Originally Posted by PuntoRex

I had the similar experiences about the trigger point.

Currently I use 1*0.5 at pre-turbo & 1* 0.6 at pre-TB, triggered at 10psi, max boost around 18~20psi.

I can be sure that the turbo spools up much more quickly in the 10~20psi region than the 0~10psi. Probably, that's the work done by water.

Which stops me make it earlier is the worse atomization in the low DC injection. I'm a little worrying about the larger droplets caused by this, especially pre-turbo.

Originally Posted by Gelf

....What im looking for

Consistantly low turbo exit temps
High IC efficiency

Originally Posted by JohnA

I don't know how you reached this conclusion.

Originally Posted by JohnA

The ideal intake temp at the throttle is around 40C mate, certainly no less than 30C

Originally Posted by JohnA

In any case, intake temps are not the whole story, especially if they max at 30C without WI
To me it looks like your engine setup at these ambient temps needs no W.I. for lowering intake temps.

Originally Posted by JohnA

It could be a different story if power is measured with different jets, while you run higher boost pressures. That would mainly reflect the advantages of in-cylinder cooling, which I'd think would be far more important than chargecooling in the winter.

Originally Posted by JohnA

Very interesting post, by the way.

Originally Posted by masterp2

WOW! Excellent work!

Totally impressed.

I am getting set to inject a v-8 duramax turbo-diesel with WI, also pre-turbo and post-CAC (charge air cooler, IC, sorry about the language). This is a workhorse vehicle in need of supplemental cooling in extreme conditions.

Your charts will be helpful. I have set out to use a modulated 1st stage for water savings and activate the full 2nd performance stage with a throttle switch. The modulated 1st stage might consist of 1 .3 nozzle pre-turbo, and (2) .3 nozzles post CAC (or just post CAC nozzles). The performance 2nd stage might be 2 additional .3's pre-turbo.


thoughts?

Thank you for this research.

Originally Posted by JohnA

Remember that intake temps consistently below 20C at the throttle will rob you of power. If that is true under boost as well, then you're compromising the engine's efficiency.

Originally Posted by SteveR32GTR

interesting. is that 100% fact even with forced induction?

on various cars ive had i always fit the charge temp sensor at the throttle and the car seemed to go better the colder the temps were, even way below 20degC on some stupidly cold nights (and very good IC/WI setup, lol).

id almost be inclined to say thats true in n/a form where you havent got boost pressure messing with the fuel/air mix, but on turbo cars?

Originally Posted by rotrex (Post 22044)

Having done some experiments with pre compressor injection last summer and fall, I can add a little to this.
Rotrex C30-94, jets tried: Aquamist 0.3 and 0.5mm. Mix 50:50 water methanol by weight.
IAT is measured after a water cooled charge cooler half way before entering the intake plenum. Throttle body is before the SC. Boost is 0.83 bar at 7000 rpm.
There is almost no influence on boost. At least nothing I can distinguish from effects of ambient temperatures. So all constant within a few percent.
The air temperatures drop as you increase jet size. Going from 0.3 to 0.5 mm introduced IAT fluctuation that got worse as ambient temps were lower. The sudden nature of those drops, you can see this effect also in the original data plots of this threat, can in my opinion be attributed to condensation in the charge cooler. The big droplets hits the IAT sensor, it has a little frame around it, evaporate and cool it down. Once the droplet is gone, it heat up to the air temp again. The fluctuations go worse as ambient temperatures dropped and as jet size increased. You feel the car modulate its acceleration as the ignition timing was advanced and pulled accelerating though a high gear due to the ignition air temperature corrections. The data logs show only some of the fluctuations, but not all of them compared to the felt fluctuations. The data rate recorded on the PC is too low.
To confirm, I removed the correction and the feelable effect was gone, but the log stills shows the fluctuations. On a 40C ambient day last summer, there were virtually no fluctuations a served with IATs as has 70C.

The observation of the original poster was that the throttle plate temp vs CC output temperatures dropped as flow increases. This is clearly a indication of evaporation.
At 50C turbo exit temperatures, much of the water and part of the methanol are likely still in the liquid phase, the charge cooler condenses the stuff further.
This big droplets collide with the wall at any bend or any obstacle, and cool is via evaporation. Unfortunalty, due to this effect, little of the spray arrives in the cylinder and mist.
During a rebuild with direct port jets on top of u-turn plenum runners. I could see the traces of the water stream on the front part of the combustion chamber and piston tops. These areas were light brown vs black.
Since moving the direct port jets on top of the plenum, but right behind the fuel injectors in the flat section, I needed to add more ignition advance. A clear sign that more mix effectively arrived in the combustion chamber as mist and not a fluid stream.

Once someone goes to significantly higher boost levels, the corresponding turbo exit temps of well above 100C or even 150C will evaporate much more fluid thrown at it. Riceracers setup is a good example. Condensation in the chagre cooler will still be present, but to a lesser degree.

Overall my little 0.3mm jet used now drops the IATs a tad, but the main work is done by now 4x 0.4mm c-type (150 ml/min) direct port jets.

Originally Posted by rotrex (Post 22044)

Having done some experiments with pre compressor injection last summer and fall, I can add a little to this.
Rotrex C30-94, jets tried: Aquamist 0.3 and 0.5mm. Mix 50:50 water methanol by weight.
IAT is measured after a water cooled charge cooler half way before entering the intake plenum. Throttle body is before the SC. Boost is 0.83 bar at 7000 rpm.
There is almost no influence on boost. At least nothing I can distinguish from effects of ambient temperatures. So all constant within a few percent.
The air temperatures drop as you increase jet size. Going from 0.3 to 0.5 mm introduced IAT fluctuation that got worse as ambient temps were lower. The sudden nature of those drops, you can see this effect also in the original data plots of this threat, can in my opinion be attributed to condensation in the charge cooler. The big droplets hits the IAT sensor, it has a little frame around it, evaporate and cool it down. Once the droplet is gone, it heat up to the air temp again. The fluctuations go worse as ambient temperatures dropped and as jet size increased. You feel the car modulate its acceleration as the ignition timing was advanced and pulled accelerating though a high gear due to the ignition air temperature corrections. The data logs show only some of the fluctuations, but not all of them compared to the felt fluctuations. The data rate recorded on the PC is too low.
To confirm, I removed the correction and the feelable effect was gone, but the log stills shows the fluctuations. On a 40C ambient day last summer, there were virtually no fluctuations a served with IATs as has 70C.

The observation of the original poster was that the throttle plate temp vs CC output temperatures dropped as flow increases. This is clearly a indication of evaporation.
At 50C turbo exit temperatures, much of the water and part of the methanol are likely still in the liquid phase, the charge cooler condenses the stuff further.
This big droplets collide with the wall at any bend or any obstacle, and cool is via evaporation. Unfortunately, due to this effect, little of the spray arrives in the cylinder and mist.
During a rebuild with direct port jets on top of u-turn plenum runners. I could see the traces of the water stream on the front part of the combustion chamber and piston tops. These areas were light brown vs black.
Since moving the direct port jets on top of the plenum, but right behind the fuel injectors in the flat section, I needed to add more ignition advance. A clear sign that more mix effectively arrived in the combustion chamber as mist and not a fluid stream.

Once someone goes to significantly higher boost levels, the corresponding turbo exit temps of well above 100C or even 150C will evaporate much more fluid thrown at it. Riceracers setup is a good example. Condensation in the charge cooler will still be present, but to a lesser degree.

Overall my little 0.3mm jet used now drops the IATs a tad, but the main work is done by now 4x 0.4mm c-type (150 ml/min) direct port jets.

Originally Posted by rotrex (Post 22061)

don't think so. I gain power with every degree advance I add.
We talk 14 deg of advance here with plenty of water/meth and 102 octane fuel on a 4 valve pent roof engine. Without water / meth I am stuck at about 10 degrees.

Why are you using racing fuel as base fuel? I use pump fuel 92 octane with the amounts of WM50 I mentioned earlie theoritically rated at 112-117octane on boost.

I map myself. I know how power develops as I advance ignition.
I also have a J&S safeguard knock control system with a display. Beside this system, knock is also easily audible in a Elise. With both I can map it right to the knock limit myself and observe changes by butt and by data logging.
Power changes are also easy to feel as the car weights less than 800kg.

I tune myself aswell and I can confirm that I have the same situation you have regarding advance and power. The "more" advance you feed it the more it pushes hard, at least 20deg over "stock" seems insanely too much. Also I run it insanely lean on boost, average of 15AFR from 1-4psi, Low 14AFR from 4-16psi, High 13AFR from 17-21psi above 6Krpm

Curious what ecu are you using (mine is Haltech PS1000) and if you mind throwing a picture of the plug burn?


I have build the engine with a 1mm copper shim to protect the cylinder ahead against indentation. Rover K heads tend to turn soft over time in high power mid engined applications, especially around the exhaust valve area. This is caused by insufficient coolant flow at idle and low revs after a high load situation.The MLS head gasket seals with a very sharp edge against the hardened alloy cylinder head.
I installed a electric water pump and controller to manage the heat in low rev and idle situations. Works a treat.

Thing with the shim is that I forgot to chamfer the lower edge of the cylinder bore of the shim. I had specified the shim with a inner bore diameter of 80mm, same a cylinder bore diameter to distribute forces uniformly. the firer ring has a diameter of about 83mm. Normally the shim also has a bore of 83mm. This puts the lower edge of the shim in the quench zone near the head gasket seal area.
I believe this exposed edge together with the reduced squish is responsible for the tendency to knock.
To fix this I need to replace the head gasket with a N-series head gasket, it employs a flat seal area with a larger surface, and remove the shim. that's it. Compression will still be a reasonable 9.2:1.
It is a full day of work. I just don't have the time or motivation to do it in the moment as the car is plenty fast as is.

https://youtu.be/onYUHH0muNc?t=25
To put it in perspective: The yellow Elise in front of me is powered by a supercharged Honda K20A2 engine from a Civic Type R with 280HP. On the Radiation straight I just left in it 5th gear.

Originally Posted by rotrex (Post 22065)

In Germany we use RON ratings for fuel. MOst stations sell 98 RON (94 US) Octane as the highest, but ARAL/BP sell 102 Octane fuel almost everywhere for 6 cents extra a liter.
102 Octane RON corresponds to a US 98. I use no race fuel. It is just a special pump fuel. Shells offers a 100 Octane, too It is regular pump stuff over here.
Stock timing of this engine naturally aspirated is more in the range of 30 to 32 deg ignition advance.
The ECU I use is a Emerald K3 from the UK connected to a STACK wideband lambda meter.
I run closed loop fueling at all times, even under boost.

I have to look for plug pictures, but they have more of a "tan" compared to yours.
Tried. An AFR of 13 with 0.6 bars of boost. Worked well, but increased thermal load on track.At 0.8 bar running leaner just makes it more prone to knock and I have to pull timing and I lose power.

Originally Posted by Flr Power (Post 22068)

parmas, Why do you run it so lean?
You are missing some power by running it so lean...

Originally Posted by Flr Power (Post 22070)

Are you sure you're Wideband is accurate?

I have two wideband sensors. One wired directly to the haltech ecu and another independent AEM wideband gauge. Both showing same readings.

How did you measure the power change with different AFR?

I do not have a dyno neither have the money to tune it on a dyno for every upgrade I made to the setup. I tune on the mainly on street and adjust for the grip on track.

I measure power by time vs speed on 4th gear.

Full throttle from 2Krpm and check how much time it got to same rev limiter or 140km.

Also power and torque is easily felt on a small car like a yaris

Originally Posted by rotrex (Post 22071)

Are you sure your reference position for ignition timing is right?



Any modern engine won't take 40 deg timing at anything but low load.
Typical numbers for NA engines past 3000 rpm and WOT are 30 to 34 deg.
Add boost and numbers drop from there.
The methanol I use is fuel for fuel cells and of very high purity. Fuel is from a major oil company (BP), sold in their stations from the pump and should not vary that much. It is used as the standard fuel in the DTM racing series in Germay. I won't test it. Not worth the trouble and the expense. It is not sold in barrels in a shop.

The timing numbers I run are not that uncommon. The methanol water injection allows for an extra 5 deg of timing. Part of this advance is to compensate for the slower burn speed.
A similar engine, but with a turbo without meth water, runs 14 deg of advance at 15 PSI of boost without water.

Originally Posted by Flr Power (Post 22074)

parmas, Did you ever check your timing for drift at high rpm with a timing light?

I checked it for very few seconds on 4Krpm

Also I just realized you said you inject at 60% water/meth fuel ratio. I have only tuned to about 30% ratio so you are in a different league.

You are right 60% WM50. The injection was gradually increased by trial and error beginning from as small as 10% until 60% which took about 2 years testing and tuning.

I am sure the mixture could handle more WM50 but there are issues I have to overcome first. I have two nozzles 350cc. The pre-turbo is set to inject later than the post turbo by installing a pressure relief valve greater than the post turbo one. Still since I inject very early the engine bogs a bit at low rpm / low boost and high gear.
Reducing injection will make the air temperature fire up accordingly.


I remember when you used to spray very little and now you spray a lot. Good work by the way!