Tuning W/I via Humidity Sensors

[rotrex]

Hi,
it means with the current w/m flow the engine starts knocking before I reach the maximum torque as I advance ignition.
If I would up the flow, I would at some point not be knock limited anymore.
As I would advance ignition triming any further, torque would just drop as peak pressure is reached to early. But there would be no knock.

Thing is that I only use a 3l tank. On a track with lots of WOT, e.g. Spa, it only lasts some 15 to 20 minutes. On tighter tracks, e.g. Anneau du Rhin or Bilster Berg 3l last for more than 30 minutes. I am usually not for longer on track within one session anyhow.

On the road it may well last 200 miles excluding longer 130mph+ streches of Autobahn driving.

[Flr Power]

That's what I don't understand. I am using w/m to not be fuel knock limited. So I spray around 30% to fuel.

Do you think it would be wise to install a 20 liters tank and then increase the w/m flow? Or is really how you want it tuned?

[rotrex]

30% water to fuel or mix to fuel?
I know I have not hit the limit in terms of what I can achieve with w/m injection yet.
I am happy then egine does not blow up in first place. :-)

It is a Lotus Elise S1, 20l is not really usefull to carry around. It is a very small and light car.
I may consider installing a additional 5l tank for track use. But doing a 5min break every 20 minutes is also nice. Drink, refuel, ...

[parmas]

Originally Posted by rotrex

Point is that the water mainly acts in the chamber or to a degree the compressor.
It needs heat to evaporate. Lots of it. as the the droplets cool down due to evaporation and the water's vapor pressure is rather low and drops with temp, water does not evaporate fast unless you provide lots of HEAT.

AGREED

This is static effects (heat of evaporation) vs. dynamic effects (heat capacity, vapor pressure vs. temperature).

Intake air cooling in the pipes is a minor effect. Folks have tried that for decades.
the temp readings barely change.

Could you define the term barely in terms of DegC?

issues like water replacing air reducing power is akin to "running rich costs power as it replaces air". It is a minor downside compared to be able to run at MUCH higher boost than without and/or getting higher cylinder pressures from more ignition advance. -> more power.
You lose a few percent of potential power and gain several times that as real power.
A turbo increases backpressure, hence costs power. As you know, that is only half the story.

Agreed, unless engine is not capable to run high boost

shoving the maximum amount of fuel and air into an engine is only one part of the story making power. You also need to be able to extract this energy and convert it into mechanical power. This needs optimum ignition timing to achieve max. cylinder pressures at the right point in time.

[COLOR="rgb(139, 0, 0)"AGREED[/color]

What you try with humidity sensors and some cooling is like trying a different cone air filter brand to gain more power. The effects are negligible compared to water or w/m in the chambers.
You ignition will have no issues unless you flow massive amounts of water.
In my engine this was at about 600cc/min pure water. Adding a tad of dwell time in the ECU in the area of the map fixed even that.

If you use methanol in your mix or pure methanol, that all changes.
Methanol has a lower vapor pressure and lower heat capacity. it evaporates much faster given the same temperature environment is provided compared to water.
Energy per time unit is cooling power. As more methanol evaporates per time unit compared to water, it tends to cool the air more in a intake path. In the chamber itself or in the turbine, the difference is of less importance as these places are a much hotter environment.

The chamber is water or methanol flow limited in most cases regarding how much they can evaporate and how much cooling you can achieve.
More fluid= lower air temp in cylinder (where it counts the most) = more power.
Of course there is a limit for a given IAT and compression ratio. At some point, it will flow out of the exhaust. The WW2 engineers tried that, too.
If you stick to the here recommended 10-20% water to fuel ration, this is of no issues. If your boost goes above 25 or even 30 PSI, you can add much more.
See RICERACERs RX7 development. His engine works in the same extreme regime as did the WW2 aircraft engines. The more w/m he adds, the more boost he can run and the more power he gets from the engine.
In his high boost setup with pre compressor injection, the IAT reduction is much more pronounced and an important element to gain big HP numbers.

you can take w/m to the extreme. Just add more boost and more fluid.
You are essentially only limited by the mechanical strength of your engine. 100PSI boost in a 1.5l 4-pot. That's what BMW and others did in the 80s F1 engines getting 1000+HP.

The boost pipes are energy limited when it comes to water evaporation. There is not enough heat to evaporate all the water. The droplets plainly become too cold to further evaporate. This is actually nice, because now they can make it where you really want to have them, in your engine.

the aquamist system in the WRC cars worked so well with pure water because the turbos were used at insanely high pressure ratios producing extremely hot charge. This was caused by the intake restrictors required.
And even they used direct port injection. They did not care about cooling in the pipes. They (and we) cared about cooling the charge in the cylinders. Even more specifically, you care about cooling unburned charge in general and eve more in the area of pressure or shock waves while the flame front advances to prevent spontaneous ignition aka knock. Shock waves compress the charge heating it further. Droplets evaporate faster, cool the mix and by that reduce the pressure. It dampens pressure waves. It also slows heat up at the flame front. Thats why it slows flame speed.

The music plays in the cylinder. It is just not that easy to get it there in a uniform way. This is where all the discussions from nozzle location come from. It is a lot about practicality vs. complexity.



The turbine can also evaporate a lot, if pressure rations are high or efficiency low.

If you use plain water, try a intercooler or charge cooler first.
Your cooling to the due point is not worth the trouble.

I have used a Rotrex C30-74, now replaced by a C30-94, with various nozzle locations and mixes.
non-intercooled I saw 100C IAT with 0.6bar of boost. just look at your compressor maps and do the math.

Injecting water does very very little to change the IATs. Maybe 20C.
Injecting methanol before or after the compressor has more dramatic cooling effects. The most extreme for me was 300cc/min 50:50 water methanol pre-compressor injection. I saw IATs actually dropping to ambient temps as boost was climbing.

I cannot see you getting only a 10C increase in IAT over ambient at 1 bar of boost without charge cooling and only minor post compressor water injection at 200g/sec of airflow.
There is something wrong or missing.


Once you use serious amounts of water that actually makes it into the combustion chamber, droplets do not like to take tight turns e.g. in a plenum with bend runners, all those factors like humidity and even intake air temp become of less importance.
See it as your cell phone playing music in the first row of a rock concert. it will make no difference to the audience.

If you want to make significantly more HP from your setup, especially without charge cooling, you need to get the water into the chambers, lean out the mix 12-13:1 and advance the ignition close to best torque or the knock limit.
if you hit the knock limit first, add more water/meth and further advance the ignition.

In the end, I now use direct port injection with 4 75ml/min nozzles 2" before the fuel injectors. Fuel i complimented by 4x 440cc/min injectors.

For the ignition, I am still knock limited, I found that up to about 60C, IAT plays little role to ignition timing. I only pull serious timing past 70C. Before that it runs really well with a particular aggressive sound.

Ignition timing at peak torque is about 23? without WI and 29? with WI, both at 0.7 bar and 6500 rpm. Engine a build 1.8l Rover K with a 8.5:1 CR in a Lotus Elise.

Mapping the same engine without WI, I can't remotely achieve similar power levels.
The engine is charge cooled, but the pre-rad is currently too small. That does not help.
A new engine radiator sized pre rad is already sitting in my basement, but the car is away in winter storage until March.
Together with bigger lines and a bigger pump, IATs will drop.
This will allow me to extract more power without WI, but also with. The gains are just lower.

Try adding more water/meth and map for it. You will be surprised how well it works.

Pure water in my experience leads to rather slow (initial) burn speed requiring you to use a lot of ignition advance. This is harder to tune.
Adding 50% (wt) methanol compensates that to a degree and makes it easier to tune. The slope of ignition advance vs. power gain is more steady with a smoother transition to knock in my engine.
At least it does for a low to medium boost engine.
Once things are really hot in a 30+PSI engine, this is less of an issue.

Thank You for your comment. I did knew that eventually the chamber is the first priority but I was not sure that injecting more water well past due point will be negligible if tuned right.

Presently I am running 18psi on a GT2560R turbo / Toyota 1.5L @ 13.5 AFRs @ 6800Rpm / injecting 350cc post turbo.

Now the issue is that I do not know how much the engine could handle more boost with no issues, that's why I tried to make a more efficient tune.

[rotrex]

ww2 engineers did the tests up to 1.5:1 water to fuel.
The slope went always one way. more boost + more water = more power.
So there is plenty of headroom from your 350cc jet :-)
What engine is it? A 1FZ-NE? Forged pistons?
Cast pistons are more often than not the first thing to go sauer.
I still suggest to install charge cooling. It combines well with water injection.

I have seen intake air temps drop 10-20C max using water injection. More with adding methanol, even more doing precompressor injection.
Doing direct port injection, there is barely any IAT reduction, but I can still benefit to a similar degree, if not even more. I have no dyno, so can't tell you in terms of % power/torque.
A 10-20C drop is rather little compared to what a charge cooler does. Even a small one will reduce IATs by 30 C or more. The decent ones get you to 10C above ambiernt, so in your case about 80-100C reduction in IAT.
That is why I suggest let a charge cooler do the charge cooling and the water injection do the in-cylinder work. They both add to the performance potential.

if your current tune is already at best torque, so not knock limited, you only can up the boost to get more flow/power in its current configuration.
If your engines knocks before hitting best torque upon further ignition advance, add more water/methanol and try again.
Where the limits of your engine build are, you need to do your own research or blow an engine.
As you run no charge cooler, you may try precompressor injection.

Regarding AFR, I have found that such a "lean tune" provides good power, but generates considerable heat. I have only observed this on track. 12:1 yielded about 85C coolant temp and the 13:1 about 90C on that particular day. Both were with w/m injection. Nice thing was fuel consumption was significantly lower in that session.
On the road doing short burst of WOT, it probalby won't matter. temp wise

My cylinder air distribution is not that uniform with Nr.1 running a tad leaner. The plenum is not designed for forced induction. The plenum chsmber has a side entry and a uniform cross section along the width. This leads to more airflow at ther first cylinder and less towards the other end.
Running the engine at 12:1 keeps them all somewhere around this number. It lowers overall combustion temps and cylinder head and valve temps with it. In your case also turbo temperatures.
Riceracer propagates 11.8:1. This is rich best torque. It adds safety and costs barely any power vs. a leaner all out tune.

[parmas]

Originally Posted by rotrex

ww2 engineers did the tests up to 1.5:1 water to fuel.
The slope went always one way. more boost + more water = more power.

I AM SURE I DID READ IT SOMEWHERE

What engine is it? A 1FZ-NE? Forged pistons?

ACTUALLY 1NZFE - ARIAS FORGED 8.5:1 Compression. Although block is open deck and cylhead is completely stock


I still suggest to install charge cooling. It combines well with water injection.

Actually I do like the simplicity of my setup. Charge cooling is not considered for now. I also like faster spool up time

I have seen intake air temps drop 10-20C max using water injection. More with adding methanol, even more doing precompressor injection.
Doing direct port injection, there is barely any IAT reduction, but I can still benefit to a similar degree, if not even more. I have no dyno, so can't tell you in terms of % power/torque.
A 10-20C drop is rather little compared to what a charge cooler does. Even a small one will reduce IATs by 30 C or more. The decent ones get you to 10C above ambiernt, so in your case about 80-100C reduction in IAT.
That is why I suggest let a charge cooler do the charge cooling and the water injection do the in-cylinder work. They both add to the performance potential.

I use the car mainly for street. Going slow and off boost will net about 35C-40C with an ambient temp of 25C. Injecting water @ 2psi and full injection @ 10psi will not let IAT to go over 35C @ 18psi. After full throttle is closed, temps go below ambient to 15C-20C for several minutes.

if your current tune is already at best torque, so not knock limited, you only can up the boost to get more flow/power in its current configuration.
If your engines knocks before hitting best torque upon further ignition advance, add more water/methanol and try again.
Where the limits of your engine build are, you need to do your own research or blow an engine.
As you run no charge cooler, you may try precompressor injection.

PRE-compressor injection is going to be used within the next weeks beginning with 150cc and leaving 350cc post turbo. I never heard knock till now and ignition advance is about 30Deg @ 6500Rpm 18psi. Again since engine block is not sleeved and cylhead is bone stock, it is likely to be boost limited

Regarding AFR, I have found that such a "lean tune" provides good power, but generates considerable heat. I have only observed this on track. 12:1 yielded about 85C coolant temp and the 13:1 about 90C on that particular day. Both were with w/m injection. Nice thing was fuel consumption was significantly lower in that session.
On the road doing short burst of WOT, it probalby won't matter. temp wise

At 12.5AFR the plug shows considerable amount of soot that it could be leaner. At 13.5 AFR seemed best. At low boost 0-4psi is also 14AFRs. The only difference I saw is in EGT temps. At 12.5, I was 650DegC while @ 13.5AFR above 710DegC

My cylinder air distribution is not that uniform with Nr.1 running a tad leaner. The plenum is not designed for forced induction. The plenum chsmber has a side entry and a uniform cross section along the width. This leads to more airflow at ther first cylinder and less towards the other end.
Running the engine at 12:1 keeps them all somewhere around this number. It lowers overall combustion temps and cylinder head and valve temps with it. In your case also turbo temperatures.
Riceracer propagates 11.8:1. This is rich best torque. It adds safety and costs barely any power vs. a leaner all out tune.

Mine is rich on cylinder one since it is the last cylinder on the side distributed plenum. The strange thing is that all the other three are well balanced. I actually do cylinder trimming on that one and reduced fuel actually.

....... Alex

[rotrex]

Hi Alex,
you probably inject a mix with high methanol content. That would explain the low air temps.
Also your turbo charger seems to run in a really efficient range.
I'd still consider adding even a small charge cooler. I use a PWR barrel charge cooler.
I barely adds any volume.
http://www.chargecooler.co.uk/index....index&cPath=67

Spool time should not really affected by a few liters of volume, especially as you have a throttle body right on the intake plenum. Your turbo is still spinning pretty fast after you press on after gear change. A front mounted one might add a few ms.
your turbo pumps out 100 liters per second at 1bar or 200l at 0bar. To fill a charge cooler of 10l of volume after a gear change at full chat from atmospheric to 0.5 bar takes mere milliseconds. From 0.5 to 1 bar an other few more. You won't feel it.

You plenum does the same as mine. Air speed drops as air is consumed from plenum intake to the last cylinder. Usually the first one gets the most and the others less.
The design itself, e.g. inlet angle and diameter, can change the distribution to a degree. But in general it fits the picture. First one gets the most air and the last the one the least.
In my case 1 is lean and 2-4 are about equal . In you case 1-3 are equal with 4 being rich :-)
The ideal forced induction plenum with a side intake is either fed by a secondary plenum and a long slit. The secondary plenum having a conical shape with the diameter reducing to 25% of the cross section.
The other solution is actually to reduce the cross section of the plenum chamber itself from 100% to 25% at the last cylinder.
This way air speed and with it pressure and flow stay uniform across all 4 runners.

see:
http://honda-tech.com/welding-fabric...folds-3016531/
and:
http://www.clubgti.com/showthread.ph...final-cut-done

If I ever being bothered I might try it, but space is really tight.

[parmas]

Originally Posted by rotrex

Hi Alex,
you probably inject a mix with high methanol content. That would explain the low air temps.

I only inject pure distilled water at this time. I have all intake and turbo housing + exhaust ceramic coated. After a run you could feel the intake plenum like a fridge cold.

Also your turbo charger seems to run in a really efficient range.

Yes, I calculated CFM according engine vs turbo in order not to choke nor surge the turbo

I'd still consider adding even a small charge cooler. I use a PWR barrel charge cooler.
I barely adds any volume.
http://www.chargecooler.co.uk/index....index&cPath=67

Spool time should not really affected by a few liters of volume, especially as you have a throttle body right on the intake plenum. Your turbo is still spinning pretty fast after you press on after gear change. A front mounted one might add a few ms.
your turbo pumps out 100 liters per second at 1bar or 200l at 0bar. To fill a charge cooler of 10l of volume after a gear change at full chat from atmospheric to 0.5 bar takes mere milliseconds. From 0.5 to 1 bar an other few more. You won't feel it.

Thank you for your suggestion, I will consider it if pre-compressor injection is unsucessful

You plenum does the same as mine. Air speed drops as air is consumed from plenum intake to the last cylinder. Usually the first one gets the most and the others less.
The design itself, e.g. inlet angle and diameter, can change the distribution to a degree. But in general it fits the picture. First one gets the most air and the last the one the least.
In my case 1 is lean and 2-4 are about equal . In you case 1-3 are equal with 4 being rich :-)

Actually mine is 2-4 are equal while 1 is rich. Throttle body is on the 4th cylinder side of plenum. I solved this issue by reducing fuel on that particular cylinder only.

The ideal forced induction plenum with a side intake is either fed by a secondary plenum and a long slit. The secondary plenum having a conical shape with the diameter reducing to 25% of the cross section.
The other solution is actually to reduce the cross section of the plenum chamber itself from 100% to 25% at the last cylinder.
This way air speed and with it pressure and flow stay uniform across all 4 runners.

see:
http://honda-tech.com/welding-fabric...folds-3016531/
and:
http://www.clubgti.com/showthread.ph...final-cut-done

If I ever being bothered I might try it, but space is really tight.

Mine is tight also, It could barely hit the front panel as it is.

.........Alex