the torque of the exhaust side.

Now, coming back to the rapid throttle/turbo response view. The quickness of the throttle is going to be mainly a function of the rate of air and fuel entering the engine. Early in the turbo's speed range, slowing the turbo's ascent in speed will be a disadvantage, prior to peak compression, in fact the impeller is an impediment to air flow if the wheel is not turning at all. Once a decent efficiency is reached though, any further cooling of the air being compressed will result in more air in the cylinder for every turn of the impeller, thus the rate of air will be faster at that instant, and the rate will increase with increasing water injection linearly. I think that is where you could see an increase in throttle response over no WI pre-turbo, but it would depend on maintaining the temperature difference as the compressor continues to spool.

I can't believe we stumped Richarl Lamb with this stuff. To his credit, it is more turbo theory than WI theory. We are all learning something I guess.

Thanks to John A for the measurement. I'll have three 0.5 mm jets, so I don't think I'll achieve his volume as it is distibuted between the jets unless I add another pump or greatly increase the pressure.

In the end, it's going to come down to testing it out. :wink:

[SaabTuner]

I think there may also be some mixing here of the term "pressure". Static pressure and flow pressure (or total pressure) are somewhat different things.

"evaporating along the way, perhaps explosively. Locally, liquid water is being transformed into gas, along with the air being compressed."

When thinking of this problem remember that the turbocharger's compressor wheel only accellerates the gas. The compression takes place in the diffuser scroll in the housing. Most of the evaporation/boiling should take place towards the tips of the impeller as that is where the static pressure should be lowest (though velocity pressure quite high). Thank the Bernoulli principle for that one.

While I'm not sure if the water's evaporation would increase or decrease the density of the gas, I think it's fairly safe to say that the evaporation of water in the compressor would significantly change the flow through the compressor.

The reason I say that is that on the forward facing edges of the blades there is much higher static pressure than on the backwards facing edges. Since more water will evaporate in lower pressure, the low pressure regions may either get MUCH lower, or get much LESS lower, depending on what effect the water has on static pressure when it evaporates. In either case, it would dramatically change the flow field through the impeller.

Adrian~

[Richard L]

I am reading and learning. Love to have some practicle examples and results.

[hotrod]

Just thought I would try and post some images of the turbo I just pulled out. It took a while to figure out how to get descent resolution images.

Richard I will send them to you by email if you could post them up as I have no place to host them.

Keep in mind a couple things:
1. On casual examination the impeller looks just fine, its only when you look close that you see a bit of errosion at the outer impeller tips.

2. I suspect I was over spraying (3%) mass air flow, and I also experimented with spraying at high idle speeds (very low throttle setting). I discovered that this would act to cool the intercooler prior to drag racing.

(note to self, very good idea for post turbocharger pre-intercooler injection point . If you could manually trigger this pre-intercooler spray prior to drag racing, or Autox event all the evaporation takes place in the intercooler and quickly cools the structure of the intercooler. Due to the engines air flow you completely eliminate any need for external fans to cause cooling of a heat soaked intercooler while the car is stationary)

I suspect during these experiments I may have had some liquid water streaming down the walls of the turbo inlet pipe. I believe, this could be a possible cause of the erosion at the compressor impeller tips.

IF that proves out to be the case, it might be useful to put a trap of some sort in the inlet tube that would re-suspend any surface water in the air flow. This could be in the form of a ridge that would stop the droplets that tend to run down the surface of the inlet tube and cause them to be ripped apart by the high speed airflow as they try to climb the lip.

3. Last but not least, it is important to note that I also made about 6 drag strip runs with this turbo using no air filter of any kind, to determine the intake air restriction the filter creates, and its effect on turbo boost. This was through a cold air intake system that had the intake in front of the right front tire and directly behind the front bumper. At the strip I run at there is a section of gravel/dirt you must drive over as you pull off the strip. I know I sucked up some dirt on at least one of those events. Shortly after that experiment, I pulled the discharge off the turbocharger and found small amounts of aluminum dust in the turbo to intercooler tube and some very fine dust. I wiped this off and it did not return after that. If you look closely at one of my pictures you will see a clear "meteor crater" impact about a quarter mm in dia. on the leading edge of one impeller blade from a piece of grit.

My personal judgement is most of the errosion is due to these no air filter tests, and the effects of the water were minimal but I cannot prove that case. After I run my current turbo for a comparable period of time I will pull the intake tube and check for similar wear without having ever made any filter free passes.


Larry

[Richard L]

Here are the pictures of larry's turbo. Picture restored - lost and found.


http://www.aquamist.co.uk/forum/compressor9eb-s.jpg

[SaabTuner]

Old Saab99 Turbos had a pre-turbo injection system factory stock. It was common after 100K miles or so to see some pitting in the compressor blade. But at that point, the turbo was going to be replaced anyway. I've never heard of it causing any problems.

Adrian~

[hotrod]

Thanks Richard!

Saab tuner
It was also used on 10's of thousands of military aircraft without serious problems.

There is however an urban legend sort of belief in the performance community accompanied with horror stories of "some kid" etc. etc. that ruined his turbo in a matter of minutes after trying a pre-turbo setup.

The main thing I wanted to demonstrate is that it does not cause immediate catastrophic damage. In my case I considered this turbo to be expendable as I already had its replacment on the floor beside my desk. After seeing the condition of the compressor and considering what I put this puppy through, I had not qualms about using pre-turbo WI on my new setup.

Larry

[SaabTuner]

Like in those NACA reports.

I just wanted to point out that it was used on automobiles, which are usually at least perceived to have longer service intervals. (Whether they actually do or not.) Of course, it was used on Saabs, which are usually perceived as having short service intervals and low reliability.

http://www.saabcentral.com/trivia/mileage.php

Oh well. :P

Pre-turbo injection is certainly much much cheaper. When I work out a WI system, it's going to be multi-port, and I'm going to do my best to get it CARB Exempted. Otherwise I'll fail emissions. (Some people pass, but not if you end up at the notorous "test only" stations. Then you're toast. I have quite a bit of experience with that ... dammit.)

Adrian~

[Greenv8s]

Quickly. Normal compression is adiabatic, which to a first order means that as the gas is compressed, it gets hot. Now this heat is one of the biggest problems with forced induction, for 2 reasons. Firstly you have to get rid of the heat, and secondly you need to take power out the turbine shaft to perform the heating.( heat is work and work is heat).

Now with the right level of water injection, the heat is removed before it builds up, pushing the compression closer to isothermal (not all the way, but closer). In round terms this is about 30% more efficient (less exhaust gas required for the same boost, or more boost at the same exhaust flow).


Still trying to get my head round the thermodynamics, but its a bit of a puzzle. The evaporative cooling takes heat out of the air and avoids the air gaining temperature and pressure. At first glance it is obvious that this means there is less energy being put into the air, but isn't it just going into the water instead? Or have I missed the point? I guess that the outlet pressure would be similar whether the charge was cooled during compression or later in an intercooler, so reduced temperature implies reduced volume, speed and KE. Is this where the gains come from?

[b_boy]

The heat energy (kinetic) is being converted into entropic energy (liquid to gas transition--the heat of evaporation (heat needed to disassociate H2O molecules)).

Ideally at the moment the adiabatic heat is generated, it is absorbed by the water evaporation. Compression occurs with less heat imparted to the compressed air.

If liquid water traverses the impeller wheel without evaporating, we are no better off injecting before the turbo as after.

Now the water droplet will spend very little time traversing the impeller wheel. We what it to completely evaporate during its journey. Hence, the dialogue about droplet size. We can calculate this time frame with some precision, but the time to evaporate a droplet is a difficult one. Pressure, temperture, velocity, increasing surface to volume ratio, and methanol mixture, will make it a multivariable calculation, one best left to empircal determination.

Aquamist at 147 psi is our best bet for small droplet size. Smaller droplets can be made, but the pressure and nozzle sophistication to make them is untenable in automobiles with current technology. I know of no sensor that could easily be rigged to detect droplet exit from the compressor wheel (a spectrophotometer could work, but we are talking at least a $1000).

Hotrod has kindly displayed his turbo impeller after 7000 miles at 2.7% water to air mass. It shows some notable erosion at the edge (with extenuating circumstance). The erosion can be viewed as a cost of implementing the impeller injection, or we can wait for creation of a nozzle that expels smaller droplets. By injecting less water than hotrod, and only at high RPM and boost, I hope to reduce the erosion substantially (using a solenoid to inject only under the chosen parameters). None the less I expect some erosion to occur. If a new turbo is needed every 20,000 mi, that is a cost of running such a system. It remains to be seen if the benefit outweighs the cost. If a 10% increase in airflow can be achieved at high RPM, it would be worth about 40 hp; that is worth it to me.

[JohnA]

... it is important to note that I also made about 6 drag strip runs with this turbo using no air filter of any kind, to determine the intake air restriction the filter creates, and its effect on turbo boost. ...
When I was careless (sorry, experimental!) enough to run my turbo (bike) without airfilter, that's the sort of wear I saw on the compressor blades.
Didn't take long to happen either... :roll:

[hotrod]

Thanks, that backs up my circumstantial evidence and theory that that damage was from the "dirt experiment".

Really T'd me off at the time, I forgot there was a segment of dirt and gravel at the end of the strip where you turn off to pick up your timing slip. Lots of folks cut the corner short and thow dirt and sand up on the paved portion of the turn out. I took the turn out at about 30 mph and was in the crud before I realized it. My air intake sits right behind the front air dam and I figured the air dam passing close over the surface of the sandy pavement suspended a lot of crap in the air.

Like you said it only takes a single event. After that first pass, I took the turn off road very wide and slow and saw no more evidence of sucking dirt.

Interesting side note is that the damage is confined to the extreme outer blade tips, probably a combination of the swirl that develops ahead of the compressor inlet centrifuging the dust to the outside of the tube and the higher relative velocity of the blade tips vs the portion of the blades closer to the hub.

There is one guy that posts on another form that says he's run a pre-compressor injection for years with no sign of compressor damage.

Larry

[Greenv8s]

Interesting side note is that the damage is confined to the extreme outer blade tips, probably a combination of the swirl that develops ahead of the compressor inlet centrifuging the dust to the outside of the tube and the higher relative velocity of the blade tips vs the portion of the blades closer to the hub.



Way off topic, but I wonder would it be feasible to use this effect to create a centrefugal air cleaner? Vacuum cleaners using this effect claim no loss of suction as the 'bag' fills, presumably in our case that would translate to no increase in pressure drop across the filter with age? For this to work it would be necessary to recover the static pressure drop across the vortex, in an ideal world it would be possible to recover this by slowing and expanding the air to convert dynamic pressure back to static, as long as there was not too much drag within the vortex. Also, straying slightly closer to topic, by passing a water mist through a vortex of this sort, fine droplets might tend to be drawn through with the air while larger droplets woud be held back until they have evaporated, broken up or settled out on the walls. Probably a crazy idea, but do you reckon it would work? How much of a swirl would be required to separate out the dirt and water droplets of interest, and how would you go about testing it?

Pete 'crazy ideas' Humphries (and a green V8S)

[b_boy]

I want to say, I'm perfectly willing to accept that Hotrod's impeller wear is from dragging with no filter. At 150,000 RPM, that's 2500 RPS, so 2 seconds of dust, dirt and you have 5000 turns of the impeller. You could do a lot of damage with 6 runs.

I'm only trying to express the possibility of wear, erosion. I would not want anyone reading these threads to go into pre-impeller injecting without this as a concern.

[hotrod]

would it be feasible to use this effect to create a centrefugal air cleaner?

They already exist, industrial equipment (cat tractors diesel trucks etc.) use them as pre-filters ahead of the normal filter element to extend its service life.


Larry

[Richard L]

This may be a dumb question. Isn't the main heat generated by the turbo is friction between the air/impeller surface? The same can be said with moisted air.

Shouldn't really compare it with the space shuttle entering the earth's atmosphere and picks up speed and heat rises as the craft speeds up, but pretty close to the air sliding passes the impeller at 150,000rpm.

[SaabTuner]

I'd say that probably constitutes a big portion of it.

Also when the air enters the scroll, or vane, diffuser section the velocity pressure is converted to static pressure by slowing the air back down. This involves a lot of small eddie currents in the air associated with fluid vorticity. Generally vorticity ends up dispersing into heat.

Adrian~

[Richard L]

I'd say that probably constitutes a big portion of it.

Also when the air enters the scroll, or vane, diffuser section the velocity pressure is converted to static pressure by slowing the air back down. This involves a lot of small eddie currents in the air associated with fluid vorticity. Generally vorticity ends up dispersing into heat.

Adrian~


"eddie current" of the air is another word for the air molecules rubbing against each other in a circular fashion and the friction between the molecules generate heat?

Sorry to look at things in a more simplistic description.

[Greenv8s]

This may be a dumb question. Isn't the main heat generated by the turbo is friction between the air/impeller surface? The same can be said with moisted air.


Some of the heating corresponds to inefficiency in the compressor, and I imagine most of this comes from small scale local variations in velocity around the compressor which are eventually damped out by friction in the air, as suggested previously. However, turbo compressors are relatively efficient so this effect would only account for a small part of the heating effect. The rest of it comes from adiabatic heating. The mechanism causing this occurs at the molecular level, where gas molecules bouncing off the inner surface of a container rebound with increased speed when the container shrinks (think of tennis balls bouncing off a moving racket). This results is increased speed and momentum in the molecules, which is essentially what heat is.

[b_boy]

Turbos are rated in terms of efficiency of compression, but they should really say efficiency compared to adiabatic compression. At best a turbo in a narrow RPM range is 80% efficient with respect to adiabatic compression. So, there is 20% of the heat that is not merely the result of compression, but comes from somewhere else--friction, turbulance, heating of the turbo materials. As the turbo moves toward stall speed (the speed of sound) the efficiency drops off exponentially. As mach 1 is reached, not just heat but castistrophic sound waves disrupt the whole stability of the spinning wheel.

All the talk above in this thread about isothermal compression, is compression without the generation of any heat. Isothermal is a theoretical idea that cannot be achieved. However, if we can cool the air with water droplets as it is being compressed we may achieve a quasi-isothermal compression. At the outer limits of the impeller speed (near mach 1) the efficiency can drop to 50% or less of the adiabatic value.

For the sake of arguement let's say we could "recover" all the energy lost to heat at these near maximum speed. We could realize a 50% increase in compression with no heat loss. Effectively, our turbo would be 50% larger. That is a major accomplishment for the injection of a little water.

Practically, we will never cool compression this much, but even a 10% or 20% gain would make a huge difference in the amount of air mass available for combustion. For a 300hp engine, that's 30-60 hp gain. Many folks would kill for a gain like that.

Ideally, I'd like to experiment with the turbo off of the car. Compare the compression with and without water injection pre-turbo and see if compression is increased (i.e. air flow) and if post turbo temperature is reduced. Also in this ideal set up, we would want not liquid water (droplets) to survive traversing the turbo (not that that "overspray" as it is called is a bad thing, we are just not measuring the change in compression/heat only).

In the real world the best we can do is measure post turbo pressure and boost with and without injection of water. My hope is that I'll see more compression (higher boost) and lower temp (adiabatic and "friction losses" absorbed my water evaporation).

We are basically talking about running the turbo outside of it's efficiency range, using water to compensate for adiabatic and turbo losses. Like, the fear of water shut down that may destroy the engine, similarly the turbo may spin out of control if water were suddenly removed.

Engine management to prevent catastrophic damage is essential, since the computer can react much faster than we can. All attempts to operate at the edge of tuning involve some form of risk. Sensors, redundant pumps, auto shut off fuel or boost blowoff, would all lead to a safer system. This is one of the reasons I was so excited about the now indefinitely postponed EcuTek-Aquamist offering. For now, I'll be using the Element Hydra to design a system as best I can (one I'm sure will not be as good as a professional product).

[SaabTuner]

"As the turbo moves toward stall speed (the speed of sound) the efficiency drops off exponentially. As mach 1 is reached, not just heat but castistrophic sound waves disrupt the whole stability of the spinning wheel."

I think what you are referring to is termed the "choke line".

As a rule boost pressure is related to the square of the compressor speed. The ultimate boost limit is not related to the speed of sound, rather by the heat at the compressor discharge. Aluminum compressors become weak above 300F degrees, too much boost and the wheel is weakened significantly.

Also, remember that the speed of sound is an entirely localized phenominon. The speed of sound on the backside of the compressor blades is much less than the speed on the front side. Even at a relatively low speed the backside surface of the compressor blades exceeds the local speed of sound.

Furthermore, a few simple calculations will yield that most compressors operate at a speed which exceeds the speed of sound at 1 atmosphere of pressure.

The choke line is more likely related to cavitation on the backside of the impellor intake side fins. Air cannot flow inside the cavitation. As the compressor speeds up the cavitation size grows for any given pressure ratio. Eventually the increase in flow generated by a higher compressor velocity does not make up for the decrease in area due to the expansion of those cavities.

Cavitation is worsened with increased flow because the faster the air flows through the compressor at any given static pressure ratio, the lower the internal pressure ratio vector in the direction of the cavitation as the flow is perpendicular to the direction of cavitation.. (Bernuelli Principle essentially.)

edit: Another possibility is that the choke limit is when the boundary layer between air and compressor is no longer stable. This is probably the most likely reason. Increased speed and flow both contribute to a destabilized boundary layer of air. Vorticity will become immense when the boundary layer is disturbed and the air will get very hot, and not flow well.

Anyhow ... sorry for that topic tangent.

Adrian~

[b_boy]

Don't be sorry for the diversion, I truely enjoyed it, and it was intimately related to the topic of tuning. I did imply and desire to turn back to tuning, but all your comments and others are welcome information.

The extent lf my fluid dynamics knowledge is weak at best, so I am aprreciative also of the more thorough explication of choke (I always confuse the two sides of the compressor maps, my apologies). I find all of your explanations highly probable, cavitation, heat, and laminar distruption--all probably play a role in destablizing air flow.

Please let us know if you think the argument still stands, that water/methanol will stretch the choke line by reducing heat and possibly slowing the compressor. I would think that heat radiating from the impeller would be a profound cause of disruption of the boundary layer in particular.

I'm putting a thermocouple in after the turbo to measure temp, a pressure transducer would be nice too, but they are too much dough and I already have the boost gauge and engine MAP sensor to follow pressure.

Thanks again for your detailed input and data!

[Macabre]

For the sake of arguement let's say we could "recover" all the energy lost to heat at these near maximum speed. We could realize a 50% increase in compression with no heat loss. Effectively, our turbo would be 50% larger. That is a major accomplishment for the injection of a little water.

I've read through this thread, and this statement is really the crux of what's being discussed here. However, what I have not seen is any evidence or argument to suggest that this effect would be any different from the heat absorbtion that occurs when injecting water after the compressor. I.e., that it isn't accompanied by a proportional drop in pressure. I don't mean to be challenging, just looking for an explaination if there is one. It seems to me that there is a fairly large logic leap involved in the supposition that injecting water at the compressor inlet will increase the MASS at the compressor outlet (that is, a drop in temperature without a drop in pressure). It would be fairly easy to test. Anyone with a WRX + DeltaDash could gather this data from their MAF/MAP sensors. However the theory of why it would happen would be interesting as well. In order for the turbocharger to be "effectively" larger, it would need to flow a greater mass of air at a given PR with the same efficiency, and I cannot get my head around the logic of that.

[b_boy]

The data you seek do not exist for our little turbos, but for prop jets and turbine engines the evidence is ample. Look under "wet compression" of "fogging" or "overspray" or "swirl flash" and you'll find the papers and studies.

In addition we have the testimony (the word comes from if you lie, your testicles are lopped off) of HotRod and some others that you can improve the efficiency of compression itself.

The only advantages to injecting pre-turbo are 1) you want a bigger turbo but are not ready to role the dice, and 2) you have no intercooler, can't fit one, or don't want one, and 3) there may be a thermal advantage to not generating the heat early in the air flow (w/out heating the intake system).

To me number 1 is paramount the others. The quest for a greater dynamic range in turbos has been a long sought pursuit--variable vane turbos, twin turbos of differing size, larger turbos with less spool friction (i.e. ball bearing), external wastegates, and so on.

That said, water injection is still a marvelous means of increasing the density of the air, suppressing detonation, and permitting combustion with more power/push (leaner AFR), even if injectin post-turbo.

It's not the holy grail, but it's sure is fun thinking about, and pretty cool if you can make you 400 CFM compressor flow 500 CFM. Penny per HP it's a huge mod.

[Macabre]

#2 can be achieved with WI regardless. #1 is still dependant on the compressor flowing greater mass when water is introduced. I find this all very interesting, and may revise my plans for adding a second water jet pre-intercooler to adding it pre-compressor. I did find a good article based on your search terms that was very helpful - http://www.caldwellenergy.com/pdfs/WETCOM.PDF particularly "With compressor inter-cooling, the compressor discharge temperature is reduced considerably allowing more fuel to be burnt in the combustor. A secondary increase in power output is due to increase in mass flow rate." . It all sounds too good to be true but highly attractive if it works.

My second concern is how to calculate water usage for a given jet size given an assumption there will be a variable vacuum present at the compressor inlet. I suppose I could relocate my boost/vac gauge there for testing.

[Forum Admin]

Turbine's Power output is thrust related, the higher the mass on the intake regardless of either water molecules or air molecules, the higer the outout thrust.

Engine's output is mainly oxygen related, mass-increase as a whole in the intake may not yield extra oxygen available for combustion if a large portion of the mass is water molecule. This is an interesting topic to discuss if there is an actual oxygen increase.

If the engine can make more power by injecting a fine mist before the impeller, it can easily be verified on the dyno?

[Macabre]

Engine's output is mainly oxygen related, mass-increase as a whole in the intake may not yield extra oxygen available for combustion if a large portion of the mass is water molecule. This is an interesting topic to discuss if there is an actual oxygen increase.

Indeed.. this is why the data I would really like to see on the subject is output of the engine's MAFS compared to MAP and RPM. If no more air enters the vehicle intake, the turbocharger has not been made "effectively larger." It still feels like a big leap to me to state that water injected into the compressor will directly increase the efficiency of the compressor (vs. just reducing the outlet temperature). It may be that this is just a peculiarity of centrifugal compressors, or it may be that details that affect one type of system are being inappropriately applied to a combustion engine's turbocharger. Dyno results can be dubious on these issues.

[Richard L]

I think there is a bit more to it. The MAF is temperature compensated but wet air will give a different read on the hot wire type.

In order to verify the theory, we really need to find a relate simple method to get some result - it may not have to be vey accurate but the claimed 20%+ air flow increase should give some positive indications no matter how inaccurate.

I really like the idea behind the claim and I think it will work but may not be a much as 20% in oxygen increase. Any thing to reduce heat is a good thing especially on the turbo impeller.

Most turbo on the market is water cooled, some benifit for sure.

[slowMX5]

I have been watching this thread with some interest. I'm beginning to become inspired to do a little testing. Currently I am only using one of the TEC3r's (www.getfuelinjected.com) mapable outputs to map WI (using Aquamists 2C system), with 2 nozzles post IC. I have another spare mapable output - and I'm beginning to think that placing another HSV in a line supplying a pre-compressor nozzle might be a worthwhile experiment. This would enable me to play with the water delivery and compare results of varying switch on points and delivery amounts independently of water that is being delivered post IC.

The only thing making me think twice at present is I am running a Garrett Disco Potato turbo (GT28RS, 0.86 AR compressor 76 trim, 0.64 AR turbine 62 trim) - so it is if anything oversized for a 1.6L engine making 280bhp (sized for future plans).

Worth making the effort? I have full data logging software that will allow repeatable power/torque graphs to be plotted if using the same stretch of road.

[Richard L]

I can send you a HSV for that purpose?

[slowMX5]

Sounds like a plan. I'll post results here once I've got them.

[Macabre]

I think there is a bit more to it. The MAF is temperature compensated but wet air will give a different read on the hot wire type.

But the air at the MAFS should not be wet. The spray would still be well downstream of the MAFS (at least, on most engine types). The data I'd like to see would be whether this practice can affect a real efficiency improvement from the compressor. That is, at the same PR & flow, there's a greater mass of air in the charge. If more air-mass exits the compressor, it had to enter the compressor, and that should be suitably measurable by the MAFS. Or am I missing something?

[Richard L]

You are correct on the MAF location in general but some aftermarket bolt-on systems do have the turbo before the MAF.

I think we shall see slowMX5's results to make this discussion more meaningful.

[b_boy]

I'll be tuning soon as well and post numbers. We will see what we get. Maybe Hotrod can be presuaded to throw a thermocouple in his set up and run plus/minus WI. He should see definitely see a drop in temp, but will he see an increase in boost is the question.

It's a tough experiment to compare because of detonation.

[hotrod]

I don't have the resources to do the thermocouple test, but I know the result. Several folks have run similar tests and the temp drop is on the order of 30 - 40 deg F for common WI injection rates. For straight water injection with no alcohol the temp drop will be about 20 -30 deg F, depending on the relative humidity.

As far as the boost, I already know the answer to that question, My electronic boost controller has a peak hold feature that tells me the maximum boost achieved on a given run. I had to turn down my boost controller to avoid over boosting after I installed the pre-compressor injection.

The mass flow through a centrifugal compressor varies directly with the absolute temperature of the inlet air. Lower the inlet temperature, mass flow goes up. Its the nature of the beast.

Larry

[ryan]

HI,

Arch my brain cell herts, na just kidding love this thread,

I was thinking that when you inject water post compressor, this cools the intake temp,

but does this cool it enough to reduce the volume (less heat less volume) when the injected water mass/volume is included,

if so are you not getting the same effect , ie compressor pumpping more air but is hot, then its cooled back down and dropped psi (less volume same mass+water mass) to the same mass/volume as the preturbo injection goes?. or fairly close to it, ie both will have water taking air space, and have lower temp-more air mass.


just a thought, sorry if this sounds dumb to you, but mayby has some relivance to this topic as far as how much more air will be moved with pre-compressor injection vs post compressor injection.

although preturbo would make for good mixing/evap of injected liquad.

and the lower temp in the compressor exit would make for less volume(same mass), so there would be less pressure for the compressor to fight at the same mass flow so less heating and more efficenecy , hmm O.K. I'll stop typing now.

Cheers
Ryan

[Greenv8s]

I've been watching this thread with interest, thanks for such an informative discussion guys.

I don't have any practical experience to contribute, but a comment by Hotrod has resolved the main issue that had been puzzling me.

I couldn't see why it made any difference whether the air was allowed to heat up in the compressor and then cooled by downstream injection, or whether upstream injection was used to prevent the air from heating up as much in the first place. At first glance it should not make much difference since a similar amount of water is evaporated (OK maybe *slightly* more evaporation in the upstream case due to the turbulence inside the compressor) so the final charge temperature should be similar. I don't think anybody has suggested any reason why the compressor efficiency would be improved by upstream water injection, and I can't think of any.

However, Hotrod has pointed out that the compressor works more effectively when it has denser air going through it. By reducing the air temperature rise inside the compressure, the upstream injection increases the air density inside the compressor and therefore increases the effectiveness of the compressor. (I think the distinction between effectiveness and efficiency is important in this case.) This means the same compressor will produce a higher boost pressure and greater mass air flow with upstream WI than with downstream WI, which I think was the original contention. My mistake was to assume this resulted from increased efficiency; I don't think it does.

Does this make any sense, or am I still missing the point?

[hotrod]

That's a good summary of the situation.

It also has a secondary impact that is not often discussed. If you increase the effeciency of the compressor, it takes less power from the turbine side to do the same work. The result of that, is the turbo spools faster and at a give boost pressure it will produce less exhaust gas back pressure.

What you have is a situation were several small, and complementary incremental improvements in the performance of the entire system make a noticable difference.

On my car, since the stock TD04L-13G turbo on a WRX is undersized by the factory, to get good midrange response. Added to that since I live and race at 5800 ft altitude, the effects of operating at high altitude made the workload for the turbocharger even greater.

The seat of the pants improvement in turbo spool and improved midrange power and torque were very noticable.

Larry

[b_boy]

Studying the thermodynamics is dizzying. There are so many variables, I just cannot predict the outcome of measurement. Evidence from multiple sources makes me believe that the turbo will actually compress more air, but proving this with heat in the mix will be difficult.

I see two ways to measure an effect. One is to tune to knock with only WI at the throttle body, and then turn the water on pre-turbo and tune again to see if more power can be had. This only establishes that more power can be had. Perhaps injection of water immediately post-turbo would have the same effect? I'm not willing to do that control experiment.

The second way is to measure pressure and temperature post turbo with and without WI. We might hope to rely on the ideal gas law to compute a gain over and above mere evaporative cooling. Since on a mass basis we are only adding a small percentage of water, the molar contribution to the gas's mass flow should be minimal when computing pressure, as well, any effect on the R value should be mininal as well.

Using PV=nRT, we will be solving for n, number of moles, or an increase in the mass of the air compressed. We will have to adjust any pressure reading by compensating for the compressed gas temperature post-turbo. If there is a difference of more than a few percent at near redline RPM we should see it. Any ability to increase power without knock will be only confirmational.

I'm buying a hypodermic thermocouple to stick into the hose that runs to the intercooler. I don't know where the stock pressure sensor for boost is, preferable before the intercooler. (Do you know hotrod? I'm in a WRX STi.)

[hotrod]

Sounds good to me.

The MAP sensor on the WRX is on the throttle body passanger side right above the throttle shaft. It has a three pin connector and is held in place with 2 bolts.

Larry

[Greenv8s]

As far as I understand it, the theory is that upstream water injection enables the compressor to achieve greater mass air flow (by reducing the temperature rise and loss of density that occurs inside the compressor).

I don't think it is possible to prove this theory by comparing with/without water injection, because we already know that water can be beneficial wherever it is injected.

The only way I can see to prove the theory is see whether we get more mass air flow using WI immediately upstream of the compressor or with the same amount of WI immediately downstream of the compressor. By concentrating on mass air flow at the compressor I think we can avoid complications from other changes in the tune. To be conclusive, we would need to ensure that a similar amount evaporates in both cases. I think it might be possible to achieve this by using only a small amount of WI so that we get total evaporation in both cases, does this seem reasonable? Since mass air flow is the thing we are looking for, it would make sense to measure that directly if possible, rather than trying to infer it from changes in temperature and pressure or detonation, net bhp etc.

[b_boy]

I think we agree. I'm talking about computing the number of moles (mass of air) and you are talking about mass flow, but we are both talking about mass.

In the end we are asking: does the compressor carry more air particles if cooling occurs during compression, than if cooling occurs after compression?

I believe the answer is yes, and here is my lastest thought experiment. It's gets right to the heart of the matter: number of air molecules compressed per rotation.

Imagine this case. The ambient air is -20C, when compressed this air, with each turn of the compressor wheel, contains more air molecules (cold is denser) than if the air were at 20C (less dense few air molecules/unit area).

Likewise, if during compaction, temperature (i.e. kinetic energy) is kept constant, the air being compacted will be more dense, and hence again with each turn of the compressor more air molecules are compacted.

The work that we are interested in is the work of compression, which is all about the pressure ratio given by the centrifugal forces than are acting on the air. Imagine that each time the impeller blade swings by is analogous to a shovel full of air. If each shovel full wieghs twice as much, then you end up with twice as much air shoveled per unit time. Similarly, with a pressure ratio of 2 (isothermic) each swoop of the impeller results in a compression of twice as much gas.

If without water the turbo is compressing air at a PR of 2, and we could completely suppress all heat during compression, we will have gained back all of the adiabatic heat loss and all of the compressor's inefficiency. The percent gain will be directly related to the PR. The larger the PR, the more gain can be had for each compression event--just like the shovels of air, 4X density air is more than 2X air.

As has been said before, the gain could be huge (40%+) if all the heat is absorbed by a small amount of water. To tie this back to yesterday's discussion. We would need a much larger turbo to compress say 40% more air, that was then cooled by WI post injection. It's all in the air density per rotation of the compressor--more dense air (cooler) leads to more air molecules compressed per rotation of the impeller.

[SaabTuner]

If the "choke line" is the point at which the boundary layer next to the vanes breaks down, then water injection will surely make the compressor more efficient.

The water droplets will make the air denser. Denser fluids have less vorticity. Vorticity is responsible for turbulence and the breakdown of the boundary layer.

Therefore denser air makes the turbo more efficient by reducing vorticity and thus stabilizing the boundary layer of air.

Makes sense, right?

Adrian~

[hotrod]

If you have the means to log a couple data streams it should be pretty simple to prove.

On the WRX, you have a MAF (Mass Air Flow) meter that reads directly in grams/second ( if you have the proper logging software) it is a 0-5 volt output so even a peak hold volt meter could be used. Just set the boost to a known value and log the mass air flow at a known rpm. Then turn on the WI and adjust until you hit the same boost setting at the same rpm and compare the new MAF numbers.

Larry

[b_boy]

So, if choke occurs later, the compression could continue at higher wheel speed. The speed of sound increases in denser mediums, the factor of which I do not know.

I'm not familiar enough with vorticity other than I can see that disrupting of either the boundary layer or added sonic turbulance would lower compression by preventing laminar flow, and add heat to that air (turbulance) without compression. For the layman, is that what you are saying Adrian?

Choke occurs when the small diameter (inlet) of the impeller reaches sonic speeds. You think that the added density with water injection will slow the wheel to under Mach1, yes? Or, that the speed of sound will increase, permitting extension (more wheel speed) of compression until choke?

[SaabTuner]

So, if choke occurs later, the compression could continue at higher wheel speed. The speed of sound increases in denser mediums, the factor of which I do not know.

I'm not familiar enough with vorticity other than I can see that disrupting of either the boundary layer or added sonic turbulance would lower compression by preventing laminar flow, and add heat to that air (turbulance) without compression. For the layman, is that what you are saying Adrian?

Choke occurs when the small diameter (inlet) of the impeller reaches sonic speeds. You think that the added density with water injection will slow the wheel to under Mach1, yes? Or, that the speed of sound will increase, permitting extension (more wheel speed) of compression until choke?

I'm not sure choke has as much to do with compressor speed as it does flow Remember that compressor speed is more closely tied to boost than flow. As an example 150,000 on a compressor with a 2" small side is nowhere near the choke line at low flow. Despite that, the tips of the compressor blades (on the small side) are travelling at roughly mach 1.37 even when only flowing half of the maximum "choke" flow. The outer tips of the compressor blades are travelling at well over mach 1 even at just one or two psi of boost pressure on most turbos.

One of the mechanisms for generating vorticity in compressable flows (like air) is the baroclinic generation term.

"The baroclinic mechanism for vorticity generation is frequently invoked to explain the existence and direction of off-shore and on-shore breezes and the generation of vorticity in compressible flows. The baroclinic generation mechanism will be non-zero whenever the density and pressure gradients are not aligned." Taken from Navier-Stokes.net

Since the evaporation of water is easiest in low pressure regions, vorticity would inherantly be absorbed. This would be because any time the pressure gradients were un-alligned by a low pressure region, some water would evaporate there and bring the pressure up to the same level as the higher pressure region. Which would make the pressure/density gradients uniform and thus reduce vorticity.

Worth a read: http://www.navier-stokes.net/nsvcr.htm

Does that make sense? In lower pressure regions the water would evaporate and raise the pressure to equal the other regions more or less. Which would stabilize the flow in the compressor and allow it to flow more easily, and with less heat generated from vorticity. That would also stabilize the boundary layer.

Adrian~

[Greenv8s]

Its clear that you guys have a far better understanding of this than I do! Your comments have stirred up a couple of thoughts though.

Gut feeling :roll: is that with WI just upstream of the compressor relatively little evaporation will occur outside the compressor, most evaporation would occur as the air is heated and shredded inside the compressor. Since the heat will be generated during compression, and the compression occurs progressively, and there will be some latency between the air being heated and the resulting water evaporation, it may be that water evaporation also occurs relatively late within the compression process. I can sort of imagine a plot of temperature against time for a typical molecule passing through the compressor, and seeing the temperature rise quickly for the dry case and slower for the wet case IYSWIM. If so, I expect the impact on the compressor entry conditions would be relatively small.

The second thought is that a centrefugal compressor is highly sensitive to density. As well as the 'each shovel holds twice as much air' analogy, it has been pointed out that the denser air will generate a higher outlet pressure for the same RPM or the same pressure at lower RPM. Does this reduced RPM (and presumably reduced air friction drag on the tubine) mean that the compressor will also (a) be more efficient and (b) spool up more quickly (in terms of flow and pressure ratio, even if it may be slower in rotor rpm terms)?

There are too many unknowns to speculate, but I would be extremely interested to see what difference it does make to have WI upstream rather than downstream.

[b_boy]

From Stealth316:

The wheel rotational speed, in revolutions per minute (rpm), at various values is shown on the flow map as a function of air flow rate and pressure ratio. When air flow is held constant (a vertical line on the flow map), faster rotation means a higher pressure ratio. When the PR is constant (a horizontal line on the flow map), faster rotation generally means more air flow. However, air flow does not appreciably increase after the outer tips of the compressor wheel are moving faster than the speed of sound. When the air flow reaches sonic speeds, the diffuser becomes choked and only very small increases in flow rate are possible even with large increases in wheel speed. Larger compressor wheels have maximum rotational speeds less than smaller wheels because of this limitation.

On the flow map above, the air flow regime to the right of the dotted line marking maximum wheel speed is called the choke area. The choke area is almost never noted on a flow map. To determine the choke area, you can drop a vertical line from about where the fastest wheel speed curve ends on the right side of the map. This vertical line is the approximate maximum air flow the compressor is capable of, regardless of efficiency or pressure ratio.

The key phrase here is just as you said, SaabTuner, it's not the wheel speed it's "when air flow reaches sonic speed." Big difference.

I also like the sentence about how the choke area is almost never noted on a compressor map (the map usually stops at 60% efficiency). That's great for us, we have some room to spare.

I wrote to a "turbomachinary" professor in England. His take is as we have noted, the gain in work with "wet compression" over and above cooling the air after compression (intercooling) is that the more air is compressed with the same amount of work energy applied--it's the shovels of air analogy. To recover, the same mass of compressed air without water injection, your need to turn the compressor more to make up for the loss in efficiency. He also said there is no combustion penalty up to 5% water:air mass. Unfortunately, he did not give much of a molecular take on things in fluid dynamics terms.

[ride5000]

what a great read.

thanks to all the posters.

ken

[Richard L]

I have been watching this thread with some interest. I'm beginning to become inspired to do a little testing. Currently I am only using one of the TEC3r's (www.getfuelinjected.com) mapable outputs to map WI (using Aquamists 2C system), with 2 nozzles post IC. I have another spare mapable output - and I'm beginning to think that placing another HSV in a line supplying a pre-compressor nozzle might be a worthwhile experiment. This would enable me to play with the water delivery and compare results of varying switch on points and delivery amounts independently of water that is being delivered post IC.

The only thing making me think twice at present is I am running a Garrett Disco Potato turbo (GT28RS, 0.86 AR compressor 76 trim, 0.64 AR turbine 62 trim) - so it is if anything oversized for a 1.6L engine making 280bhp (sized for future plans).

Worth making the effort? I have full data logging software that will allow repeatable power/torque graphs to be plotted if using the same stretch of road.

Slow MX5 has now all the tools to do whatever you consider to be a real-life test on the theory, must need some careful planning how to set it up properly so that the results can be analysed. It is so lucky thart he has data logging controller.

Do you think a temperature probe is a good addition?

[ryan]

Well, if the shoveling more are per turn is correct, I guess you would get a good power increase due to turbine does not need so much pressure to supply same air mass/pressure, less backpressure on motor will increase power output too.

this all sounds good, let's see the results, I would try on my car but I have twin turbo v6, so would need more hardware ie atleast 1 more hsv, but really 2 would be the one to make sure misting is o.k., - turbos are on oppersite sides of engine bay.

also, would injecting straight arfter air cleaner/maf sensor, not just before compressor be even more benaficial, ie longer time to cool the air down makeing it even more dense before it enters the compressor ie shoveling even more air per turn, or is the most benerfit from cooling the compression stage. maybe that could be tested also.

Cheers
Ryan

[hotrod]

A have a friend that has a small number of dual probe temperature gauges that can display delta T between the probes or either probe temp.

I will be buying one from him soon. I think its the best way to confirm the WI is running (ie if no temp drop, means back off the boost).

They are a bit expensive but if anyone wants to try one, let me know and I can get you contact info.

Larry

[SaabTuner]

I think a potentially easier idea (for a mass produced unit) would be two PTC resistors (change resistance based on temp, artificially heated, used in GM MAF sensors) ...

Put on a few inches before the WI, and one a few inches after ... when the resistance values are equal a light could come on, and the ECU could be informed that no WI was taking so that it would use a more conservative map.

Adrian~

[hotrod]

I agree for a mass produced unit something like that would be the best. I need to get the gauge to do some testing.

The only issue I see with temp sensor safety systems is the natural hysteresis time delay in reaction.

Suppose you had a logic circuit that gave a 1 ( WI okay) when the delta T > 20 deg F between the upstream and down stream sensor. It would give a 0 ( WI alarm) when the delta T was <= 20 deg F.

Set that up so if you sense power to the spray relay, and get a logic 1 then use the high boost map. If power to the spray relay and get logic 0 drop to a safe map and set a Check W I light.

possible situations:

A --- first use of WI in some period of time. both sensors stabilized at IAT (intake air temp). The down stream sensor would begin to cool very rapidly but there would be a small but finite delay in its temperature drop. So you would want a small wait interval following start of injection before the CWI light circuit checked for delta T between the sensors.

B --- WI in use, sensors stabilized at max delta T between up stream and down stream. How often does the CWI logic check the temps ( every 1/2 second ??)

C --- Sudden total failure of WI, down stream sensor stabilized at max delta T when WI fails. I see two issues. First if the sensor surface is well wetted with WI fluid, there will be a small period where its temperature will hold low due to evaporation of this fluid film and any fluid that drips from the nozzle ( ie partially blocked nozzle or surface wetting of interior of intake path). Second, if flow is not completely blocked but only impaired you would still see cooling but the amount would drop.

Looking at above there would need to be some tests to determine the ideal spacing between the temp sensors to get best detection of WI failure with minimum time delay, and minimum surface wetting of sensor surface to reduce detection errors.


Another possibility would be to use an IR diode detector pair that looks across the intake path down stream of the spray nozzle. There should be a very significant reduction / scattering of the IR light off the mist. If you have one detector in direct line with the emitter its sensed output should drop dramatically when the mist plume passed between the detector and emitter. If you also placed a second detector at right angles, or even in line on the same side as the emitter beam it would only see scatter if the mist plume was present due to back scatter off the mist dropplets.

You might also be able to detect the presence of the mist plume with a capacitance system that would detect the change in the capacitance between two plates on opposite sides of the intake path. Or a similar inductance system that detected the change in inductance of a coil surrounding the intake path when the plume was present.

A third possiblity would be a charge transfer sensor. Place a voltage potential on the injection nozzle, and a collector screen down stream in the mist plume. You should get some small electrical current transfer in the mist dropplets from the nozzle to the collector screen. This charge transfer should vary directly in proportion to the mass volume of spray dropplets that impact the collector screen.

I think a combination of a 2 or 3 way detector composed of the delta T photo optical back scatter or intake air capacitance/inductance , and charge transfere detectors would be the most fool proof.

It would just require some testing to see the range in detection values and possible false alarm potential for each system.

Okay Saabtuner and I have discribed 5-6 workable detection concepts, all you experimental electronics types, --- go build some prototypes and see if any of these work well enough to use. :twisted:

Larry

[hotrod]

Richard:

Perhaps, the last couple posts belong over in the WI safety discussion, or I could double post it over there if you prefer, because there is some value to seeing these discussions in context as they were developed.

Just a thought --- Saabtuner -- what do you think?

Larry

[SaabTuner]

I agree. Some of the posts should be moved as they are more safety related than pre-comp related.

The problem with the sensor being wetted is moot, it is heated to 220C above ambient to prevent this already. The amount of current needed to maintain that temperature determines the mass-flow through the system.

Even if it were wetted, so too would the rest of your induction be. Therefore the WI effect would taper off slowly. Which is exactly what was seen in that dissertation I posted from Linkoping on WI.

Since these are resistors, and very accurate ones at that, all you really need to do is this:

1. Have the sensors active at all times, but passive.

2. Have a boost pressure solenoid that activates the safety circuit just as your WI comes (at a fixed boost pressure) on and deactivates as it goes off.

3. Have the first PTC resistor (RH) heated to 220C above ambient. (Ambient temp is measured by a third special PTC resistor (RC)) The amount of voltage required to keep that sensor at that temperature differential will depend on the mass flow of the air across the (RH) sensor.

4. Have a third resistor post-WI "wet" (RH) which is heated to 220C above the temp of the special resistor (RC), and if the value is equal, or near to, your "dry" PTC (RH) resistor the power to your boost solenoid should be cut (thus limiting you to base boost, which should be knock free) ...

5. If the voltage required to maintain the second "wet" (RH) PTC resistor increases significantly while the safety system is active (step 2), which means the power required to keep it hot has dropped, power should be cut to the boost solenoid. The level of drop which causes this should be adjustable becuase the WI may not produce a 100% constant cooling effect.

Make sense? Should be relatively simple to setup with the right tools and electronics knowledge. The parts could be scavengened from any GM Mass AirFlow sensor. (Each contains two PTC (RH) resistors and another PTC (RC) resistor.) Pricing is in the $200 though.

If your car is injecting pre-turbo and already has a MAF sensor, you'd just need a second MAF sensor. The value of the second one should just be significantly higher.

Adrian~

[dsmtuned]

Thanks to everyone for your input in this informative and idea inspiring thread!

I am wondering how colder ambient air temps would affect the evaporation of pre-compressor water/ETOH droplets. Hotrod and I live in Colorado and can experience cold winters. Larry, have you had any experience with your pre-comp injection in the winter time? I just wonder if the water mixture would have more of a chance of not evaporating fully and possibly damaging compressor blades.


Another thought relating to the idea of increased air mass and O2 content with pre-comp WI. Some posts here have suggested that the increased air mass due to colder/denser air would not have more O2 content because more air is not actually entering the system through the intake. Wouldn't the O2 content of the air be increased with the addition of WI prior to the turbo because H2O and methanol themselves contain O2? I thought this was one reason for using methanol injected after the turbo because it has a high O2 content (in addition to it's knock suppression qualities). What happens at the molecular level to the O2 in the water and/or methanol at the time of evaporation?


-Craig

[hotrod]

I usually shut off my WI when temps get below about 55 deg F. I simply don't need it for knock supression at those temps with my current set up, and it makes the mixture too cold -- leads to compressor surge problems on my setup in cold temps.



The oxygen in the water is not available for combustion so you can ignore it. The methanol does reduce the oxygen demand slightly as it consumes less oxygen as it burns than an equivalent weight of gasoline.

The colder intake air mass flow does increase the amount of oxygen available to burn simply due to its increased density, but the percent oxygen in the intake air is essentially unchanged from normal atmospheric 21% O2. You can ignore the very small dilution effect of the extra water vapor.


For example if you have dry air (0% RH) at sea level pressure, 59.9 deg F and 29.9 in/hg air pressure and call that 100% air density, if you bring it up to 100% humidity, the density is now 99.35%.

The effect on engine power from humidity change is quite small. If you take an engine with intake air temp of 102 deg F, 0% RH, and absolute air pressure of 29.9 in/hg and call that 100% engine power. If you changed only the relative humidity of the intake air the engine should make 92% of its rated power. The dyno correcton factor for that humidy change would be 1.087.

Okay you lost 8.7% power, but if you now look at the temperature drop effect of Water injection on power output, and drop the intake air temp to 62 deg F. you get your power back. This would be due to the cooling effect of the water evaporation in the intake tract, you now get an engine power output of 102.2%. So you have a net gain in power of 2.2% in spite of the dilution effects of the water thanks to its intake cooling.

Obviously in the real world you'd never start at 0% humidity, but on the same token most of us get extra evaporative cooling from the methanol in the WI mix. The example does show that even in a worst case condition you get more back than you loose.

This is totally ignoring the advantages you get on a boosted engine with increased detonation threshold and being able to crank up the boost.

If you want to play with some numbers and read a fascinating web page on atmospheric pressure effects on engine power check out:

http://wahiduddin.net/calc/calc_hp.htm

Follow the links he has a TON of information on his web site.
In fact we probably should invite him to participate in this discussion.
I'm going to send him an email and see if he'd like to join in on this thread.



Larry

[slowMX5]

Slow MX5 has now all the tools to do whatever you consider to be a real-life test on the theory, must need some careful planning how to set it up properly so that the results can be analysed. It is so lucky thart he has data logging controller.

Do you think a temperature probe is a good addition?

As Richard says I now have a HSV sitting in my garage. I hope to install the HSV and run the wires to control it with a second GPO this weekend - which means that I will be in position to start experimenting next week (assuming I have dry roads for power tuning and WOT data logging). This will provide me with 2 HSVs, independantly mapped, one supplying water post IC and the other supplying water pre-compressor.
So the questions are:
- where to install the pre-compressor nozzle?
- how much water to map to the pre-compressor nozzle (and when to switch water on)?
- what combinations of injection should I try (post IC water off, pre-compressor on etc)?
- anything else?

I have full data logging ability as well, so checking for improved spoolup during say a 5th gear low rpm roll on can be done as well, and I should be able to supply other data that may be of interest.

So guidance and suggestions will be appreciated. My car can be seen at the sites below to help with deciding where to site the pre-compressor nozzle (how about pushing the 4mm ID nylon line through the rubber end of the filter? (doesn't leave holes behind in my intake piping!).

BTW I'm injecting a 20/80 methanol/water mix

Steve.

[Richard L]

I am surprised that there is so far no offers to your questions.

A long thread on theory and the some practical answers could be just a few posts ahead.

I guess you will have to go alone and place the jet in front of the turbo. I would try a 0.5mm jet first (pre compressor) and plot the result against a 0.5mm jet (post compressor) and post IC with the same jet. The logged data can be compared.

What do you think?

[hotrod]

Sorry for the delay getting back to the board, just started a new job but on 12 hour midnight shifts so --------- Argggg --- its friday.


Lets see:

- where to install the pre-compressor nozzle?

My nozzle is positioned about 20 inches so if you can I'd suggest about 0.5 meters for your setup.

- how much water to map to the pre-compressor nozzle (and when to switch water on)?

Start with about 2% of max air flow mass, that will give you near the ideal 3% or so at peak torque rpms. I'd turn it on about 50% - 70% of max boost. If you turn it on too early it can send the compressor into surge, If it operates close to the surge line at mid range rpms just as it comes on boost. Low air flow, but fairly high boost, places the knee of the air flow plot at its closest approach to the surge line on many turbos. I'm getting into surge on my new turbo and may need to install an rpm window switch so the pre-compressor WI can't come on until I get over a certain rpm.

Summit racing has a cheap rpm switch that I'm looking at.

- what combinations of injection should I try (post IC water off, pre-compressor on etc)?

For documentation I'd like to see the Pre-turbo only values, then numbers for both pre-turbo and post IC. I suspect you will want to delay turn on of the post IC injection until you reach the boost level that begins to stress the Intercooler and increase manifold air temps.

- anything else?

Yes, a couple passes with the pre-compressor injection on with pure water, and the common 50/50 mix would be nice if you can. That would give 3 reference points so we could guess-ti-mate what other injection mixes would give.

I'd throw out a pre-test guess of around 25 - 30 deg F drop on water alone and around 40-50 deg F drop on 50/50 mix.


Looking forward to your numbers.

Oh almost forgot, please record ambient temp and humidity conditions if you can.

Larry

[Richard L]

Does anyone agree that having two identical jets, one before and one after the impeller should give a very good indication of how the two work.

Once the data are logged, it will be nice to compare the results.

[SaabTuner]

Does anyone agree that having two identical jets, one before and one after the impeller should give a very good indication of how the two work.

Not necessarily. Only if the two jets are both flowing the same amount of water. The same size jet on the high pressure side of the turbo should flow a bit less water shouldn't it?

If the water amount injected is equal, it should give a good comparison of the difference in effect.

If the same jet-size is used, it will still be useful in determining proper jet-size before and after the impeller.

Adrian~

[Richard L]

Does anyone agree that having two identical jets, one before and one after the impeller should give a very good indication of how the two work.

Not necessarily. Only if the two jets are both flowing the same amount of water. The same size jet on the high pressure side of the turbo should flow a bit less water shouldn't it?

If the water amount injected is equal, it should give a good comparison of the difference in effect.

If the same jet-size is used, it will still be useful in determining proper jet-size before and after the impeller.

Adrian~


This is perfectly true, the pressure must be the same for both jets at the point of injection. The water pressure can easily be adjusted between 2-10bar on the 2c/2d. The post-turbo jet will run at a pressure higher, equal to the boost pressure.

[Richard L]

An example of Bert Waite using pre-compressor water injection to increase the power of his turbines engine.

http://www.aquamist.co.uk/forum/gallery/bert/1.jpg

http://www.aquamist.co.uk/forum/gallery/bert/2.jpg

http://www.aquamist.co.uk/forum/gallery/bert/3.jpg

He has also added this on his email to me:

I finished the Turbine project that I was working on. We went from 1,400 HP to 1,800 HP with the water injection system. The addition of the water to the saturation point not only increased the turbine compressor efficiency, but allowed us to add additional fuel due to temperature management. THANKS FOR ALL YOUR HELP!


He has been using water injection for many years and here are a few of his past and present projects:

In the Lotus application, I injected water before the turbo and after the intercooler. Injecting before the turbo or any supercharger/compressor , increases the efficiency of the compressor by increasing the density of the air. This one one of the main advantages in the turbine application.

http://www.aquamist.co.uk/forum/gallery/bert/4.jpg
http://www.aquamist.co.uk/forum/gallery/bert/5.jpg

I am working on a new project now that I am back on my feet. It is a boat like the turbine project. It is powered by two 6.6 liter small block Chevy engines with Whipple super chargers. It is intercooled and running about 1 bar of boost. The engines make 850 HP on a mix of 100 LL av gas and 110 octane racing fuel. The boat is a 8.5 meter cat and should run about 241 km/hr.
I would like to run the boat on straight 100 octane av gas with the addition of a water injection system. According to my calculations, each engine would need between 300 ml and 500 ml of water per minute or a total of up to 1,000 ml. Does that sound correct to you? At that rate, I am considering using the Shurflo 8030-813-239 pump which is capable of 9.5 bars at 3.3 l/m.
http://www.aquamist.co.uk/forum/gallery/bert/8.jpg

[Richard L]

So it is agreed that tow identical jet but alter the water line pressure for each jet so the back pressure will be the same for both jets.

Anyone else think there are other factors that we need to consider before slowMX5 starts his test?

[PuntoRex]

Installed mine this evening :D

I started yesterday, prepared the jet adaptor fitting:
http://home.pchome.com.tw/personal/rexchiou/JetAdaptoronHose.JPG

The adaptor is fitted on a piece of 2mm alu. I hammered the alu board into shape & then drilled/tapped for the adaptor. Some silicon adhesive (gasket maker) was used to fix the assembly onto the hose. ( That was really a mess to work with ) ( And the hose was drilled previously of course ) Additional nylon ties secure the whole thing one step furthur. I hope the fitting can be trouble-free.

I did hesitate a lot about the jet location. For better access, it would be too close to my hot wire MAF. I was afraid, on such location, I'd experiece the hiccup as Hodrod did. Avoiding that, I'd have to place the jet just a few inches ahead of the compressor.

That's just what I did. On the picture above, the air flow is from left to right. The branch facing down in the middle is the outlet of recirc BOV ( which is 1" ID FYR ). The elbow on the right leads to compressor. You can see the jet is quite close to that.

The following picture shows it better:
http://home.pchome.com.tw/personal/rexchiou/JetonElbow.JPG

It's about 6" to the end of the elbow, another 2" to the compressor blades.


I took 0.5mm jet on the pre-comp injection, 0.6 for pre-throttle. ( I ran 0.7 on pre-throttle only previously )

Since I have no check valves on hand right now, so I can just T the two jets directly. I know there would be some cross-flow & siphon, but I can't do anything about it. So... I arranged bigger jet on the pressured side, hopefully it could offset the cross flow somewhat. (I may be totally wrong & just fool myself)

Yet another thing that's not so optimum is that the pre-comp jet is quite far from HSV, about 3ft.

Anyway, it works!

Lukily I didn't experience the WOT bog down as Gelf. Yet not so lucky as Hodrod to get a rise in boost. Car just pulls harder on boost, especially in the midrange. The performance gain is easily sensible with a quiter turbo hiss.

This is really a very effective & worthy mod to me. Two thumbs up. Now I'm just waiting for Richard's mail of my check valves & other accessories to make the whole thing complete.

Thanks every contributor on this thread :wink:

[masterp2]

I have been following this thread with great interest. I am trying to develop a WI process for the duramax TDiesel with 30 lbs of boost. I have had initial reports that HP increases with it are not as dramatic as gas, especially the "tuned" trucks. I'm not sure why this would be, maybe someone could explain.

In any event, going to use it to reduce EGT's that are quite high, and to aid tow vehicles that overheat (seems the CAC is imparting too much heat to the rad). Considering 2 stages. Stage 1-towing-15 psi-2 nozzles-post turbo pre-CAC. Stage 2-performance-throttle switched at the floor-2 additional nozzles, 1 pre turbo-1 intake

Michael

[hotrod]

You can probably get some additional good feed back in the diesel forum as well as on this thread.

Just need some back ground info. I'm not familiar with the acronym CAC I assume your refering to an after cooler ?

Most of the conventional wisdom is that injecting between the turbo and the intercooler/after cooler is less effective than injecting after the last stage of intercooling/after cooling.

Are you using a pure water injection or a mixture of water and alcohol?

Water and methanol mixtures are supposed to produce some very impressive performance gains and lower EGT's. The bad news is the bump in power is addictive and too much methanol can push the engine to the point of blowing head gaskets due to high cylinder pressures, but the increase in torque is supposed to be very impressive.

The pre-turbo injection is most useful when your maxing out the airflow capacity of the turbo compressor at high rpm.

Right now my system is set up as pre-turbo only, and triggered with a pressure switch, but I plan on changing over to a system that is a bit more complex.

I'm going to add an rpm switch in line with the pressure switch so the pre-compressor injection only comes on after the engine gets to about 70% of max rpm. My experience recently also leads me to consider adding a temp sensor in the intake tract so that once the intake air temp drops below about 70 deg F. the pre-compressor injection is completely disabled.

Larry

[masterp2]

CAC, charge air cooler, IC. The stock Garrett turbo has 25 psi stock, 30 when chiped. A few people have played around with WI only getting 15-30 HP, 5% increases, from the 300-400 that are coming from the stocker, with 1200-1400EGT's.

That is using WW fluid, with meth.



Michael

[hotrod]

Thanks I figured it was and intercooler, but couldn't figure out the usage. Unfortunately each automotive enthusiast group develops their own common usage acronyms and some times the alphabet soup is hard to translate.

Second question, is your CAC an air to air system postioned infront of the radiator? If so then the heat transfer to the radiator would be due to the CAC pre-heating the air going through the radiator.

Two options come to mind, first under the heading of keeping it simple, you could also look at direct water spray on the outside of the CAC or on the radiator or both, rather than injection in the air stream between the turbo and the CAC.

The other way to take heat load off the cooling system is to add an oil cooler. The oil moves quite a bit of engine heat. On the WRX in stock configuration it uses an oil cooler that passes heat to the engine coolant. Folks have had good luck adding an air cooled oil cooler and passing that heat directly to the air rather than the radiator through the coolant.


When you refer to "gas" are you talking about propane injection?

The pre-compressor injection will only increase the max air flow by a few percent, so your 5% gain seems about right in that regard.

As I understand the effects on diesels you need to add some fuel to make full use of the air that you are already flowing so if you upped your fuel flow and then held the EGT temps in check with WI you should get more power than 5%

Larry

[slowMX5]

I must apologise for being slow to get my testing done - but a combination of a dead laptop (couldn't data log) and poor weather here in SW England has slowed me down. Doing 3rd gear WOT power runs on damp/wet roads isn't possible (it's dangerous and wheel spin will mess up the power/torque calcs) - so I am waiting on completely dry roads - or at least for my test stretch of road to be dry.

Hopefully this will happen by next weekend and I'll get the tests people have asked for done. I'll either post details of each test with torque and power figures or post complete comparison power/torque graphs overlaid on a website. PuntoRex's findings are encouraging.

[masterp2]

Second question, is your CAC an air to air system postioned infront of the radiator? If so then the heat transfer to the radiator would be due to the CAC pre-heating the air going through the radiator. Yes, exactly

Two options come to mind, first under the heading of keeping it simple, you could also look at direct water spray on the outside of the CAC or on the radiator or both, rather than injection in the air stream between the turbo and the CAC. But requires too much water and not as much fun

The other way to take heat load off the cooling system is to add an oil cooler. The oil moves quite a bit of engine heat. On the WRX in stock configuration it uses an oil cooler that passes heat to the engine coolant. Folks have had good luck adding an air cooled oil cooler and passing that heat directly to the air rather than the radiator through the coolant.


When you refer to "gas" are you talking about propane injection? No, gasoline vs diesel

The pre-compressor injection will only increase the max air flow by a few percent, so your 5% gain seems about right in that regard.

As I understand the effects on diesels you need to add some fuel to make full use of the air that you are already flowing so if you upped your fuel flow and then held the EGT temps in check with WI you should get more power than 5%

Larry

Correct, a diesel is just a big air pump, meter in some fuel (fuel is subject to timing vs a spark) and compress until it bangs. At first glance, detonation appears to not be the issue it is in gassers.

I guess I am not clear why water-meth solution added pre turbo, vs post IC, results in less power added.

[hotrod]

Pre-compressor injection is a special case situation.

In a perfect world the ideal place to inject WI water/methanol mix would be after the CAC and as far as practical before the intake manifold. That gives you time for the spray to mix well with the intake, and you get the additional cooling of the air charge from evaporation . That additional reduction in charge air temp, increases the VE of the engine (ie you actually get more air in on each intake stroke for the same manifold pressure)

On the other hand, pre-compressor injection is a trade off. By injecting ahead of the compressor it increases the mass flow through the compressor so it is very helpful on engines that have undersized compressors. The down side of pre-compressor injection, is that it reduces the effeciency of the CAC slightly.

An intercoolers heat dissipation rate is controlled by two things, the mass flow of cooling air through the intercooler, and the average temperature differential between the charge air and the cooling air. Cooling varies as a log function of the temp difference.

If you inject the WI mist prior to the compressor you end up with a cooler discharge from the turbocharger for a given boost condition. That means a smaller temperature differential between the charge air and the cooling air flow --- result, your manifold air temp can actually be slightly higher with pre-compressor injection only, than it would be if you injected the same amount of fluid in the post CAC location.

If you have an adequate sized turbocharger compressor using pre-compressor injection is not as productive as the same injection rate post CAC.

On a car with a significantly undersized turbocharger, the increase in mass flow through the turbocharger out weighs the slight reduction in cooling from the intercooler.

Hope that makes sense?

That is why with my new turbo I will be limiting the pre-compressor injection only to high temp, high flow situations, as at moderate boost and engine rpm the turbo has plenty of flow.

Larry

[masterp2]

I completely understand your explanation, nicely done. As a Chemical Engineer, I have a good working knowledge of the thermo dynamic.

Now the balancing act for my unique application: How to design one WI system that serves 2 purposes. Cooling enhancement for hi-load towing focusing on lowering CAC and egt temps, plus a second (stage 2)performance aspect (has to be fun afterall :lol: ).

Question: Can you make a guess at how much flow to dedicate to pre-CAC, post turbo misting to cool that charge 100 degrees (350 to 250) on a 30% RH day? I assume it would be the same flow, whether pre-turbo or post-turbo? It's a 6.6 l motor (diesel). Diesel posts seem to suggest 400 cfm moving at full tilt, but not sure on this.

Using your rationalle, I don't want to overfeed this stage, just kill the fire a bit. Save the big water for the 2nd performance stage. Unless you have another idea.

[hotrod]

Well I used a 3 and a 4 gal/hr nozzle on my pre-compressor WI setup on a turbo that was rated with a max flow of 360 CFM. The 4 gal/hr was a bit too much, and the 3 seemed to be about right. Since I was pushing that turbo very hard a lot of that water went to cool the compressor discharge to sane levels.

In your case you have:
6.6 L displacement @ 30 psi, lets assume 4000 rpm --- engine needs about --

6.6 L = 403 CID x 2000/1728 = 466 CFM, at 30 psi equals a pressure ratio of `approx 2:1 . Your engine should need about 932 CFM at 4000 rpm or 64.3 lb/min of air flow. (this is air flow into the compressor inlet at atmospheric pressure).

Not sure what the VE for a diesel engine is so this is at 100% VE, for a gasonline engine you'd mulitply by about .85 to get a real world air flow.

If we do a quick on the napkin computation of the specific heat of that air flow we get:

64.3 lb/min x .024 btu/lb x 100 deg F delta T = 154.3 BTU/min deg F

Latent heat of water = 970 BTU/lb deg F at atmospheric pressure. (it will be a little lower at high pressure but this is close enough for a ball park computation).

To absorbe that heat flow you, would need to evaporate about 0.16 lb of water /min or 73 ml/min of water. That would be just over a 1 gal/hr water flow, or about the same as an Aquamist 0.4 mm nozzle at 40 psi line pressure.

Not trying to be condescending by working it out, but just wanted to show where the numbers were coming from, so everyone could follow the reasoning.

Larry

[TurboGTi]

I;m not sure if this was disccussed but can i run straight nethanol and still inject Pre-turbo?

Will it have any effect on the turbo or the inlet temperature?

[hotrod]

Yes you could inject pure methanol pre-compressor, but you would get more cooling effect with a water methanol mix.

Larry

[masterp2]

Larry, excellent work on explaining the math. How about another?

How much water does it take to bring 900cfm from 10% humidity to 100% humidity? What is the resultant temp if ambient is 120 degrees, assuming a 70% efficient, 30 psi turbo.

[hotrod]

I can't give you an exact number as I lost a reference book I used to have on compressed air systems. If you need an exact answer you might try calling a company that specializes in industrial compressed air systems. They have to work out the water weight in the compressed air to design, drying systems so the user does not end up getting wet compressed air at their tools.

I found one reference that gives some ball park values on a chart of lb water in lb air at different temps and pressures.

Airs ability to hold water vapor changes significantly at higher pressures so it is not a simple relationship.

As you can see below as pressure goes up, the amount of water vapor the air can carry at a given temp goes down. Also as the air temp goes down so does the water capacity.

Bottom line if you saturate the air going into the compressor, as it leaves it will be quite dry, while it is still hot, but after being cooled in the CAC it will become super-saturated as the temp approaches the compressor inlet temp. In other words if you had a transparent intercooler you'd probably see a dense fog going through the throttle body.

A secondary conclusion from the above fog formation on cooling just came to me ! The temperature where your charge air becomes supersaturated and forms a water fog will be a hard limit to the minimum temperature the intercooler can reduce the charge air temp to. Once water fog begins to form, the latent heat of condensation of the water vapor released as the charge cools will exactly balance the heat taken out by the intercooler. This is likely the reason pre-compressor injection can appear to reduce the effeciency of the intercooler.

900 CFM would be about 63 lbs / min air flow at 1 atm


From the chart in the Chemicals Engineers handbook (Perry and Chilton)
5th editon, Fig 3-4

Air at:
1 atmosphere and 100 deg F holds 0.04 lb water / lb air
2 atmosphere and 100 deg F holds 0.03 lb water / lb air
3 atmosphere and 100 deg F holds 0.015 lb water / lb air

1 atmosphere and 125 deg F = 0.09 lb water /lb air
2 atmosphere and 125 deg F = 0.045 lb water /lb air
3 atmosphere and 125 deg F = 0.03 lb water /lb air


1 atmosphere and 200 deg F holds > 1 lb water / lb air
2 atmosphere and 200 deg F holds 0.43 lb water / lb air
3 atmoshpere and 200 deg F holds 0.23 lb water / lb air

Assuming your intake air is completely dry:

If you take a small turbo like I have on the WRX which has a max flow of 35 lb / min, and the inlet temp is 100 deg F, and the exit temp from the intercooler is 125 deg F (at 2 bar boost = 3 bar absolute pressure), that 35 lb of air could hold a max of about (.03 x 35) or 1.05 lb or water per minute. That is just over 475 cc/min of liquid water completely evaporated.

If your intercooler is heat soaked, then that same air flow at 200 deg F and 3 bar absolute would hold almost 8 lbs of water per minute or nearly 3654 cc / min of liquid water completely evaporated.


We know that the pre- compressor injection drastically cools the intake charge, so if the inlet air is cooled to 50 deg F by the water evaporation before it hits the compressor, that cool air can only hold a max of about 0.008 lb water / lb of air, or .28 lb of water or 127 cc/min will be able to evaporate before it arrives at the compressor. (that is equal to about .8% of mass air flow).

If the air leaves the compressor at 200 deg F and 3 bar absolute pressure then on exit it could hold a max of 0.23 lb water / lb air. That would equal 8.05 lb water / lb air (3655 cc/min).

If your injecting at 3% of air flow then you injected only 1.05 lb water, so the relative humidity of the air leaving the compressor will be less than 13%.

Now you cool it down to 125 deg as it goes through the intercooler and you have that same 1.05 lb water but it is traveling in an air stream that is now is only capable of holding about 0.03 lb water / lb air, or 1.05 lb water --- surprise surprise your at 100% humidity as the air leaves the intercooler.

( I didn't plan this to come out at exactly saturation but I'll take the lucky break ;) )

For your example of 900 CFM you have 63 lbs/min air flow in. If you get 30 psi boost ( 3 bar absolute) and it exits the CAC (intercooler) at 125 deg F then at 100 % RH it could hold 0.03 x 63 = 1.89 lb water / min or 858 cc/min at saturation.


Someone check my numbers but I think that is all correct!
Larry

[masterp2]

This is so funny, I was browsing threads (after posting earlier today) and came across this. Said to myself, this is the response to the question I posed earlier. It took 5 minutes to figure out that this was the same thread!

Now I have to go back and read it better. Thanks Larry.

[masterp2]

Yes, the hot air really holds a LOT more water. Then it stands to reason that, if misted efficiently, more total water can be evaporated post-IC only, vs pre-turbo plus post-IC??

Does the turbo performance enhancement of pre-turbo misting make it better performing anyway?

[hotrod]

I think the answer to your question can only be answered in context of a specific engine turbo setup, and your specific performance goals.

If the turbo is of adequate size, so it can feed the engine without straining at max rpm, and your going for max power, then post intercooler would likely be the best in my opinion.

If the turbo is a bit undersized and your pushing its air delivery capability, then I think either pre-compressor or a bit of both would be best.

There is also the consideration about ease of implementation. If its nearly impossible to squeeze a spray nozzle inbetween the intercooler and the throttle body (which is almost the case for the WRX) then the simplicity of pre-compressor injection has to be considered as well.

There may also be an advantage to over injection pre-compressor or pre-intercooler and then cooling to super saturation. It would be hard to beat a true fog of microscopic water dropplets for detonation protection I suspect. Only testing will tell for sure.

It's obvious that the WWII military aircraft did quite well with pre-compressor injection, so it may be that the only thing that really counts is the suspended water fraction and it may not matter much how it gets there once its in the cylinder.

I guess the question is, which of several problems is the most urgent for you to resolve on YOUR setup. Detonation (not an issue for a diesel), might depend on the degree of evaporation vs suspended dropplets in the combustion chamber, high EGT's (probably only dependent on the total water fraction, max power ( depends on highest VE so pre-intake valve cooling would dominate).

I think we sometimes forget that WI can be used to resolve several different problems and the setup that most effectively solves one, may be less effective for another.

Larry

[masterp2]

Well put. Evaporation would seem to me important also from the combustibility aspect. Using a meth solution, meth drops (liquid) are worthless to cylinder power enhancement (correct me if I'm wrong). Only once it is a vapor, is it explosive. Dry air (desert SW) would seem to be a real advantage, as so much water AND meth can be vaporized. (thinking out loud)

A friend on-line claims he has reached 127HP using WW fluid, in dry Colorado, 6000 ft altitude, won't share much of the details, but he claims to have unusual nozzles and very high pressure.

[hotrod]

There are several WI diesels up here that are very quick for their size. A couple of them run in the 13 second range. It's kind of funny to see a heavy duty diesel pickup beat a supposedly fast car in the quarter mile. We also have a turbocharged diesel dragster up here that runs in the low 10's high 9's. It's so much quieter than the AA fuelers that you can litterally hear its slicks chirping the full length of the strip.

As you know from where you live, ultra-low humidities are pretty common up here, and in the desert sw. During our forest fire summer a couple years ago, the fire fighters were absolutely stunned when they got morning fire weather reports that they had temps of 80+ deg F and RH humidities of 6%.

That is one of the reasons I have such good luck in the summer with the pre-compressor injection.

Your correct that only the vapors are flamable, but dropplet size is also important. Very small dropplets of fuel vaporize very quickly during combustion due to their very large surface area to volume ratio. It's only an issue if the fuel dropplets are too big to evaporate during the combustion time. This becomes a problem with fuels that have high 90% evaporation temperatures, (like xylene spiked pump gas) that can end up blowing still burning fuel drops out the exhaust valve, giving very high EGT's.

Larry

[masterp2]

I guess the question is, which of several problems is the most urgent for you to resolve on YOUR setup. ... max power ( depends on highest VE so pre-intake valve cooling would dominate).

Larry

Could you elaborate on this part? You are talking air charge density I think. Drops vaporizing in great quantity at the 1000 F valve opening to do what, lower temps and condense more air into the cylinder before detonation? The liquid is important then. Does the meth enter in?

[hotrod]

There are some differences between diesels and gasoline engines so that needs to be considered. As far as VE increase is concerned, it is clear that charge air cooling due to evaporation helps inprove VE all else being equal.

In one of the NACA studies they tried injecting the WI mix both well before the intake valves and immediately before the intake valve. The postion that was "pre-vaporization tank" was several feet from the intake valve and the other injection point was only inches from the intake valve.

Study E5E18 page 5.

"For vaporization-tank injection, the mixture temperature decreased as more internal coolant was injected until the fuel-air mixture became saturated (fig 12) and then the mixture temperature remained relatively constant. In a corresponding manner, the power at peak-power spark advance increased because of the increased charge weight inducted into the cylinder until the internal coolant-fuel ratio for saturation was reached ( fig 13), at which point the power leveled off. After complete saturation of the incoming mixture with the internal coolants, any additional cooling of the mixture must occur after the intake valve closes, which makes it impossible to increase engine power through an effect on air flow. From this point, the power obtainable at peak-power spark advance dropped slightly as the internal coolant-fuel ratio was increased because some heat of vaporization was extracted from the air during the compression stroke, resulting in a decrease in cycle efficiency. When the internal coolant was injected at the manifold elbow, the power increase at peak-power spark advance was relatively small (fig 13) because there was insufficient time for charge cooling before the intake valve closed."


As you can see on a gasoline engine the intake charge cooling due to evaporation of the WI mixture can have an effect on maximum cylinder filling and hence power up to the point you reach 100% humidity (saturation) of the intake charge, AND if there is suffecient time for the evaporation to take place. Spraying beyond, that will increase the cooling effect on the engine due to reduction in combustion chamber temperatures, but at a slight cost in power. You might be able to recover that loss in power by upping the boost more to levels that couldn't be reached at lower injection rates, so you need to consider the fact that you are not only changing cylinder filling but also how much boost, and spark advance you can get away with. This interaction of multiple variables means you have a very complex system to optimize.

Now in your case, your dealing with a diesel, which has too much air to begin with, so the power limit is mostly determined by the amount of fuel available to burn and maximum EGT numbers, and how much cylinder pressure you can tolerate before you start blowing head gaskets or lifting the heads.

If you went with a WI mix that was pure water or low in alcohol content you would maximize cooling at the cost of some reduction in the power gain possible. If you upped the alcohol content you would be increasing the fuel load, and power.

If you increased the alcohol content injected by overspraying a lower alcohol mix, you should be able to get both a power increase and a strong cooling effect on the combustion temps. Only some experimentation would tell you the ideal balance of the variables. For testing purposes you may want to experiment with a 2 nozzle system where you inject water alcohol from one and straight water from the other. That would allow you to sort out the ideal total injection weight of water and alcohol for your goals.


My guess is that on diesels you will need to spend more time sorting out how much extra fuel you can accept (ie alcohol ) to reach your power goals without blowing head gaskets and then control cooling by increasing or decreasing the total WI spray mix and increasing or decreasing the alcohol fraction so you don't exceed your max additional fuel level.



As far as engine cooling due to WI, this paragraph is worth noting.

NACA report 756 ( page 71 , Conclusions item 3.)

Water injection had a marked cooling effect on the engine head and cylinder. The exhaust-valve guide was the only point on the head at which the temperature showed a tendency to increase with indicated mean effective pressure. The temperature was less, however, than that obtained with a straight fuel permitting equivalent power.

Larry

[masterp2]

This some very dynamic conversation. I do appreciate someone who understands the science behind result. I am going to read those reports, they (indirectly) seem to make a case for SOME pre-turbo injection for it's contibution to evaporation. Heres one for ya:

Meth, as a more volatile liquid than water, evaporates at greater rate than water. Given a droplet of mix, won't a greater fraction of meth evaporate per unit of water? Seems as though water will still be liquid when the meth is mostly vapor. Hypothetically, a WW fluid mix, 70/30, might be 90/10 by the time the remaining liquid hits the cylinder.

BTW, I don't plan on playing with proportions. Will use water or WW fluid.
Study E5E18-where can I see this?

[hotrod]

Your correct, the various chemicals will evaporate independently of each other. The methanol does not know or care what the water vapor content of the air is, and likewise the water does not know about the methanol.

(okay there is a very slight impact because methanol is so hygroscopic but not enough for us to worry about).

Due to this you are correct that as the air cools and water evaporation falls to near zero, methanol will still see an environment it can evporate into. That is one of the reasons your temperature drops in the pre-compressor intake tract, exceed the expected 30 deg F you would get with water only before it gets to 100% RH.

For any of the NACA studies you can usually find them with a google search of the form: +NACA +(study number)


A google on:
+NACA +E5E18

Returns the link in the first hit.

If the NASA server does not work -- they are in the process of "improving" the site. You can try the UK mirror. I personally think the recent changes in the report server are a step backwards but what do I know.

http://naca.central.cranfield.ac.uk/

It still has the old interface and lets you search on key words.

Larry

[masterp2]

Thank you. I thought I would mention, I used to fly B-52G models, which had WI for takeoff high GW.

10,000 lbs of water in 90 seconds (8 engines). I'll have to break out the manuals and look at where the rings were placed.

Funny looking at typewriter print. Now on to the patent office website to scout out other ideas.

[doctorfrag]

This thread really has been one of the most impressive threads I have read, thanks to all for a read, that has led me to getting littel work done this afternoon! :wink:

However, apart from a couple of subjective reports, there does seem to be little objective evidence for the use pre-impellar in cars, such as rolling road printouts etc.

I currently am running a Fiat coupe 20vt with a T28 hybrid 60 trim comp wheel, EVO6 uprated IC and aqumaist 1s system with DDS2.

I can run 340bhp with some mild detonation, 320bhp without, and have trimmed the base ignition map back to 303 bhp for safety.
However, although I get excellent spool up, just over 3000rpm, my current turbo simply runs off the edge of its comp map at 1.3bar and 6000rpm, and as you see torque drops off post 5500rpm quite dramatically.

http://homepages.which.net/~j.machen/Coupe/dualtorque.jpg

The Top line

I've been thinking of increasing my jet size from 0.4 mm to beyond 1mm to get on top of the detonation, but this thread got me thinking as well. I could inject pre-impellar, and as well as cooling the charge, I will increase my relatively (still small) turbo efficiency, and keep the same sppol-up, sounds all too good to be true!!

I would activatee it on boost says over 1 bar like I have at the moment and I can place it about 15" pre impellar.


Main concern will be as my IC entrance and exit are lower than the turbo ,will I run a risk of my IC simply filling up with condensed water?, and then running a risk of hydrolastic lock?

If I do this, if I have the money for experimentation, then I will run RR power graphs before/after so we can some objective results :)

Joe

[Richard L]

Near the bottom of this page, it shown power increase and temperature reduction also.

http://www.aquamist.co.uk/phpBB2/viewtopic.php?t=645

[masterp2]

Those are good concerns Joe.

If injection pre-turbo is implimented with moderation, I don't think you will have any problem. Freezeup more of a concern for many. I have been thinking along the lines of a temperature switch post-IC that would inhibit pre-turbo spray if not hot enough.

You really should play around with atomization, though. You need a very fine mist, more like fog. The biggest danger is impingement on the blades from larger particles. Maybe try fog nozzles under the highest pressure you can get, at least 100 psi. Multiple nozzle array vs single large one.

I am going to be trying some of this as our weather warms up.

[doctorfrag]

Thanks, hopefully I'll get a chance to chat with Richard as he really knows his stuff ;)

I need to get a mist rather than droplets as you say. Gelf got some moderate results, it is hard to know whether that was just due to more WI, not specifically pre-turbo. Also I think the pre-turbo will only work if you are out of your efficiency island with your turbo, in other words the turbo is too small

http://photos.fotango.com/p/eba00496747f00000002.jpg

Here you can see, on the right vertical line ,that at 6500 rpm, even at 1.2bar I am actually off the map :shock: :wink: , although I normally will only run 1bar or so at the redline. I am mapped to 1.3bar, so really my turbo is only efficient between 3000 and 4500 rpm, so if I can shift the island to the right I will be pleased :)

Joe

[masterp2]

Can you post a better image?

Present day feeling is that the effect of misting will move it to the right as you say. Usefull for all temps and speeds however, apparently most helpful at increasing compressor efficiency, so more of the work performed is applied to compression, less to heat loss: hence higher comp ratio (boost). Add a little fuel (meth), lower charge temps, all = power increase.

[doctorfrag]

http://www.turbocalculator.com/compressor-maps/t3-50.jpg

this is for a T3 '50' trim, as I can't get a compressor map for my turbo.

Mine is a T25/28 hybrid, the standard turbo on fiat coupe 20vt, and using a larger compressor wheel '60' trim, has a T25 inlet.

I'm going to try a 0.3mm jet preimpellar to see if this makes a good difference.

do you have any further thoughts? :)

regards

Joe

[masterp2]

It is my opinion that a nozzle that is oriented with the air stream is much more effective, than a nozzle pointed at the opposite wall of a conduit. Keeping water from recombining into larget drops and facilitating evaporation, rather than pooling so much, as evident in videos. A nozzle with a narrow angle of spray, vs a wide angle. 2 smaller nozzles, perhaps back to back, one pointing into the wind, one pointed leeward (with the wind). A simple plumbing T could do it. Just need some way to keep the fitting aligned, concentric with the conduit.

If you do something like this, i am very interested in your results. Just remember that you need heated air to hold the water, so be sure pre-turbo misting cannot occur till relatively high boost. The lower the boost cut-in, the more likely it is you will have condensation in the IC. And this will occur at higher speeds where the IC is most efficient. In reality, since misting is only a short duration event, I don't see any real need to be concerned about condensation, since even if there is some, it will quickly evaporate (within seconds) in a heat soaked IC. The atomization challenge is the big hurdle.

There is another article on this board about swirl flash atomization you might find interesting.

[doctorfrag]

Thankyou Michael, for you points. :smile:

I have a short induction route pre-turbo with a metal pipe and could run a jet into there, or jets.
My current jet comes on at 1 bar boost, and I would probably keep the same there, you have reassurred me about the condensation, as it would only come on when the boost is on, in test we recorded post turbo temperature of 135degC, so I don't think much water would hang around in the intercooler for long.

I will speak to richard about the atomisation, I run at 1.3 bar midrange which drops to around 1-1.1 bar at the redline, fitting a bigger downpipe will help the turbine efficiency, but this science has got me thinking about improving the efficiency, I amy even see more boost?, but as said befoer ned to get the water atomized.

regards

Joe

[masterp2]

1.3 bar pump pressure? Doesn't the aquamist go higher?

the following increase atomization effectiveness (smaller particle size):

higher pressure
lower viscosity (warmer fluid)
lower surface tension (analagous to lower viscosity but using surfactant)
smaller nozzle flow capacity


ideally 500-800 psi and fog nozzles would get you all the "fog" you need, but the practical limit to pressure, hoses and fittings is a limitation. Where this may not be so important in post IC misting, your effectiveness will be measured in your ability to achieve the smallest micron size possible, for the pre-turbo application. Some think it is only important in order to keep blade damage at bay. Not so. With only milliseconds of interaction during compression, it happens so fast that so much of the water cannot evaporate, because even a 50 micron droplet takes time. And the turboine efficiency will depend on the amount evaporating during compression. There is no help to turbine efficiency if water is evaporating after the heat has been produced, and the particle is downstream. Hence the repetition on stressing atomization.

I believe aquamist did not design for pre-turbo. While I have no numbers on aquamist nozzles, I doubt they are set up to "fog" under low pressure. Some nozzles are better than others though. Perhaps someone can host a particle distribution chart on aquamist, I have not seen one. Bete makes a PJ nozzle that is a impingement fog nozzle. I will be experimenting with them under 100-200 psi. I have some charted data on those. If interested in researching further, get aquainted with Sauder Mean Average particle size. You may find that even at 100 psi, 50-60 micron is a common SMA. Fog is considered 20 or less. So it's not an easy challenge. But you can play around. Maybe a little soap will make a big difference. Do an easy experiment. Get a narrow funnel and a quantity of water. Add a drop of DW detergent, see how much faster the water flows thru.

just some thoughts. Post your results. Looking forward to it.

[doctorfrag]

thanks Michael,

I meant 1.3 bar turbine pressure.

Yes, from what I've read Atomisation is the key, it is no point having larger drops, as you may as well have the nozzle downstream then.

I am running the Aquamist race pump, will keep you informed :cool:

regards

Joe

[janis]

I've got all the stuff at home to start pre-turbo injection, but I'm still not sure about a few things.

So I've got a few questions (and forgive me please if the answers are to be found in this thread already, although I read it completely, I might have missed something):

1) Is it wise to inject methanol into the turbo (80W/20M)? Isn't the stuff too aggresive for the poor turbo?

2) I got a suggestion (from Richard, thx) to inject pre airfilter. But I have a few reservations:
2a) Will my MAF survive?
2b) My airfilter is a Cold Air Intake, foam with oil. Will the methanol ruin the foam? Or will the water stay in the filter until it forms puddles and ruin both my MAF and the turbo?

Thanks,

Jains

[PuntoRex]

I suspect the air filter would gather the droplets into much bigger ones. The droplet sizes are supposedly bigger then the meshes of filter, thus physical contacts should happen & slow the droplets down....

And no matter how the droplets change, the MAF would certainly be affected by the water, especially hot wire type. It's hard to predict the changes to sensor reading & it's reliability.

:(

[Greenv8s]

It's hard to predict the changes to sensor reading & it's reliability.:(

I'm fairly confident the changes would be adverse. Hotwire sensors are very vulnerable to surface contamination, so water droplets reaching the element would cause an instantaneous overread and also affect the accuracy subsequently.

[PuntoRex]

Oh well, I think I must quit this pre-compressor injection camp, too. :sad: .....

I've been using the pre-compressor injection since early Nov. last year, so it's more than 5 months.

The jet was 0.5mm initially & swapped to 0.4 later, placed near an elbow in front of the compressor, only about 20cm away. It injects when high flow/high boost, controlled by a comparator circuit & a solenoid, metered by system 2D, and shares the pressure source with another pre-throttle jet.

The turbo in my car is installed deeply between fire wall and engine block, hardly accessable. Today I finally have some time to dig these all out & take a look. Actually I can not see it directly, through a camera instead.

Here's the pic:
I have contacted PuntoRex to retore this picture

As can be seen, the outer edges of the blades are damaged obviously.

I think the pressure is still not high enough, thus the droplets are too big for this application.

Anyway, in these days, the turbo was working fine & giving me 22psi of boost. And, on the driver's seat, I always feel the turbo & engine love this pre-compressor injection.

However, seeing this damage, I can not persuade myself to keep using it. I disassembled the jet, plugged the hole, took out the hoses & solenoid...

[Greenv8s]

That looks rather nasty. I've no experience with turbos, but I assume that much erosion is abnormal and is definitely caused by the water?

Obviously the blade speed would be highest at the outer edges which would explain why the damage occured there. If somebody clever could think of a way to focus the water droplets towards the center of the turbine, would that reduce the problem? Perhaps it would be possible by mounting the jet axially quite close to the turbine, or something like that? Always easy to speculate when it's somebody else who has to do the work and foot the bill. ;)

[PuntoRex]

Strangely enough, with this damage, I never felt the turbo was working badly. I didn't see boost drop or slow spool up.

So I'll keep using this turbo until it's bearing & seal go bad.

[Richard L]

The picture posted has exceptional details, goog steady hands.

It is most annoying that this has happened because we have had great results from it. I would if I can develope special 5-50um nozzle for you to continue this test rather let it continuing eroding the blades.

I need some time but I have a few ideas how to design this nozzle.

[hotrod]

Its my suspicion that that sort of damage ( which I also had) is due to small water dropplets collecting on the walls of the intake tract and running down the surface toward the turbo compressor like rain drops being blown up a windscreen.

Note that only the lead blades are eroded the second set of blades look nearly perfect. My theory is that the front most blade blasts these dropplets apart and they never make it back as far as the second set of blades.

If that theory is correct, than I think there is a solution but I need to build a test rig to verify it. If you installed a small step (about 1mm high) in the wall of the intake tract facing upstream toward the air filter, when the water dropplet hits that step is should be ripped apart by the fast moving air as it tries to climb the step. There are other possible configurations, like golf ball dimples on the interior of the intake tube ahead of the compressor to try, but I really need to build a flow bench with a few vacuum cleaner motors and some plexi tube so I can see what the water spray does.


It will cost me several hundred dollars to put together a home brew flow bench but I think it would be really useful to get live action shots of the water spraying into a fast moving air stream to see what nozzle locations and designs do in the real world, not in free quiet air.

Larry

[masterp2]

Punto Rex, How many miles on the turbo? How many miles using preturbo misting? It must be said. Unless you shot a pic of the turbo before you started misting it, you can't make a conclusive statement on the cause of the damage.

Larry,

Superb observations. You beat me. I am sure you are 100% correct. And the biggest problem with the hardware is that the nozzle shoots right into the opposite wall. Right on target! The water just collects into large drops or a stream more likely (water surface tension is a very strong force..)

Richard, you may be able to reduce the drop size a bit, but I don't think that will solve the problem unless you can get an 80% distribution under 30 micron (true fog). I hate to say it, and be a doubting Thomas. I haven't seen anyone do that under 500 psi. I don't really feel that particle size is the biggest problem.

(Thinking out loud) Even fog shot to the opposite wall with any velocity may still get to the wall. And while a larger particle does erode more effectively, this picture (nice pic) is classic leading edge erosion from FOD, caused by large drops or wall stream. It looks just like my lawnmower blade. Some degree of the wear must be considered normal wear and tear. While some impingement damage is visible across the entire leading edge, the majority of the damage is on the outer 10% of the blade, no doubt from something traveling up the wall.

Good ideas Larry, not knocking your step idea larry, I hadn't thought of that. In my mind, the real solution is aligning the nozzle(s) into the airstream, from the CENTER of the conduit, and using a narrow cone pattern (to minimize wall pooling), and then only at the highest practical MAF threshholds. If we can brainstorm to create a dual nozzle setup that will help even more. a smaller capacity nozzle (better atomization) then can be used at the highest practical pressure. Maybe put the nozzles back-to-back.

Larry a leaf blower can be used to show how 2 different nozzle orientations (within a conduit) will have dramatically different pooling characteristics. I will have to do this myself. Just collect the pooled water at the end of the conduit over time and compare.

Michael

[hotrod]

I agree, a co-axial spray setup is one of the things I have been thinking about.

Imagine a carburator venturi type nozzle suspending centrally in the air intake. The high air speeds in the venturi should keep the spray as a relatively compact plume that should travel down the center of the duct, with the mist impacting the compressor impellers near the center where their radial speed is relatively small.

In my next version I will be only using the pre-compressor injection at high engine rpms and boost ( pressure switch and rpm switch in series ). I could do the same with some sort of a maf voltage, trigger as well.

Larry

[PuntoRex]

Punto Rex, How many miles on the turbo? How many miles using preturbo misting? It must be said. Unless you shot a pic of the turbo before you started misting it, you can't make a conclusive statement on the cause of the damage.

Well, fair enough, I totally agree.

Here is my records:

2/28/2002 83125km The first turbo service
12/28/2002 99094km The second turbo service + engine rebuild
11/06/2004 131593km Pre-compressor injection installed
4/2/2005 139671km Damage blades were found

I have a photo of the turbo before the 2nd turbo service (12/28/02). Which is, it's 10-month-old & had serviced for 15969km.

Unfortunately that picture is not so detailed as this one. (it's a hard copy & I don't have a scanner to share it here) However, on that picture, the outer edges of the blades were obviously in a much better shape than this. The tips are mostly in a pretty sharp right angle, instead of rounded out like this current one.

Pre-compressor injection has been deployed for 5 months & approx. 8000km on it. TBH, it's only a short service time.

It's a pitty that I missed the chance to take a picture when the first service to the original stock turbo with 4 and half years and 80 thousands km of service. And I don't have many experiences with old/abused turbos... Maybe someone around here with more experiences could share more info.


Richard, thanks for the compliment. This photo is chosen from a collection of about 10 shots.

[PuntoRex]

This is(was) my pre-compressor jet seat:

http://home.pchome.com.tw/personal/rexchiou/JetonElbow.JPG

The compressor inlet is on the left bottem of the picture. In the engine bay, the whole thing actually turns 90 degree...

As can be seen, the jet is facing a half-open space. Even when no air is flowing, near half of the spray cone should hit the compressor directly, well, sort of. The spary cone is so wide that unavoidably hitting the wall.

Without significantly higher pressure, samller droplets, much narrower spray pattern & precisely injection angle, I think we can hardly get away with the damages. :sad:

As to close co-axial installed jet, in my car, it's almost impossible to deploy. :cry:

[janis]

I tried to make pictures of my 4 year old turbo that has 112.000 Miles on it but I don't seem to be able to make sharp pictures like you PuntoRex.

Anyway, my point is. My compressorwheel doesn't look as damaged as yours, the edges are still sharp, well, not 100%, but better than your wheel.

My pre-turbo-injection plans just went into the fridge...

Sorry to hear it didn't work out in your case!

Janis

[Richard L]

How accessible is your turbo inlet area. Is there a room to accommdate an inline narrow angle nozzle (super-fine droplet) right in front of the turbo blades?

[janis]

How accessible is your turbo inlet area. Is there a room to accommdate an inline narrow angle nozzle (super-fine droplet) right in front of the turbo blades?

The hose to my turbo looks very much like puntorex's. I'd have to mount the nozzle in the bend of the hose.

[Richard L]

hmmm... not so simple.

May be puttng a 1mm ridged sleeve inside the bore of the turbo inlet will do the trick, just try to explore hotrod's idea.

http://www.aquamist.co.uk/forum/sleeve.gif

[PuntoRex]

janis,

I took that picture by a Canon digital camera, with the built-in flash light & close up mode.

With the turbo in car, it's very hard to get a good angle, I just guessed & saw what I got.


Richard,

How narrow spray pattern you can get with a fine enough droplet?
At the elbow near inlet, I think maybe it's more easily to aim the target, like this:
http://home.pchome.com.tw/personal/rexchiou/PreTurboInjection.jpg

Or in a straight pipe, the co-axial nozzle is more difficult to install. However, the air flow would probably tilt the water injection pattern near the turn.

[Richard L]

PuntoRex,

I can try making it as narrow as possible - need to do some experiemnts after designing it.

Cound this sleeve adaptor arrangement work on an existing nozzle?

http://www.aquamist.co.uk/forum/sleeve2.gif

[masterp2]

You guys are really talented. Very impressed with this concert effort.

Another 2 cents? Any mod in the airstream coaxially (off the wall) is going to be hydraulicly restrictive, defeating the improvements sought. Especially anything that reduces the apparent inlet diameter at the turbo inlet mouth (such as an ID sleeve), as this part of the housing is probably finely engineered to work correctly by an aero engineer.. IOW, I believe you may very much disrupt the intended airflow characteristic, angle of incidence, to the blades. These are just my educated guesses. I never designed turbos, so a better opinion is needed. But I do understand aero and hydraulic engineering concepts.

I can't see the benefit of this sleeve, then perhaps I'm not understanding. If the purpose of it, is to form a step that diverts pooled water toward the center of the blade, how can that be done without severely pinching off air flow/reducing inlet ID?

[Richard L]

May be it is just effective to bore the ID by 2mm just short of the turbine tips.

[masterp2]

janis,

http://home.pchome.com.tw/personal/rexchiou/PreTurboInjection.jpg

Or in a straight pipe, the co-axial nozzle is more difficult to install. However, the air flow would probably tilt the water injection pattern near the turn.

I like! Concepts to keep in mind. The heavier the drop, the harder it is to change its direction, or put another way when considering spray before a bend: a more coarse spray will have more collisions at the bend (collecting a larger stream). Hope that makes sense, it is a key concept.

The finest spray possible, in a conduit with the least change in direction, when concentrically implimented, using the largest diameter conduit, smallest practical cone pattern, at the highest possible conduit airspeed, will produce the least pooling. Nonetheless errosion WILL occur if a significant fraction is over about 40 micron, due to impingement. I do believe that the above outlines how to drastically (90%) minimize the occurence.

Another thing to consider. If a turbo ranges 0-15 lbs of boost, what good does it do, from a performance perspective, to spray it at 10 psi? Answer: none

You don't get the maximum vehicle performance until the foot is on the floor, 15 psi. To enhance top end, spray at 13 or 14. That change alone just reduced exposure and water damage 50% maybe (even using the wall nozzle). Moderation, properly implimented, is key. You guys with photographic baselines, probably are going to be the likely candidates for testing something here, not much to lose at this point. Please consider it.

[masterp2]

May be it is just effective to bore the ID by 2mm just short of the turbine tips.

Here again, I will make my opinion, but in no way should you stop the thought process. This is a wonderful collaberation.

I am fairly certain that the turbo compressor is designed around the concept of "specified inlet" flow. (if someone can say I am wrong then the following is meaningless)

A notch at the turbo inlet could be turbo suicide, if it induces stall. If it does, you will know right away, water or no water. On the other hand, maybe you stumbled onto a very rich patentable idea, like the golfball dimples.

Edited: for clarity

[hotrod]

I agree with most of the preceeding statements. My intention was to have the drip step well back from the turbo inlet. I would be very reluctant to place it right up against the compressor impeller as it could create all sorts of strange effects on the compressor map, moving the surge line around and so on.

This is the mental image I formed. The injection nozzle is placed .5 -.8 meters back from the compressor inlet. Some reasonable distance down stream from that, you flare out the inlet pipe some. This slows airflow speed and encourages any large dropplets to drop out of the flow. After a short distance you began to taper back in, creating a venturi. Liquid water would tend to collect at the point the intake tract begans to converge again. The liquid would be forced to move "uphill" on the converging section. If air speed is low enough at that point it would be impossible for the liquid dropplets to run on the surface. If you place a small lip near the end of this converging section, you would stop any dropplets that try to climb the "ramp" and as they get pushed up over the lip they should get ripped into a fine mist again by the high speed airflow at the throat of the converging section.

A secondary possibility would be to place a small drain at the bottom of the wide point to allow any liquid water to drain away. This may require some form of active extraction, or simply a gravity drain with a vacuum check valve would suffice. For active extraction you could just plumb a simple windshield washer pump with its suction to this low point and its outlet back to the water reservoir or suction side of the WI pump.

For reference, Oakos Automotive on their web site ( http://www.oakos.com/wrx/installs/AEMCAI/index.htm ) did a measured air pressure drop test on a WRX. The stock intake created a 26" of water pressure drop at max flow, the high flow AEM intake reduced that to about 12" of water pressure drop. Any drain system would need to have a check valve so it did not create a vacuum leak post MAF due to the intake pressure drop but allowed the collected liquid to drain away.

My current setup (old version) had the spray injected from the outside of a horizontal 90 degree elbow, with the spray at right angles to the air flow. When I pulled the turbo out that I photographed earlier, I could see clear evidence of liquid water flow on the inside of the intake tract. There were slight but unmistakable markings left by the evaporating water as it ran on the surface. The water mark basically went downward from the jet turning down stream and then followed the bottom center of the intake tract.

In considering the physics of what was happening I remembered some old info I had picked up years ago about water flow in a stream as it enters a turn. The fast moving flow moves to the outside bank and then due to drag with the outside river bank the current turns and moves down toward the river bottom then comes back upward on the inside of the turn toward the surface. In the case of air flow in a duct you would have this same hydraulic flow in the full 3 dimensions of the tube, with the air scrubbing hard against the outside center of the duct at the point of the turn then splitting and 1/2 going up and 1/2 going down (on a horizontal bend) and curling back toward the inside of the bend a few diameters down stream. This sort of flow would carry most of the water dropplets down the bottom center of the duct. In the faint water marks on the inside of my duct I could see exactly this same type of movement. I suspect a large fraction of the liquid drop out occured during the turn on and turn off period of the spray event when atomization is not good, not to mention some after drip that I imagine every nozzle would have.

The most simple solution if you have the room would be to create a P trap intake like on a sink drain where the injection occurs on the downward section of the P and then any liquid water would collect at the bottom as it could not climb the vertical leg of the trap. Air flow would likely quickly evaporate any water collected but it would be a bad thing if a large quantity of liquid were to collect there allowing a water slug to be pulled up the intake during a hard "yump" or other violent change in direction.

I also agree that Pre-compressor injection is best reserved for use only during high airflow situations where the benefits it provides would be best used. My 10 psi turn on was too low and as mentioned above I will move that turn on point up toward peak boost where I need to maximise the compressor effeciency.

FWIW

Larry

[Greenv8s]

Your diagrams still seem to show the water spraying onto the whole face of the turbine. I assume that the water erosion could be avoided by ensuring that the water didn't impinge on the tips of the blades, could this be achieved by having a narrow jet on the center line very close to the turbine? I mean within an inch or two. I'm envisaging something like a pitot tube inserted into the air intake close to the turbine.

On another tack, to stop droplets formed on the walls from hitting the tips of the blades it may be possible to pinch an idea from a Porsche crank scraper. Imagine the air intake as a vertical tube with the turbine at the bottom. Step the wall in just above the turbine, and shape the wall so it forms a gutter. Angle the gutter so it forms a spiral rather than a circle. At the 'lowest' point, put a rib taking the gutter out to the middle of the turbine, still with the gutter profile so that water has to run out to the end of the gutter and can't just drip off the sides. Assuming the air speed is high I think aerodynamic forces would be quite high compared to gravity so the water might be pursuaded to go 'uphill' towards the axis of the turbine (if you see what I mean). From point of view of air flow it would look like a straightening vane with relative little drag.

Just a thought, what do you make of it?

Oh, by the way, can anyone put any numbers to the turbine tip speed and the water droplet speed at the jet? I'm wondering whether there's any mileage in tangential injection ...

[masterp2]

http://home.pchome.com.tw/personal/rexchiou/DamagedCompressorWheel.jpg

On second look, I don't think I buy this being water damage. What is your filtration? Did you run w/o a filter ever? Lose a nozzle screen ever?

i am no expert, but this looks like grit damage to me. Sorry, just that I have to question if we are chasing our tail.

Larry, you have the same turbo, yes? You had nothing like this, correct?

[masterp2]

greenv8s

I am not able to envision what you have described.

But, for it's simplicity, I like the center-mast stream, or narrow cone (preferred) idea. But again, anything located close to the blades will be an airflow disruption, and I think a good rule is to keep at least 2 diameters away from it. Then pray your install never get's loose! For many, there is a bend in the intake close to the blades (considered bad design) and that may limit how close you can get anyway.

If only someone can come up with an easily serviced coaxial alignment. The wall nozzle has that advantage. Someone can do it, to make this idea practical.

[PuntoRex]

...What is your filtration? Did you run w/o a filter ever? Lose a nozzle screen ever?
....



I've been using K&N cone filter for years, and never run without it.

I do hope, too, this is not caused by the injection, but now I hesitate.

And I remember Larry had a similar picture here somewhere. The damage was not so severe, though.



Michael,

Why don't you try it yourself & see how it goes? :twisted:
Maybe you'll have better luck :wink:

[hotrod]

And I remember Larry had a similar picture here somewhere. The damage was not so severe, though.

My compressor erosion pictures are on page three of this thread. My damage is less severe and I KNOW that some of my damage was due to dirt ingestion during a couple unfiltered drag strip runs.

Your diagrams still seem to show the water spraying onto the whole face of the turbine. I assume that the water erosion could be avoided by ensuring that the water didn't impinge on the tips of the blades, could this be achieved by having a narrow jet on the center line very close to the turbine? I mean within an inch or two. I'm envisaging something like a pitot tube inserted into the air intake close to the turbine.

Yes it is likely it can, you just discribed the setup used by one guy that has been running pre-turbo spray for years. He injects directly on the compressor shaft nut with a narrow jet. The rapidly spinning shaft nut simply blasts the water stream into a fine cloud. I suspect the effect is not dissimilar to a 1000 psi impingment fog nozzle. The advantage of that system is as you say little or no chance to wet the intake tract walls, and the water is ejected tangientially outward inline with the compressor blades so little relative speed difference should exist. This is also very similar to the configuration used in WWII combat aircraft where they injected both fuel and WI "into the eye of the compressor".

The problem with that setup is it is not easily setup on many systems as the available space near the compressor inlet is quite limited in many cars, and service (ie to change jet size) would be quite difficult.

One of the things that may have limited my compressor damage is that I set the system up with a constantly pressurized system and the flow was controlled with a solenoid only about 4 inchs from the jet, so after drip would have been essentially eliminated, and rise time to full spray pressure at the jet should have been very short. Mine was also a continuous spray rather than a modulated spray as you would get with the HSV turning the spray on and off for folks using that style of system.

I also had a manual test spray button that at times I would briefly test the spray at low rpm to confirm my system was working. Based on this discussion, I now consider that to be a bad practice and will avoid it in the future, unless the engine is at a fairly high air demand, so that intake velocity would be likely to keep the plume from heavily wetting the intake walls.

Larry

[Richard L]

Would this method work?

http://www.aquamist.co.uk/forum/sleeve3.gif

[masterp2]

Nice Richard,

Would it "splice" into the conduit? Can the plumbing be routed within the support structure?

[Richard L]

Nice Richard,

Would it "splice" into the conduit? Can the plumbing be routed within the support structure?

Haven't thought that far yet. It wil be quiet a simple fit, just slice it between the inlet hose and turbo flange, possibly with a few slots on the ring so it can be tightened up.

[masterp2]

Do you think it could be used on different size ducting?

For universality (all diameters) and ease of maintenance:

how about a device that can be inserted in a standard size hole drilled into the conduit? Insert, and then have a band clamp of sorts on the opposite side that would pull it tight around the periphery, for a seal. Would also minimize drag interference I think.

[PuntoRex]

..... He injects directly on the compressor shaft nut with a narrow jet. The rapidly spinning shaft nut simply blasts the water stream into a fine cloud. .....


Larry,

Is it possible to have a picture about how he install the whole thing?
I guess it'd be pretty similar to Richard's proposal above.

But, how can we deal with the restriction to the airflow & the fixing reliability?

Yet another idea, the supporting spoke can be designed as a set of "wings" to be a fixed turbine. Can this be done, or has any advantages? (sorry for the naive imagination on the aerodynamic)

[hotrod]

Is it possible to have a picture about how he install the whole thing?
I guess it'd be pretty similar to Richard's proposal above.


I've never seen the installation but it is explained in some detail in another forum : (posts are about 2 years old)

http://www.eng-tips.com/viewthread.cfm?qid=72284

Look for the posts made by " turbododge " --- I haven't figured out a way to invite this guy over here yet, as I don't see any sort of a private messaging system on the eng-tips forum --- maybe I'm missing something, but I would like to see him join this thread if he happens to browse this forum under a different username.


================== quick summary of his key posts ======

turbododge (Automotive)
30 Oct 03 19:51
I just got referred to this thread by a friend and being a long time user of water injection thought you all may be interested in what I have personally seen over the years. I run a twin turbo, now EFI (was carb), twin turbo 340 Mopar in a 70 Cahllenger, strictly street, 14 psi boost, intercooled.

I use an old Edelbrock Varijection system that has been altered for instant on under boost conditions at 7 psi. It varies the amount of water based on rpm and absolute manifold pressure. It does not have adequate pressure to spray directly into a pressurized intake manifold, so I spray directly on the turbine blades. The turbos are lower than the inlet ducting to the throttle body, so even if I have a valve failure, I cannot water slug the engine. I have run this system for over 15 years on carbed and EFI turbo engines, and would not go without water injection. I can run 8 to 1 compression at 14 psi on 92 octane all day without worrying about detonation.

I find the talk about atomization, hummidty, where the water evaporates, etc to be very interesting, as I have watched my system extensively over the years. When I spray on the turbine blades, the water (with 50% isopropyl) vaporizes instantly, before it even touches the blades. I have 50K mile turbos that show no erosion at all, and even the intercooler does not get any significant amount of moisture in it. The ductwork all stays dry. The low pressure area at the turbine inlet easliy evaporates the water spray. I found no difference with EFI or blow thru carb in this respect.

When I was running a suck through carb setup, I was spraying directly into the carb throat with the water, and found that the carb venturi would vaporize the water right along with the fuel from the venturi. Subsequent running through the turbo also further atomized both elements. All was good unless the inlet ducting was cold enough to condense out the fuel and slug the engine.

In an N/A engine, you would spray straight into the carb with most systems like a Spearco. An Aquamist you could go into the manifold below the carb as they atomize better, but most folks don't do that. Unless you are dumping very large amounts of water into the carb, it will atomize, if it did not, it would build up in the intake and give a big slug to the engine when you turned, etc, and probably thermal shock or hydraulic the engine to distruction. I think it would be a very risky business to try to get, or count on getting unvaporized water into the cylinder.

As far as the power is concerned. As I mentioned before, my setup is instant on, off a pressure switch, or an override switch in the cockpit. I have done lots of testing at boost levels to 12 psi (max without detonation without water) with the water on and off. I give people rides and turn the water on and off, then off and on to see if they can feel the accelleration rate change. I can tell you that no one has ever not been able to tell that the water came on and the accelleration rate increased. It is very significant. I would be surprised, however, if you would be able to readily feel the difference in an N/A engine, as when I turn the water on and off while holding the boost down with the throttle, I cannot feel a difference at all, in my low compression engine.

My net conclusion is that on a boosted car above 10 psi, water injection is necessary if you want to be able to run decent compression, timing, and mixtures, plus you will get more power do to charge cooling, even if you are intercooled. On an N/A car, I would use water only if you need it to kill detonation because of high compression, bad gas, etc. I don't believe it would give you enough more power to notice.

==========

turbododge (Automotive)
31 Oct 03 0:18
Turboice: You are absolutely correct in that one big factor with water injection is killing the detonation to allow you optimize other conditions. It is very common in the turbo crowd to try to drown the detonation with extra fuel, even down to 10 to 1 A/F. All it does is cost power and give minimal results. The is a very good chart in the Hugh MacInnis book Turbochargers on the affects of water on detonation, boost level and air fuel ratio. One look at the chart and you can become a water booster for life.

Concerning the non erosion of my blades, it may have to do with the fact that my water only comes on when I really need it, so it is on a very small % of the time, but I am also very carefull to send the nozzle stream (it is not atomized) at the CENTER of the compressor shaft so that any impact is on a slower speed, less critical area, and the water flow moves out the blade rather than impacting it. Also, with Varijection, there is less water at lower rpms and boost levels and more at higher speeds and boost. A single output volume, set to deliver enough water at full load, like a Spearco, can easily overwhelm a turbo spinning at lower rpm and impact the blades harder. With the EFI controlling spark and A/F (12.5 A/F under boost)and careful tuning of the water we normally will only use a gallon of water per 1 to 2K miles of street driving, as we are good to 10 psi very safely without any water, (and 12 if the gas is good). So putting is enough water to get to 14 psi is pretty easy.

==============
turbododge (Automotive)
31 Oct 03 12:40

...

The erosion of the blades discussion is very interesting, and may apply to others differently than me. I am running plain old T04, oil cooled turbos, and have been for many years. The newer turbos with the aforementioned lighter blades and perhaps more brittle alloys may be more of a problem. With those I have no experience. There was a post that mentioned someone who had a friend who did get blade erosion (post probably gotted whacked because it referenced another site). Was this person, by any chance, running a water/methanol mix in his system? From what I have seen and heard, methanol will take out turbine blades, throttle body plates and bearings, and can even corrode the tips of fuel injectors enough to cause pattern problems. That is why I use Isopropyl in mine.

============
In thread http://www.eng-tips.com/viewthread.cfm?qid=82878


turbododge (Automotive)
5 Jan 04 18:51
I have been injecting water/isopropyl alcohol on the turbine blades of my twin turbo 340 Mopar for 15 years now. This kind of system was very popular in the past, as it was much more difficult then to inject into the boost stream because of the high pressure. The blade erosion issue has always come up in discussion, but the only ones I have seen that had problems were either on stationary units that ran the injection constantly, instead of only under boost like in a street car, or if they used methanol in the injected liquid. Methanol is much more corrosive than isopropyl. My turbos have over 40K miles on them with no signs of erosion at all. I use about 1 gallon of water/isopropyl mix per 1000 miles. My system is also variable with speed and boost, so I don't overload the airstream with too much liquid at any time. It also atomizes better if you aim a single spray nozzle directly at the shaft of the turbine, the liquid usually is totally vaporized by the time it is 1/4" from the nut on my setup.

I can run with or without the water very easily, and if I am holding a boost level and turn on the water, you feel a very positive increase in accelleration. I am also running an intercooler, so this is beyond the intercooling affects.

If you do a good job of setting up the system, I would not be afraid to inject ahead of the turbo, as I have seen very good results that way.


============
turbododge (Automotive)
7 Jan 04 0:01
Yes definitely on the compressor wheel, poor phraseology on my part.

I am not up on the viscosity/vapor pressure etc with methanol vs isopropyl, but I know I have seen methanol systems corrode the compressor wheel, but never an isopropyl setup do the same. Downstream corrosion is even more of a problem with with methanol, and I have even seen corroded throttle shafts stick in the carb base. Intercoolers will also sometimes condense out a small amount of water or alcohol in cold weather and the methanol will also go work on the intercooler. I would guess that the compressor gets wet enough with the methanol someplace along the line, either at cold start, cold idle or such. The corrosion I have seen covered more of the depth of the blade than impingement damage which tended to be very close to the entrance edges of the blades.

The other thing that goes with this is that all the other systems that had problems with corrosion also were on/off, non variable, systems, which can heavily overload the compressor at low rpms, cool air conditions. I have literally seen liquid drip from the compressor housing hose connections after a run.

I am sure there are lots of folks that have successfully run methanol in their setups, but for me the isopropyl makes it much easier and more reliable, with very little downside.

===============

turbododge (Automotive)
7 Jan 04 20:36
Overrange: We drive the car a lot, but spend a very, very small percent of the time at throttle positions that would turn on the water. If we do a 500 mile "cruise" through the countryside, it is very possible that we will not use any water at all. My system does not come on until 8 psi of boost, (out of 14psi maximum),and since the system matches water flow to boost and rpm, water flow at that point is quite low. I can tell you that a 340 CID V8 will accellerate very quickly at 7 psi, and it is easy to stay within that range by controlling the throttle. To use the full 14 psi on the street requires a very good road in a remote area as the tires can go loose at any time, in any of the first 3 of 5 gears, depending on the traction. Even if the water does come on, it is only on for 10 or so seconds per use (maximum) before you are way over 100 mph. If I am testing and tuning, I have used a much as a quart in a day.

Although I did not mention it earlier, as the question I was addressing was the feasibility of injecting into the compressor, is that my system is not primarily used for inlet charge cooling, as I have an intercooler for that. My system is to allow me to run decent boost, without detonating, on 92 octane pump gas. Without the water I will get detonation at 10 to 12 psi, depending on the gas and weather conditions. It takes much less water to reduce detonation (as long as you have an intercooler to cool the air) than it does to cool the air enough to make power. That said, if I run up to 10 psi with the water turned off, and then turn it on, you can feel the car pull harder, so I am either gaining from cooling of the air further, or because of a better combustion process in the cyliders. I don't know what it is, but you sure can tell if the water is on or off.

The car also has a very good cold air system on it that takes air in through the boxed wheelwells that a fed from spoiler ducts, with only 6" of ducting before the turbos, after the large K & N filters.

===============

As you can see by the above he sprays a solid stream directly on the compressor shaft nut, and allows the spinning compressor nut to beat the stream into a microfine mist moving radially outward. His system is in my estimation an ideal case of wet compression as there is no time for the water mist to evaporatively cool the air stream before it arrives at the compressor, but it will modify the compression due to the high presence of water in the air flow.

I on the other hand needed to maximize the density increase due to air charge cooling pre-compressor to get the most out of my undersized turbo.


I might add that I am not too concerned with his observations about methanol corrosion as he was obviously talking about setups that grossly over injected fluid and kept the intake tract wet all the time. Any system like mine that only comes on intermittently at high air flow demands will quickly dry out any condensate in the intercooler etc. That is one of the reasons I turn my system off in cold weather. I don't need the detonation suppression when the weather is cold ( I don't run high boost when its snowy and icy out ).

Perhaps a little of both would be best one very small spray jet well back from the compressor with some sort of a dropplet trap to keep any but smallest dropplets from passing into the compressor, and a hub injection setup to get the advantages of wet compression.

The other obvious question is are some of us willing to consider the turbo compressor a wear part and to plan on replacing it on a regular basis? I know many of the hard core performance nuts upgrade turbos on a very frequent basis, and for them replacing a $40 compressor wheel (or what ever they cost) every 2 years would not be a big deal in exchange for the performance advantages. I'm sure the WRC folks or people that compete in similar levels of competition would have no problems at all with that sort of replacement cycle. Serious drag racers might go through several clutches, a couple rear ends and a motor or two every year. A little R&R on the turbo compressor is not that big a deal in the grand scheme of things.



Larry

[masterp2]

[quote]

Perhaps a little of both would be best one very small spray jet well back from the compressor with some sort of a dropplet trap to keep any but smallest dropplets from passing into the compressor, and a hub injection setup to get the advantages of wet compression.


Larry

I agree, why don't you get your density contribution post IC instead of pre-turbo, the convention? I am planning a 2 stager. Variable post IC, plus the boost activated 2nd stage pre-turbo (for when you just can't have enough!)

I just took a look at trying to get close to the diesel turbo, not going to be easy. There is a 90 only 5 inches from the blades. The upside to not requiring fine atomization, is that the nozzle doesn't have to be finely engineered. Mybe that will prevent some clogs???

[hotrod]

I agree, why don't you get your density contribution post IC instead of pre-turbo, the convention?

There are two reasons!

1.) In the first case when I started experimenting with this, I was trying to maximize the performance of a stock undersized turbo just to see what was possible. The only way to increase mass flow in a system in choke flow, is to increase the upstream density. It will still be choked to the local sonic air speed at the critical flow point, but with the increase in density, mass flow will still go up slightly.

2.) Due to my living at 5800+ ft altitude (and can easily reach over 12,000 ft altitude in an hour or so of driving) every turbo that is currently on the market is working far outside its design compressor map at these altitudes. At max boost, I am not on the far right side of the compressor map, I am completely off the page to the right. The intake cooling and wet compression pushes my operating point back toward the left.

Larry

[Greenv8s]

At max boost, I am not on the far right side of the compressor map, I am completely off the page to the right. :lol:

[masterp2]

Larry do you feel you have been successful in humidifying the slug of air in the 50 milliseconds it spends between filter and turbo inlet? Can you support that?

just my opinion that with these coarse sprays, and so little residence time, that nearly no evaporation actually occurs, especially considering the low water/alc temps used. Maybe you can change my mind.

Flash swirl sounds like what you need to explore. What do you think? swamp cooler?

[hotrod]

Larry do you feel you have been successful in humidifying the slug of air in the 50 milliseconds it spends between filter and turbo inlet? Can you support that?

just my opinion that with these coarse sprays, and so little residence time, that nearly no evaporation actually occurs, especially considering the low water/alc temps used. Maybe you can change my mind.

Interesting question --- My short answer is enough to make a useful difference.

The objective is NOT to humidify the air, the objective is to increase engine (and in this case turbocharger performance) due to the sum of multiple effects. One of which happens to be evaporative cooling.

After considering you question for a while here's my view.

First very very small amounts of evaporation are all that are necessary to make a major change in the inlet conditions of the turbocharger. The specific heat of the intake air is .25 Cal/gm deg C (give or take a bit) while the latent heat of evaporation of water is 540 Cal / gm.

The maximum intake air flow on the stock WRX turbocharger is approximately 360 CFM or about 25.2 lbs / min ( 0.42) lb/sec or about 191 grams /second.

If you are spraying at a rate of 4 gallons / hour that is approximately 4.2 grams/second of water.

If you lower the intake air by 11 degrees Farenheight or about 6 deg Centigrade you increase the air density by 1% which will give you an increased mass flow through the turbocharger of approximately the same and give a net power increase (all else being about equal) of 1%. On a 250 hp engine then, each 6 deg C drop in intake temp is worth about 2.5 hp.

Given the ratio between the specific heat of air and the latent heat of water you have to evaporte only a very small fraction of the water to bring down the intake air temp significantly. Lots of people have recorded intake air temp drops on the order of 30 - 50 deg F (17 - 28 deg C) with initial air temps near 105 deg F (40 deg C).


So just to persue your question lets look at what we know. The air flows at a rate of 360 CFM max in an intake tube of 2.5 inch ID --- this gives a peak air speed at redline of about 176 ft/second (120 mph). The spray nozzle is about 2 ft away upstream so flight time for the dropplets are on the order of 11.4 milliseconds.

Now how many drops are in that air column and what is their surface area?

One cubic centimeter of water in the form of 50 micron dropplets (if my numbers are right) consists of about 424 million dropplets with a total surface area of about 333.3 cm^2. We are spraying at a rate of 4 gal/hr or 4.2 cm^3 / second so we are putting about 1,780 million dropplets into the airstream per second, they have an effective lifetime of 11.4 milliseconds in the air column, so the air column at any moment contains about 0.0477 cm^3 of spray or 20.44 million dropplets with a total surface area of 15.9 cm^2. Plus we also have the evaporation that occurs off the wetted interior surface of the tube, which would amount to 1216 cm^2 of surface.

How much water evaporates every 11.4 milliseconds, from 1232 cm^2 of water surface when exposed to a 120 mph wind ??

That intake air column has an air volume of only 0.068 cubic ft, which converts to about 2.166 grams of air, it obviously only takes a miniscule amount of evaporation during that 11.4 millisecond flight time to make a huge change in the temperature of a slug of air that only weighs 2.166 grams.


To lower that 2.166 grams of air 6 deg C, you only require .54166 calories of energy per deg C. To get that amount of cooling by evaporation of water you only need to evaporate 0.001 gm of water in 11.4 milliseconds/ deg C, or 0.08799 grams per second/ deg C which equals .5279 grams of water evaporated per second for a 6 deg C intake air temp drop.

{ all this ignoring the higher volatility of the alcohol component and its lower latent heat of evaporation but --- lets keep it simple for discucussion sake}


The interesting thing about this exercise is that it is obvious that the wetted surface of the intake tract dominates the evaporative cooling process by a large margin over the cooling that takes place at the dropplet surfaces.

Your swap cooler comment is right on target, and I have been considering similar ideas. It might be useful to intentionally maximize the wetted intake surface both to gain this cooling and to capture large mist dropplets.

I have considered putting pieces of fine mesh brass or stainless steel screen wire placed at extremely oblique angles to the air flow. (intention to give a very large open surface area but little or no direct path through the screen) Any dropplets that couldn't make the "jog" due to their large size, as they approached a screen wire would be broken up and wet the screen wire surface. Smaller dropplets would simply pass through the screen unmolested.

In a perfect design the screen would be treated so it wetted well with water alcohol mixes and had high capillary attraction to pull liquid water up off the floor of the intake and into the air stream. This is basically what is done inside heatpipes.

Larry

[masterp2]

I started down the road to verify your numbers and it suddenly occured to me... I'll get to that. First, nice quantifiable arguments, as always. What I say in no way detracts from your well explained facts, but something is left out. Nowhere really do you represent where there is a reason for the liquid to evaporate. There is no "driving force" equation representing the change of state. You could have a billion droplets at -500 C and all the exposure in the world will still require an ice-age of time to evaporate (sublime) them.

Yes we have heard of 30-60 F drops, but not with ambient air, heated air (that is also dry). Ambient air at 80F will evaporate water at 80F, and if the air is 60% RH, it may take minutes, not milliseconds, to change state of a fraction of it. Perry's or the CRC might be worth looking at. See in an environment where compressed heated air (post-IC) is plentiful at 5% RH, there is a much greater force evaporating the liquid. But what you said, about the wall exposure was brilliant, and maybe breakthrough.

Here in Sunny Phoenix, I can lose 1/2" of water out of my 90 degree pool on a windy 115 F DAY. Yet not notice any loss, after a week of calm winter weather, with 50 degree air and water temps. Naturally if the water was at boiling, air at 100C, I could evaporate the whole pool in hours.

What quickly became obvious to me is this: a lightweight droplet, entrained in a moving column of moist cool air, is going nowhere (except through the compressor). Ah hah! The moisture on the wall has nowhere to go but to humidify. (the assumption that the entire conduit wall is wet is optomistic without a system settup to make it wet, one nozzle isn't up to the task, agree?) Without some nice dry hot air blowing past the drop, it quickly retires to a fate of drift, until it is exposed the hot, dry, dynamic conditions that are waiting downstream.

Keeping clear that anything that increases the apparent restriction in the plumbing is not going to work. It WILL kill the performance sought in the form of pressure drop. Screens and such in the column will not be a solution, without drasticly changing the conduit dimensions. 176 ft/s is not the place to add any kind of change, since as we know, most hood ornaments will be cast off at that speed, certainly flies don't survive it.

FWIW, I grant some evaporation can occur post filter, but this must be so miniscule as to be negligible. Swamp cooling that much air, done in a conventional way requires square meters of wicking material, and a puller fan that will wreak hovoc on your alternator. The pressure drop across swamp cooler pads is significant, and the process is not an on-off process. It takes minutes to wet the pads befor they start providing the desired result, etc.

What does come to mind, is along what he have discussed, with a twist. A 1/8" perforated SS tube that runs the much of the length of the conduit, concentrically. Ever seen a soaker hose? each .008" perforation a pseudo-nozzle, maybe 10-20 of them depending on flow requirements. You could even put a nozzle on the end of it, pointed into the blades.

What it might do for you is keep the wall lightly wetted, and still provide for airborne mist, and a direct inject nozzle. Manufacturing it, without significant restriction, will have to be a different book.

[hotrod]

Well your comments are interesting, but I feel there is plenty of driving force.

Typical conditions here at the strip or on a hot summer day, outside air,measured well away from the pavement 85 deg F, at 12% humidity. I frequently measure actual intake air temps above 105 deg F.

That puts the intake air humidity as it passes the spray nozzle in the dry air catagory with a dew point of less than -39 deg F.

At dew points like that with pure water your wetbulb temperture would be about 53 deg F. (ie this is the temperature the air would cool to if you raised it to 100% RH by evaporation of water)

This is ignoring the issue of alcohol evaporation or the lower air pressure which makes the water have a higher relative vapor pressure.

The relative air speed between the dropplets and the intake air being in the 100 mph plus range will result in a nearly instantanous evaporation of enough water to drop the air temp by 10's of degrees F.

Its difficult to measure the true air temp because of problems caused by wetting of the surface of the temperature probe, but there is absolutely no doubt that there is a very large driving force for evaporation when the dry air and the rapidly depressurized water dropplets first encounter each other at high relative velocities.


If you want to dig through Perry's or CRC handbooks to figure out the realtive vapor pressures and all that ---- please do ---- I have neither the time or the interest in trying to validate what is obvious to me based on my experience. When going to the pre-compressor water spray, my midrange boost onset was so dramatically improved I had to change my driving style to keep from running over people when accelerating from stop lights. I'd come into boost so quickly I had to change my boost controller settings (turn them down) for daily driving.

Simple proof of driving force for water evaporation when your car is mass flow limited for air by the choke flow capacity of the turbo.

Without water spray in the intake car goes x mph in 1/4 mile, with water spray pre-compressor car goes y mph in 1/4 mile.

x is > than y ----- spray must work!


Larry

[masterp2]

OK, I buy it. What do you think of the spray bar/tube idea?

[hotrod]

I'm not fond of the soaker hose idea, --- seems to be too much complication, and an absolute gurantee that you would end up with liquid water running down the bottom center of the duct. By the time you get that long tube supported so it does not fatigue fracture and go through the turbo I think your drag losses in the intake would go sky high.

At this point I think some controlled tests on a test rig would give us more useful info in 2 days than we could glean from months of brain storming.

I just need some time to cobble together something that will allow some pictures and measurements to be made.

New job it keeping me busy. ;)

Larry

[masterp2]

Oh well, a revelation for me alone then, I guess.

Hope I was clear. Each pseudo nozzle in the SS line would be much smaller flow than any w/i nozzle presently marketed. Large droplets would wet the wall, small mass droplets entrain in the stream. It has the advantage of saturating using all means, given the short residence time before turbo.

Not sure why a SS line would be in danger of coming loose. Of course, this is unconventional, out of the box thinking. That is, no doubt where the solution is for such a challenge.

Perhaps you have seen the loops that are fastened to high speed fans at football games, for quick cooling on the sidelines. I'll find one and post it.

[masterp2]

http://www.koolfog.com/images/home_middle_fan.gif

[Roverdose]

wow interesting read. some thoughts i had that im not sure were covered though..

i think there is always a step from turbo inlet pipe to compressor housing (unless the rubber pipe flares out just at the compressor inlet).

if im running one pump and 2 hsv, what happens with the pressure difference at the nozzels?
say i start injecting the main jet (after the intercooler) at 18psi and the pre-compressor jet at 23psi. engine running a max of 28psi.

would the water be more likely to travel to the compressor jet as there may be a vacuum in front of that one and +23psi in front of the main jet?

Drew

[masterp2]

Any progress Larry? Or anyone else?

[dsmtuned]

Larry I was wondering what size nozzle you are running pre-compressor. If I remember correctly you have a Shurflo pump, correct?

Do you have post-turbo injection as well?

Drew I have also thought about what will happen when I run a dual stage set-up (1 pre, 1 post compressor) when I change to a pre-compressor injection nozzle. I wonder if the Shurflo pump can maintain pressure so the nozzles continue to atomize correctly. I know lowering the pressure of the system by adding a second nozzle decreases the amount that the nozzles each flow, but what is the effect on atomization?

I currently run a dual-stage post-turbo WI setup with no problems. I just worry about changes to the atomization when I move one of the nozzles to spray in front of my soon to be new turbo :? .

I have a turbo with turbine wheel damage from a manifold cracking and throwing chunks downstream, so I am willing to sacrifice it in the name of science :wink: . The only problem is that I am continuing my year long travels in SE Asia and will not return to Colorado until October 2005, so we will have to wait until then.

Craig (currently in Cambodia)

[Roverdose]

i was thinking of maybe adding one of those inline check valves in the pre-comp line.

something along the lines of if i got 15+ psi after the intercooler, i could try and balance the flow between the 2 jets with a 15psi valve at the compressor.

although just read aquamist's website..
http://www.aquamist.co.uk/sl/plist/pic2/806-248/806-248.html

Recently it has been used for twin jet application where water is again being siphoned between the jets due to pressure in-balance and causes unnecessary delay during injection. Twin-jet application requires two check valves and must be installed immediately after the jets.

maybe thats what your supposed to do anyway.

Drew

[masterp2]

I would like to see a 70-90 psi crack pressure in a check valve. That would help prevent inneficient atomization for the critical pre-comp stage.

...or a solenoid run by a presssure switch on the water line, etc...

[hotrod]

Larry I was wondering what size nozzle you are running pre-compressor. If I remember correctly you have a Shurflo pump, correct?

Do you have post-turbo injection as well?

Well at the moment I'm getting ready to re-configure things. My old setup was pre-compressor only and a 3 to a 4 gal/min nozzle. That was too much water in retrospect so I decided to rebuild the system.

I will be adding pre-throttle body nozzles and reducing the pre-compressor nozzle size probably to a 2 gal/hr or a 1 gal/hr rate.

My system uses a shurflo 100 psi pump and a shurflo 24 fluid ounce accumulator (pn 181-203) and the trigger solenoid sits within inches of the nozzle, so the turn on time is very short, with a large accumulator at 80-90 psi feeding the solenoid.


Larry

[dsmtuned]

I have been thinking about how I can get rid of pooled water in the intake pipe during pre-compressor injection. One idea I thought of is creating a place for the water to pool, like bottle cap sized addition to the intake pipe that is lower than the compressor intake. Here you could place a small check valve that would close while you are WOT, but open when you let off the gas, allowing some of the pooled water to drain out.

I don't know what kind of check valve would open so easily that water could drain out, but I'm up for ideas. I also don't know for sure that there would be adaquate time between shifts and low enough suction in the intake pipe to allow the valve to open.

Here is my other idea.... You could use the vaccum from your throttle body to suck out the water from the pooling area (for lack of a better term) into a small catch can or something.

The only thing I worry about with the pooling area is that huge drops of water would be sucked out of it because of the high speed of the air in the intake tube. I think the depth of the pooling area would be the only way to reduce this. I also don't know if the pooling area would create a bunch of turbulence directly in front of the compressor. I guess it depends on its design.

I would appreciate any feedback that people have to offer.

-Craig

[masterp2]

If you can design something that will evacuate excess, and not reduce pressure in the inlet tract, that would be interesting. Anything that sucks out water, is going to cause a performance degrading pressure drop I think. Also, can't have unfiltered air infiltration when no water is present.

Sorry i am not more help.

[Roverdose]

just inject into the centre of the turbo.

Drew

[dsmtuned]

If you can design something that will evacuate excess, and not reduce pressure in the inlet tract, that would be interesting. Anything that sucks out water, is going to cause a performance degrading pressure drop I think. Also, can't have unfiltered air infiltration when no water is present.

Sorry i am not more help.

I was thinking that the suction would only happen when you let off the gas, as to not greatly affect the MAS "counted" air mass during WOT.


-Craig

[skiericski]

WOW! I made it through all 13 pages! Great forum here!

Now this forum doesn't apply to me very directly, since my compressor is oversized for the boost/flow I'm currently running. However, my engine setup would be perfect for pre-comp injection, hardware-wise. I currently run a very large K&N cone filter(~6" diameter and 9-10" long) attached directly to the turbo inlet(4")

I got to thinking of all the ideas I've read here. If I had an injection nozzle in the end of the filter(pointing directly at the impeller) then any spray hitting the walls would be hitting the filter element(read swamp cooler filter) and since the filter has a large step where it necks down to the turbo it'd be about impossible for non-atomized water to flow down the turbo inducer(besides the small amount that hits it directly)

So what I'm saying is that you should all ditch your MAFS, buy a Megasquirt kit(or other MAP-based system if you like wasting money) and put a large cone filter directly on your turbo inlet. :D

Anyhow, thanks for the extremely interesting reading!

BTW, what ever happened to the MX5 tester guy? As has been said before, a few simple tests would be worth more than all 13 pages of this thread!

Edit: Pardon the obnoxious first post :wink:

[dsmtuned]

I've been thinking more seriously about how to mount a WI nozzle aimed into my compressor blades.

How do you guys think a nozzle and the hardware to keep it in place will affect the tornado like flow of air entering the turbo?

It seems like you would not want to disrupt this swirl as that would make the turbo have to work harder to suck air in.

Any thoughts?


-Craig

[TELVM]

An interesting article on pre-compressor WI for gas turbines:

http://www.alphapowersystems.nl/NewOrleans.pdf

"By injecting atomized water in the compressor of a gas turbine the parasitic work of the compressor is reduced due to quasi-isothermal compression.

Approximately 2% water to the mass flow results in a significant increase in the gas turbine?s power output.

... round-the-clock water injection could cause problems with erosion, water separation etc. The droplets must therefore be small.

The water to be sprayed is pressurised and heated up. Then, as it spouts out-of a swirl nozzle, explosive flashing takes place. The result is a surprisingly fine hot plume of tiny water droplets, ready to evaporate as they enter the compressor.

By using a swirl nozzle and by supplying pressurised hot water, the combination of spraying and flashing results in droplets roughly ten times smaller in diameter and thousand times smaller in volume and weight than the droplets of a normal swirl spray device.

The amount of heat extracted from the compressor air by evaporation is much greater than the amount added through the hot water spray. As a result, the temperature drops and the compressor discharge temperature is reduced. This results in less parasitic work of the compressor ..."


Seems the higher the pump pressure & the finer the atomizing, and the warmer :shock: the injected water, the best.

[dsmtuned]

I searched more about these "SwirlFlash Nozzles" and it sounds like you have to pressurize the heck out of the system to get the droplet size they have. 40-150 Bar!! :shock:

Here is a quote:
-----
The patented SwirlFlash? technology is based on a simple but robust principle. A liquid is pressurized (typically 40-150 bar), heated-up to about 200 C and fed to a swirl nozzle. Due to the swirl movement the liquid (for example water) spouts out of the nozzle in a typical spray pattern which has the shape of a cone. The droplets size is about 25 micron. However, when the water is significantly above the boiling point at the ambient pressure, it starts boiling violently (flashing). As a result each droplet of 25 micron explodes in a thousand fragments, each having the size of about 2.5 micron. The typical spray cone of a swirl nozzle changes as a result of partial flashing to a parabolic shape. The ultra fine spray ensures almost instant evaporation and cooling.
-----

Here is where it was:
http://www.alphapowersystems.nl/swirl.htm

I apologize if this is what is contained in the pdf file, but the computers over here in China (travelling now) will not let me open it up. :roll:

Has anyone used these SwirlFlash Nozzles?

-Craig

(edited for clarity)

[PuntoRex]

Quite a while ago, someone has posted this "SwirlFlash" here somewhere (sorry I don't remenber who).

I think the key is not at the nozzle, but the pressure & temp. These two are also the hardest portion for us to deal with. :cry:

And the whole system is definitely bigger & heavier than what we are using now. That would be a bad thing. Maybe sometime it's the difference of doable & mission impossible. Especially in a crampped engine bay like mine.

[Psi-Fi]

Quite a while ago, someone has posted this "SwirlFlash" here somewhere (sorry I don't remenber who).

I think the key is not at the nozzle, but the pressure & temp. These two are also the hardest portion for us to deal with. :cry:

And the whole system is definitely bigger & heavier than what we are using now. That would be a bad thing. Maybe sometime it's the difference of doable & mission impossible. Especially in a crampped engine bay like mine.

not so undoable... run pure water in your radiator fluid and run a 30 lb cap. tap the aquamist feed from the radiator so now you have heated water 180-220 degrees f and its also pressurized to 30 psi above atmospheric. to fill the system just keep the overflow can filled to spec.

[PuntoRex]

That's really a good idea!

But (a big one), it's too "agressive" for a road car, isnt't it?

I don't need anti-freeze in where I live, that's alright, but I do need anti-corrosion & lubrication from the coolant.

Seeing rust in the injected water would freak me out.

[Richard L]

If you just heat water in a strong chamber to a specific temperature and maintaining it electronically, you will automatically generate the required pressure, I posted a chart sometime ago regarding water pressure against temperature, I believe at 370C, the natural water pressure is something like 200 bar!!!

[PuntoRex]

That's impressive!

Richard, can HSV take this pressure ? :twisted:

[Richard L]

Defintely a no no no. The HSV is not suitable for this job.

I too having troubll finding the chart I posted on water pressure against temperature in a closed vessel on another post.

I think you need a diesel injector to do the job.

[JohnA]

Shame how little this thread has moved on since I last checked.

PuntoGT's errosion pics are definately not conclusive proof -- they could well be due to the poor filtration characteristics of K&N cones, which become laughably bad after cleaning them once or twice (just stick it in front of a strong light and you'll see what I mean!)

I've now got a SupraTT which I'll fit with precompressor injection and see how far the stock turbos can go.
They are good candidates for such experimentation, as they are undersized for 400+bhp, in fact they leave their efficiency islands right after 1bar.

The shape of the intake plumbing is kinda weird, so I need to figure out where to optimally locate the nozzle(s).

Fingers crossed.

[PuntoRex]

...just stick it in front of a strong light and you'll see what I mean!
....

There're holes everywhere, yes, but that's the way it works!

According to K&N, the filtering is not done by "blocking" the particles, but adhering them by the oil & the thin thread end splits. So you always need oil in such filter. (and do not use them after 20 times of cleaning, at that time, those thin hairs are mostly gone)

I've been using K&N for years without filtering problem. Yes indeed, it's not that clean as the stock paper filter can provide, but very close. And it out performs a lot in flow.


Let's admit it, those obvious radial traces on the blades can only be made by "fluid", not "particles" from bad filtering. Marks by particle impacts should look like dots (think sand blasting), not lines. Because they are always bouncing around when contacting solid surfaces. Fluid would mostly stay on the surface & flow down, that makes line marks.

I know the corroded blades in the picture is scaring (to some at least), but I absolutely hope to see other's successful results. (And actually I'm still using this same turbo now & it seems OK. )


I remember Larry had used pre-compressor injection for quite a while & the condition of his turbo was far better then mine.

About the jet location we were talking over a lot previously, I have a reverse idea that maybe getting the jet(s) further upstream would be better.

Why? Simple, droplets have more time to evaporate before touching the blades, and evaporation makes them smaller. As to the water which gathers on the inner wall, I can only guess the issue would be minor. Also as mentioned, there's always a step in the turbo inlet.

My pre-compressor jet was quite close to the blade, about only 8" far. It was located in the elbow and mosly "aimed" at the blade. In contrast, Larry's jet is further upstream, and injects vertically to the flow.....

That's what I'm thinking....

[JohnA]

There're holes everywhere, yes, but that's the way it works!
That's not my definition of a 'working' air filter mate :lol:

According to K&N, the filtering is not done by "blocking" the particles, but adhering them by the oil & the thin thread end splits.
I know their advertising literature mate, I've wasted money on them like so many others before me (and after!)


I've been using K&N for years without filtering problem. Yes indeed, it's not that clean as the stock paper filter can provide, but very close. I could explain how they reach their filtering percentages, but won't bother (it's a con)
Have a look on my website under 'intake' if you're really interested.

And it out performs a lot in flow.
no it does not.
I've got the pressure measurments to prove it.
Let's admit it, those obvious radial traces on the blades can only be made by "fluid", not "particles" from bad filtering.
I've seen worn compressors that looked exactly like yours, and these people had never used W.I.

I know the corroded blades in the picture is scaring (to some at least), but I absolutely hope to see other's successful results. (And actually I'm still using this same turbo now & it seems OK. )
The compressor efficiency is a bit lower, no big deal, don't worry too much about it. Just use the stock filter from now on.

About the jet location we were talking over a lot previously, I have a reverse idea that maybe getting the jet(s) further upstream would be better.
I've just set up mine as upstream as possible (right after the MAF)
There is a 'split' afterwards, with the pipes going to each turbo (say 120degree turns) Then there is a 180degree turn for each turbo.
Who knows how the atomisation will be after all this torturous path...

When I dismantle the turbos in the future, I'll stick a W.I. nozzle right opposite to each compressor. Then we'll know which one works better.

Wish me good luck :wink:

[PuntoRex]

...

Just use the stock filter from now on.

Thanks for your advice, but I just can't.

The space for the stock air box has been occupied by water pump for years, and nowhere else to mount it. (if you get a chance to glance round the engine bay of a Punto, you'll know what I mean)

...

Anyway, about the air filter "myth", I've read some. I'd say they are very "interesting". Opposite opinions are everywhere, just like everything else.

In my own car & my own experiences, the cone type K&N works the best. (with proper heat shield)

About the so-called "stock type" plate filter from K&N, I think it's indeed useless. The area is just too small. In this case, stock filter does sometimes perform better. I know it because I had tried it. But cone type is another story.


This somewhat injured turbo on my car is the 3rd one I have. And I had also used K&N with the first 2. Their compressor blades were not like this.

So, it's quite "interesting", eh?:wink:


(Actually, as I'm getting older & older, I found it's harder & harder to distinguish truths & myths :? )



...
Wish me good luck :wink:

Yes, I surely do! I do wish you good luck & am also looking forward to see your results. :smile:

[JohnA]

(Actually, as I'm getting older & older, I found it's harder & harder to distinguish truths & myths :? )
I'm probably even older, and can now see many myths from miles away -- especially those fuelled by commercial interests


I do wish you good luck & am also looking forward to see your results. :smile:
I'll post pictures later on, I've now got the LED in the cabin to indicate when WI kicks in.
One good thing is that you can time a run with W.I., then another one with a different activation point, then pull the fuse and do another without WI

[JohnA]

here's more reading on the subject :smile:

http://www.turbineinletcooling.org/intro.html

http://www.turbineinletcooling.org/News/ETDec03.pdf

http://www.patentstorm.us/patents/5867977.html

http://www.patentstorm.us/patents/5930990.html

http://www.patentstorm.us/patents/6173564.html

[JohnA]

OK, here's another issue that's been bugging me. Ideas are welcome.

Suppose that we want the water injection stream to hit the compressor blades straight on. If there is a 90degree turn of the pipes right before the turbo this might be easier than I thought -- my engine has two compressors, each one 'blessed' with such a steep turn :wink:

This is how we'd like things to be

http://www.max-boost.co.uk/stuff/PreTurboInjector_ideal.jpg

However, since the air is moving at seriously high speeds and it has inertia, this is what we'll probably experience:

http://www.max-boost.co.uk/stuff/PreTurboInjector_real.jpg

In that case, perhaps we could account for this by setting the nozzle a bit offset:

http://www.max-boost.co.uk/stuff/PreTurboInjector_fix.jpg

What do you guys think?

I've found some pictures of compressor blades erroded by injected water in the wrong size/pattern. OK these were power-generator turbines and they were being injected continously, but the principle is the same. (the leading edges of the front compressor blades were eroded, not unlike Punto's pic lol...)

[rarson]

The problem I see with that is, by directing it toward the wall like that, I think you'd have the puddling problem with water streaming down the pipe as hotrod explained. I agree that your second picture seems accurate, that the water would not flow straight, but I don't think the third picture solves the problem.

I honestly think the best design, for a system with less than optimal atomization, is a nozzle very close to and aimed at the shaft nut. Not only would the nut help "atomization" but it would also alleviate injecting water at the blades... it would seem to me that the water would more likely go through them instead of at them.

[JohnA]

The problem I see with that is, by directing it toward the wall like that, I think you'd have the puddling problem with water streaming down the pipe as hotrod explained.
Maybe it won't be aiming at the wall, this is a rough sketch to show the idea of offset aiming.

As the water is at a pressure differential of over 10bar it should have strong resistance to the air rush that is below 1 bar (precompressor, so no boost!)

Has anybody measured the vacuum before a turbo?

[cheekychimp]

Having just worked my way through all 15 pages of this thread, I feel somewhat reluctant to pose a question that is moving away from the CURRENT focus of the questions at hand. That said, having gained a huge amount of really useful information by following this thread for so long a very crucial matter is staring me right in the face and I really just have to ask.

Most MAS systems place the MAS immediately after the air filter and prior to the turbo. A lot of DSM vehicles are however using the larger bore and less restrictive GM-MAS with a MAF-T in what is generally referred to as a 'blow through' setup. That is it is placed in the intake tract between the intercooler outlet and the throttle body.

My question here is that since this is essentially a hotwire MAS, is any WI whether it be pre-compressor, pre-intercooler or post-intercooler but PRIOR to the GM-MAS going to have a detrimental effect on the MAS readings and consequently going to cause problems with the A/F ratio that the ECU calculates?

Paul

[masterp2]

really good question

[hotrod]

Hot wire mass air flow systems will not tolerate any liquid mist in the air stream. I can say this from personal experience having gotten a bit of water inside my air filter box which got pulled past the air filter under high intake vacuum. It led to an instantanous CEL and fuel cut. The engine would hardly run until the system dried out.

In the tests on wet compression, only a fraction of the water is evaporated in the compressor proper, so I would expect some small dropplets to survive.

You might ????? get away with one down stream of the intercooler. But I have my doubts. I know it would not work just down stream of the turbo if any mist survives the trip through the compressor scroll. As far as WI post intercooler, and before the blow through sensor --- don't even try it ----- won't work.

Larry

[cheekychimp]

Hot wire mass air flow systems will not tolerate any liquid mist in the air stream.

I am not going to dispute this, because I quite see the logic, but then what happens in rain, fog or high humidity conditions? At least some water must get into the intake tract? If not, how would we ever know to make a comparison between the effects of using WI and driving a car on a cool misty evening in autumn?

Like I said, I am NOT disputing this, I just want to know if injecting water at a very high temperature as was suggested above (i.e. at boiling point and under pressure) would result in sufficient evaporation or if the particles of water/alcohol would be of sufficiently small size to negate any adverse affect on the readings of the MAS.

[Richard L]

To day, we had a chance to experiment on fabricating a holder for pre-turbo injection.

Haven't yet finalise the method of clamping.

Here are a few pictures:

http://www.aquamist.co.uk/forum/gallery/precooler/1.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/2.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/3.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/4.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/5.jpg

[JohnA]

very interesting.
How about an aluminum fitting to replace the hose sitting in the pressurised airstream? Would the standard bits survive the conditions?

[Richard L]

very interesting.
How about an aluminum fitting to replace the hose sitting in the pressurised airstream? Would the standard bits survive the conditions?

The air pressure will be less than a bar. The intention was to use two jubilee clipd, one to clamp the disc bracket and the other clamps the silicon hose to the turbo flange.

We might try to make a lip to go over the turbo flange (interference fit). This bracket will be tested on a twin turbo skyline about 800 horses. The car in concern has inlet heat soak problem becuase it has pot injection fitted.

The goal was to achieve 900 horses with twin 400bhp turbos - it will show if the pre turbo idea work of not. We have found that running 3 boost yields less power than 2 bar boost - nearly 70 horses loss!!!

[JohnA]

Correction: I said 'pressurised' but technically it will always be in vacuum of course.
Still, it won't be sitting in engine bay under atmospheric conditions like it normally does, will it?

[Richard L]

Correction: I said 'pressurised' but technically it will always be in vacuum of course.
Still, it won't be sitting in engine bay under atmospheric conditions like it normally does, will it?

I am not sure how much air pressure the 3mm x 6mm fins will withstand - the more aerodrnamic the sharp the less friction and less force acting on the fins.

We have ordered some new cutters to profile the fins properly.

[JohnA]

next question: how is the pressurised water pipe going to get through the air intake hose? (gracefully I mean, without DuckTape :lol: )

[Richard L]

next question: how is the pressurised water pipe going to get through the air intake hose? (gracefully I mean, without DuckTape :lol: )

Interesting question.

Since it is experimental, the 4mm hose will enter the inlet tract throght a small hole on the air filter or silicon hose. Ducktape may be employed- only joking.

The final solution will depend on how well the theory works. I have great confidence that it will work - just want to know by how much.

We have unlimited use of a mustang dyno - in fact the owner of this skyline owns the dyno - he is very interested to follow this project throught.

Here is his company: http://www.gtart.co.uk
and his car: http://www.aquamist.co.uk/phpBB2/viewtopic.php?t=838

[JohnA]

Where is this dyno based?

I'll do a few dyno runs myself on the supra in a couple of weeks (hopefully) with precompressor injection on/off/various activation points.

[Richard L]

The link above has the exact address of GtArt. It is between Lewes and Brighton. About 10 minutes form my home.

Are you long way away from Sussex?

[JohnA]

I could arrange to be there, if we could squeeze some dyno time to try 2-3 variations of the pre-compressor theme.
It doesn't take long to swap settings, everything is easily accessible.
It would be very interesting, I haven't seen such figures anywhere else. :D

[Richard L]

He charges 70 pounds an hour to trade users - he will do exactly what you want him to do - just plan your events clearly or it will be a lot of time wasted.

How far do you have to travel to Brighton?

[JohnA]

One hour, if there is no traffic.

There are no pictures of the dyno on the site, not even a full address.
Is it a hub dyno?
It's hard to get traction on the supra when the second turbo kicks in.

At ?70/hour I'd hope that we would get a printout for each run, I want them for my website.

With some preparation, I could be over in 1/2 hour max.

[Richard L]

You need to ring him to get the direction because there are too many people turn up without appointment.

The dyno is a 1000BHP Mustang dyno - I know very little about it.

Here is a picture:


http://www.aquamist.co.uk/forum/gallery/gtart/dyno.jpg

[Richard L]

here is some information of the dyno.

http://www.gtart.co.uk/news.htm

[JohnA]

The supra will spin like a bitch most likely :lol:

I may ask Abbey what they charge per hour on their dyno. It might be cheaper on a 2WD and it's a hub dyno, so no need to strap the car and no wheelspin either (much more repeatable figures)

[masterp2]

That is really nice.

[masterp2]

I have a 4 temp data logger now, using it for cooling stack air flow mods, and increasing CAC function. I would love to try something like this. While I have a garrett, no doubt they all have different inlet dimensions. Would that prove impractical? I am totally willing to test.

[masterp2]

PS where is all that data logging that gelf (I think) did?

[dsmtuned]

I worry about what the 4mm line will do inside the intake. I mean the hose is not very areodynamic itself. Do you plan on running the line next to the inside wall of the intake?

Nice job on that jet holder by the way!

-Craig

[Richard L]

I have a 4 temp data logger now, using it for cooling stack air flow mods, and increasing CAC function. I would love to try something like this. While I have a garrett, no doubt they all have different inlet dimensions. Would that prove impractical? I am totally willing to test.

If the theory works well, I am going to develop it further. At present, we just use an extra clip to hold the bracket in place. The OD of the bracket can be machined to match the OD of the turbo flange.

At present we just see what the set up can achieve to further the turbo's capability, it is stuck at about 790bhp (twin turbo) at 2 bar. Higher pressure ratio (3 bar) produced less power.

We shall see.

[Richard L]

I worry about what the 4mm line will do inside the intake. I mean the hose is not very areodynamic itself. Do you plan on running the line next to the inside wall of the intake?

Nice job on that jet holder by the way!

-Craig

Not sure what to do yet with the 4mm line, the bracket will only serve to obtain some results on the dyno. I think the 4mm line will exit the inlet tract a little way towards the air filter.

I am sure it will flap around but when the line is pressured at 7 bar, it become quite stiff. Ideally the water channel can be incorporated into one of the fins of the bracket.

[Richard L]

Made some more improvements.

http://www.aquamist.co.uk/forum/gallery/precooler/11.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/12.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/13.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/14.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/15.jpg

[Greenv8s]

That's a beautiful bit of machining, but I'm a little concerned about the stiffness of those three mounting fins. I would guess there is plenty of vibration to excite them, and if it hits resonance at any point I could see the whole lot dropping into the turbo. It looks as if you have space to make the fins two or three times deeper (axially) and this would strengthen and stiffen them immensely. It would also give you the opportunity to give them a more streamlines teardrop cross section if you wanted.

[Richard L]

That's a beautiful bit of machining, but I'm a little concerned about the stiffness of those three mounting fins. I would guess there is plenty of vibration to excite them, and if it hits resonance at any point I could see the whole lot dropping into the turbo. It looks as if you have space to make the fins two or three times deeper (axially) and this would strengthen and stiffen them immensely. It would also give you the opportunity to give them a more streamlines teardrop cross section if you wanted.

I was surprised how strong those rails are, you can't move them very easily with your fingeres. I do see your point when it hits resonance (not sure when) it will cause it great distress.

The rail will be much thicker when the cutting tools arrive next week, it will be in tear drop shape. At present, we use what tool we have available.

The rails are 2.5mm thick 4mm tall. When we are putting the water way inside one of them, it will be around 3mm x 6mm tall. We have plenty of room to do that as you mentioned.

This quick job was for just one dyno test next week to see how if it can puch the turboes beyond its bottle neck at just under 790HP at the wheels (twin turbo set up).

If it does, we will invest more time on it.

[JohnA]

Yeah, teardrop profile, that would be ideal (tears of joy :lol: )

[masterp2]

Just my 2 cents. This will be a significant restriction, with air moving over it at over 100 mph. I see a possible flow issue. I would taper each side to a V.

Also, if it is possible to integrate the mount, such that no cross sectional area is lost at that location (larger diameter chamber that flairs back to normal) that might mitigate head loss.

[Richard L]

Which "side" are you referring to?

The ID of the fitting is the same as the ID of the turbo flange.

Just sized the images after anodising. You should be able to see the sides better.

http://www.aquamist.co.uk/forum/gallery/precooler/21.jpg

http://www.aquamist.co.uk/forum/gallery/precooler/22.jpg

[masterp2]

When airflow gets to the fitting, it will either have the additional obstacle (restriction) to overcome.

That pic looks nice, a sharp V on the fins, if the backside tapers back to a V the same way, good job. I would just "round" the transition, instead of the flatspot or angle change. If you like, you can widen the fins, in the effort to make them as thin as practical, the idea being invisible to the airstream.

With respect to the ID of the fitting, if you can enlargen it just a bit, in a blended curvature, so that no x-sectional area is lost (from the addition of this beautiful cut piece) that will help minimize head loss as a suction side restriction (the worst kind).

The last thing you want is a failure due to cavitation. that is what I fear. And larger diameter, at the fin location, may help.

[Richard L]

Thanks for the advice and useful suggestions. I will thicken up the fins for the next time but will do for the time being.

At present the holder ID is identical with the turbo flange ID. See picture below. There are no breaks - red circle. As far as the air is concern, it cannot see the gap at the intersection. The chamfer on the OD is gone on the anodised version.

I will also round off the tapered fin to minimise any notch developing. It is knife sharp at the moment.

http://www.aquamist.co.uk/forum/gallery/precooler/17.jpg

[masterp2]

Like I said before, I'm impressed. Good luck!

[Richard L]

I will report the results as soon as I have it. I am keep my fingers crossed and hope the fins hold up. I cannot not rob any more machine room time anymore.

[Greenv8s]

That's a beautiful piece of machining and I look forward to hearing how well it works. As a non-turbo user this is the first time I've looked at the insides of a turbo this closely. It looks as if there is an abrupt contraction just downstream of your new fitting. Is that real, or is it an optical illusion? If it's real I imagine it's there for a reason but I can't see what the reason would be.

[Roverdose]

that does look fantastic. just wondering about the water feed pipe.

how do you get those small filters in the jets? do you screw the barbed pipe part into the base?
if you did, could you then machine a thread onto the turbo side of the plate, and screw the jet onto it.

so you would have a thread on the outside of here..
http://www.aquamist.co.uk/forum/gallery/precooler/21.jpg

and a cone on the other side to help flow.

then machine a hole into one of the fins to get the water to the jet.

if you see what i mean?

Drew

[Richard L]

The water will be channeled into the jet via the rails, possible redesigned the jet jet to accept water on the side, similar arrangement as the side inlet fuel injectors.

The tail of the jet will be a cone shaped nut, as described by you.