Pre Turbo Injection Recorded Temps

[Gelf]

I borrowed a dataharvest from work this weekend, ive recorded over a hours worth of data.

I used four thermocouples, one before the WI point, one before the IC, one after the IC and one at the manifold.

I started with a 0.4 jet and worked up to a 0.9 jet. I noted the time on a stop watch after each jet change and drove up to the top of a long steep hill, which allowed me to hold boost for a considerable amount of time.

Finally i did the last run with WI off.

Should be able to post the results on monday :smile:

I want to find out the optimum jet size for cooling the inlet charge.

Ive noted that the turbo was a lot cooler (quieter, not hissing) when i pulled over to ghange jets from 0.6 upwards, also the post IC temps were higher than the post turbo temps at these times. ( So i kept the engine running while changing jets from that point onwards).

[Richard L]

Pity the weather not as hot as one like but at least it wil be consistant.

I look forward to your results. Look out for nails :wink:

[Gelf]

Ive testing been driving with the spare on, not ideal really, but wouldnt effect inlet temps :smile:

[Gelf]

Pity the weather not as hot as one like but at least it wil be consistant.


I'll repeat the exercise next summer, then i'll now whichs jets to use in summer and winter.

[Gelf]

Richard,

im not sure how to post the excel file ?????????

[JohnA]

Richard,

im not sure how to post the excel file ?????????

If it doesn't work here mate, just email it to me and I'll host it on my site.
Send it to "john@max-boost.co.uk"

[Gelf]

Im going to repeat the exercise again this weekend, the temperature was being sampled every 15 seconds, my work colleague will set up the equipment up to sample once every second for this weekend.

I will still post the first set of results as soon as i can.

JohnA: I'll email you this coming weekends results, which should be more comprehensive.

[PuntoRex]

No matter it's a table or graphic, you may cut & paste onto a PowerPoint template. By that, you can "save as" a .jpg file, that would be much more convenient on this forum.

[Gelf]

No matter it's a table or graphic, you may cut & paste onto a PowerPoint template. By that, you can "save as" a .jpg file, that would be much more convenient on this forum.

Done it, 200k jpeg.
http://www.imagehdd.com/d1y2004/9473Slide1.jpg

Interesting that temps exiting the IC have cooled to below ambient by the time they reach the throttle plate. Evapouration is continuing down the 20" pipe from the IC to the throttle. Thats after the water travelled 10" to and through the turbo, and 18" to and through the IC.

[Richard L]

WOW, that is one 1.5 hours of full boost :twisted: I think the magenta peaks are the on-boost period?

These datas are so good, thumbs up for posting the chart.

Please remind me where all the jets are or it is only one in fornt of the turbo?

Awesome work!!! :D

Richard

[Gelf]

This is just the pre turbo jet in use, i think i need to move this jet much further back from the turbo and the throttle jet back to just after the IC and move the inlet temp sensor that sits on the exit from the IC (where recorded temps are much higher than at the throttle plate) to just before the throttle plate.

Should have a much better graph next week for temps sampled more than once a second. I will also lower the trigger point to 10 psi so WI comes on earlier.

[Richard L]

Observing the post IC temperature, iy appears that the larger the jet, the air is hotter after the IC. :?:

[Gelf]

Observing the post IC temperature, iy appears that the larger the jet, the air is hotter after the IC. :?:

Could be the IC suffering from heat saturation from boost conditions before WI triggers (13 psi, too high now), once triggered the cooler air now leaving the turbo is being heated by the hotter IC.

Best wait and see what the graph looks like when temps are samples at one or more times per second. Temp spikes could be much higher and vice verser( troughs much lower) At one point 0 degrees Celcius was recorded at the throttle.

What is the ideal inlet temp we are looking for :?:

What is the effect of air temps reaching the cylinders 10 to 15 degrees C less than the ecu thinks :?:

[Gelf]

Ok, cant sleep for thinkin about this.

Ive got two 2d systems, one pre turbo and one at the manifold (which looks like its not needed).

One is wired into cylinder 2 and the other cylinder 3 injection pulse.

I propose that i place both jets together before the turbo but another 12 or so inches further back. Use matching jet sizes for each (i.e two 0.5's or two 0.6's), they will inject alernately giving a move even flow into the inlet charge and give a finer spray, both desirable.

Finally move the ecu inlet pressure combined temp sensor to just before the throttle to allow maximum temperature reduction to be recognised by the ecu.

Added advantage if one pump failed or had a blocked jet, the other would still be injecting reducing the chance of sudden detonation.

Did i mention that my ecu uses independant cylinder knock detection :D

And maybe, like Bill Shurvinton said, remove the IC altogether...........

[PuntoRex]

Is there any subjective difference in driving with these jets?
And the difference between the pre-throttle injection?

[marck_c]

hello all I'd like to add my input questions to the discussion.

I run a circuit racing mitsubishi lancer evolution here in Jamaica. The car is pretty much a dedicated race car. The engine is fully built (rods, pistons, headwork, bigger valves, stainless steel exhaust manifold, hybrid Garrett turbo etc, etc etc. This engine has dynoed at 420whp/330 ftlb tq at 1.5 Bar. I've tried experimenting with water injection and water spray and logged via the Haltech datalogger the following results:

with 2 aquamist .8 jets; one spraying at about 3" downstream of the intercooler, and the other about 3" ahead of the throttle plate, and another two .8mm jets spraying the intercooler core; the intake temperature fell by about 45deg F. From 140F to 95F. These temps were measured by a sensor on the intake manifold. The ambient air was about 85F and there was a smallish fan blowing on the I/C. With this combination the power fell to 395 whp. There were however no adjustments made to the timing or fuelling. There was a slight bog at high rpm when the water was injected ( I think I may have sprayed too muc water).

I tried spraying on the I/C alone and noted no appreciable temperature drop.

At the race track, with much more available airfolw on the I/C the water spray alone was tried and the follwing results noted:
After about 10 minutes of continuos high boost operation, the intake temperature stabilized at about 155F (90F ambient) Without the I/C spray the temperature would normally be about 170F. I think the water spray helps with the I/C heat soak for prolonged operation, but can't act quickly enough for a 10 second dyno pull.

[marck_c]

I am planning to experiment next with just one .8 jet into the intake manifold. triggered by an aquamist pressure switch (set at about 15 ps)i. and a haltech GPO set to cycle between 38C and 42C (targetting 40C). I would also like to experiment with a pre turbocharger jet triggered once the engine goes from vacuum to boost. The plan is to increase the compressor efficiency, and reduce the turbo lag. I'm shooting for a constant 40C intake temperature. Once I can maintain that I'll play with the fuel and timing to build more power.

Anyone have comments or suggestions?

[JohnA]

very interesting chart! :D


Two things stand out in my opinion:
1. the largest nozzle produced post-ic temps the same as no WI! The smallest nozzle had the best results here.

2. as nozzle size goes up, the difference between pre and post-ic temps is squeezed. I believe this to be because we are compromising the ic efficiency (which is max when temp difference is max)

I look forward to hearing the comments of others, as we are a bit in unchartered waters here.

[Gelf mate, if you have a website, can you give me the address, so I can reference it from my Water Injection page.]

[PuntoRex]

The efficiency of IC drops, seems a bad thing....

But wait, on the other viewpoint, it means IC could be eliminated. (maybe a good thing, eliminating pressure drop & long routing)

Moreover, post-IC temp can not explain the efficiency shift of the compressor itself.

Averagely, pre-IC (i.e. compressor outlet) temp actually drops as the jets get bigger. So we might roughly say the compressor efficiency benefits somewhat by more water (within the range of this experiment).

(However, do we really need the final temp below ambience is another story)

As Larry (Hotrod) mentioned in another thread, 2~3% injection rate may be ideal for the pre-compressor injection. Maybe someone can help on this in this experiment.

In addition, I'd really like to see MAF info along with the pre-compressor injection, if possible.

[TurboGTi]

I am planning to experiment next with just one .8 jet into the intake manifold. triggered by an aquamist pressure switch (set at about 15 ps)i. and a haltech GPO set to cycle between 38C and 42C (targetting 40C). I would also like to experiment with a pre turbocharger jet triggered once the engine goes from vacuum to boost. The plan is to increase the compressor efficiency, and reduce the turbo lag. I'm shooting for a constant 40C intake temperature. Once I can maintain that I'll play with the fuel and timing to build more power.

Anyone have comments or suggestions?

Hey Marck_c good to see another jamaican on this site, Ok re the above test that you want to do with the .8 jet, Have you tried to inject methanol? and lean your fuel while the methanol is being injected?

I hear that this will result in some big HP figures ....

I too want to do some datalogging to find out what my engine is doing, but as stated in anothere post i'll document this at a later stage.

[Richard L]

John A,

I can't help making a comment about your compressor wheel (avator), isn't it about time you change the bearing? one day it will fly off the edge of the screen, I hope it won't be mine.

[Gelf]

JohnA: Sorry no web site, but the data i collect will be freely available for anyone to use.

PuntoRex: Dont have any means to record ecu data at present, but i am cosidering

http://www.espautomotive.com/trolleyed/index.htm

[PuntoRex]

Gelf,

As previous query, did you "feel" any difference among all these jets?

[Gelf]

Gelf,

As previous query, did you "feel" any difference among all these jets?

Sorry for not responding :oops:

I can only say that i percieved drop in performance with all jets, no matter what size (at WOT). At say 3/4 throttle im sure i expierenced increased performance with the 0.6 and 0.7 jets.

Not conclusive im afraid :?

[Gelf]

Something else the graph showed, the cold air pickup was drawing hot air from the engine bay on occasions. Some investigation needed there :sad:

[PuntoRex]

...
I can only say that i percieved drop in performance with all jets, no matter what size (at WOT). At say 3/4 throttle im sure i expierenced increased performance with the 0.6 and 0.7 jets.

Not conclusive im afraid :?

That's totally unexpected!! :shock:

With pre-comp injection only, bogging down shouldn't happen, at least on those smaller jets.

Would it be "too cold" of the incoming air? At 5~15C, not quite, I guess.

It's odd!

If we can see the A/F info might helps.

-------------

Once I made the fuel pressure way too high, the acceleration force declined obviously. It seemed the energy produced by the engine was held back by itself, like holding fists tightly with popped-up veins but doing nothing.

Another case was injecting too much water in cold winter, it seemed the fire was put out somewhat. Bogging down was easily sensible. It felt weakened, diluted, emptied. The original feeling of solid strength was gone.

Net effects seem similar, but the "feeling" are different.

[Gelf]

At WOT, when WI activates, i can see the needle on the boost gauge drop substancially :?

[PuntoRex]

Supposedly there would be an on-board air temp sensor somewhere on the intake track or manifold which affects ECU's action on fueling/igniction (even boost maybe).

Is your boost control actuated directly by ECU? Or you use a third party controller?

[Gelf]

Supposedly there would be an on-board air temp sensor somewhere on the intake track or manifold which affects ECU's action on fueling/igniction (even boost maybe).

Is your boost control actuated directly by ECU? Or you use a third party controller?

You can see the sensor in this pic, where the IC connects to the blue pipe.

http://www.imagehdd.com/d1y2004/4025IM001159.jpg

Yes, boost is contolled directly by the ecu.

[JohnA]

...I can't help making a comment about your compressor wheel (avator), isn't it about time you change the bearing? one day it will fly off the edge of the screen, I hope it won't be mine.
It's perfectly balanced, Richard, freshly reconditioned. :lol:
Apparently it's the shadows that move around, giving the impression of imbalance. I'll see if I can fix this, but it's purely cosmetic. It holds boost fine!

[Richard L]

...I can't help making a comment about your compressor wheel (avator), isn't it about time you change the bearing? one day it will fly off the edge of the screen, I hope it won't be mine.
It's perfectly balanced, Richard, freshly reconditioned. :lol:
Apparently it's the shadows that move around, giving the impression of imbalance. I'll see if I can fix this, but it's purely cosmetic. It holds boost fine!

It appeares that the world is spinning around when you focus on the shadow - perhaps it did - otherwise the wheel is perfectly balanced, as long as you are make good boost, that is all that matters.

[Richard L]

...
I can only say that i percieved drop in performance with all jets, no matter what size (at WOT). At say 3/4 throttle im sure i expierenced increased performance with the 0.6 and 0.7 jets.

Not conclusive im afraid :?

That's totally unexpected!! :shock:

With pre-comp injection only, bogging down shouldn't happen, at least on those smaller jets.

Would it be "too cold" of the incoming air? At 5~15C, not quite, I guess.

It's odd!

If we can see the A/F info might helps.



I just remember the days when I tune cars ( too long to remember when) without and a/f ratio meter lambda probes. The throttle is a good indicator on a/f ratio.

If you are experiencing less power at WOT them 3/4 throttle, it could be your fuel supply is running out. You can also test it by applying full throttle first and slowing easing the paddle out and you will experience a surge of power starting to develop. If this is the case, you should consider incease you fuel pressure temporarily and see if it makes any difference.

Another common cause is the airflow meter maxing out, you will notice that almost all injectors runs out of duty at 4000-5000 rpm - this is just a electro-mechanical contraint - the opening and shutting time of the fuel injector dedcates.

The System 2d has a fuel injector duty cyclee monitor circuitry in-build. You can connect a volt meter (analogue type is easier to read) to the purple wire and see how wide the injector are opening eniminate the "airflow sensor" maxing-out theory. 0-5V = 0-100% duty cycle - 5% accurate but good enough for this purpose.

[JohnA]

..It appeares that the world is spinning around when you focus on the shadow - perhaps it did - otherwise the wheel is perfectly balanced, as long as you are make good boost, that is all that matters.
I've edit it a bit and it's slightly better.
Still some shadow spinning, but boy does it boost faster now. Less inertia.

[Richard L]

Sorry, I like the other one better :lol:

[TurboGTi]

...I can't help making a comment about your compressor wheel (avator), isn't it about time you change the bearing? one day it will fly off the edge of the screen, I hope it won't be mine.
It's perfectly balanced, Richard, freshly reconditioned. :lol:
Apparently it's the shadows that move around, giving the impression of imbalance. I'll see if I can fix this, but it's purely cosmetic. It holds boost fine!

It appeares that the world is spinning around when you focus on the shadow - perhaps it did - otherwise the wheel is perfectly balanced, as long as you are make good boost, that is all that matters.


Talk about OFF topic :?

:D

[PuntoRex]

....

The System 2d has a fuel injector duty cyclee monitor circuitry in-build. You can connect a volt meter (analogue type is easier to read) to the purple wire and see how wide the injector are opening eniminate the "airflow sensor" maxing-out theory. 0-5V = 0-100% duty cycle - 5% accurate but good enough for this purpose.

I've done that already.

It frequently rushes to 80% DC when I rev it to the red line. I even saw 90% once or twice :shock:

I guess the poor little motor has been strained for all these years. Maybe I'll need bigger fuel injector for the coming pre-comp injection.

[PuntoRex]

....

Yes, boost is contolled directly by the ecu.


I suspect the boost control portion in the ECU is affetced by the temp senser. Maybe I'm totally wrong. But nowadays electronic controls on cars are really complex.

When you're on 3/4 throttle, where you sensed gain, was the boost any different from how it was before?

(on my modified manual boost control, I can get 90% boost on part throttle)

[Gelf]

....

Yes, boost is contolled directly by the ecu.


I suspect the boost control portion in the ECU is affetced by the temp senser. Maybe I'm totally wrong. But nowadays electronic controls on cars are really complex.

When you're on 3/4 throttle, where you sensed gain, was the boost any different from how it was before?

(on my modified manual boost control, I can get 90% boost on part throttle)

Yesterday while i was recorded temp data, i observed that i could hold 15 psi at 3/4 throttle, but if do the same at WOT, im only seeing 10 or 11 psi. Ive moved the jets as far back from the turbo, now i need to move the temp sensor as close to the throttle plate as possable to see if that helps. (Water vapour effecting the sensor ?) Ive always felt power was down at WOT since i had it remapped (generic) going back for a full RR remap very soon, FOC too to correct this issue.

Richard said i could be running lean at WOT, i will get hold of volt meter and check the IDC. I thought leaning fuel was the idea though, because of over fueling.

I only experienced bogging (hesitation) yesterday using twin 0.9 jetss :shock:, i know thats alot of water, but i thought id go all the way to maximum injection while i had the recording equipment running.

With twin 0.8 jets, the car really pulled in 5th gear above 4K rpm, (at 3/4 throttle opening)

[hotrod]

Keep in mind what your discribing could be due to the fact the turbo has reached its maximum sustainable flow. If the engine could use 400 cfm of air at 20 psi, and the turbo can only deliver 360 cfm of air at 20 psi your going to end up with a manifold pressure equal to the point where the turbo can deliver some value in between. For example if the turbo hits max flow of 380 cfm at 17 psi and the the engine wants 380 cfm at 17 psi, that is what the boost will drop to.


On the stock WRX turbo boost taper at high rpm is mostly a function of this situation where the engine needs more air than the turbo can deliver.

Larry

[PuntoRex]

I got an idea on the injection method :idea:
(or I'm not sure has anyone proposed this before? )


example:

main: 1* 0.6mm jet at pre-throttle
aux1: 1* 0.5mm jet at pre-throttle
aux2: 1* 0.5mm jet at pre-compressor


Stage1:
Main & aux1 are triggered by boost simultaneously.

Stage2:
Up to the pre-set rpm (or other conditions), aux1 switches to aux2, while main jet still dose its job continuously.


Reasons & tips:

1. At the fast-rising of boost in the midrange, engine needs water the most for det control. Full cooling capacity is used on this.

2. As the rpm up to a point, huge mass air flow makes compressor efficiency suffer, part of the water is switched onto here to give helps.

3. Total water flow vs air & fuel maintain within a certain range.

4. Jet sizes can be widely variable. Different effects & preferences can be played around.

5. RPM switch can co-work with other trigger, e.g., temp senser or anything else.


Minimum h/w requirement:

1. Aquamist 2D system with 3 jets
2. 1-in/2-out switchable splitter (or 2 solenoids? )
3. RPM switch
4. Others as you like....


Seems insteresting.
Any thoughts?

:razz:

[hotrod]

Yes that is a good concept. I have alluded to that sort of setup on a couple of occasions here.

I will have something very similar when I get my system finished.
The multiple nozzles should give you more flexibility, and safety regarding nozzle blockage.

Larry

[masterp2]

I am moving toward the config P-rex suggests. Would like the option of doing more vaporizing, which can't be very effective at the throttle with so little time. The hi temps assocaiated with the turbo compression will solve that. 8% humidity here, need it bad.

[PuntoRex]

Been using the pre-compressor jet installation for a while, I did suffer from the surge problem in higher gears, especially 5th in the mid rev range on the highway. Tuning up the injection point (10psi to 12) helped some, but would suffer from the other problem - midrange det.

The weather was chilly these days, and my car det! It never act like this. I suspect that maybe the partially vaporized water from the compressor oulet condenses back to liquid within the cold IC, then the heat is back. And the liquid water in the thin IC passages is definitely not a good thing. While slightly later injection point provides insufficient in-cylinder cooling after all these chaos. It's just not right on time, especially in this fixed w/f 2D system.

So, pre-compressor WI should not be too early & too much. Especially with an upgraded compressor wheel or bigger turbo. I think this is for sure.

I managed to make a system which turns on pre-compressor jet a little later than the main (the pre-throttle) one, and one additional jet also on pre-throttle for slightly larger w/f ratio at the first injection stage. (as above)

Using the 2D's injection duty cycle voltage output as the switch reference, i.e. 3V as 60% IDC. I think this is a good point for my car which rushes to 80% at red line.

By a comparator circuit, I compare the IDC related voltage with a pre-set 3V reference, then drive a 3way/2position solenoid valve to guide the flow.

There're 3 jets:
A- main jet, running full time, pre-throttle
B- mid-load jet, pre-throttle
C- high-load jet, pre-compressor

My car is running at 20psi boost max. At the 1st stage, begins at 8psi (about 40% IDC on my car), A+B are working, for maximum in-cylinder cooling. 2nd stage, begins at 60% IDC, A+C are working, for improved compressor efficiency & less det control needs.

By doing this, it avoids the surge line on the midrange operation & stretched the choke line on the hi-load/hi-flow situation. Effectively, it extends the overall area of the compressor map, I believe.

I just got it done today (2:30 am). It's running very well in the first impression. Earlier injection point makes the car livelier on part throttle acceleration & eliminates the midrange det. 5th on highway speed is now smoother on small throttle inputs. WOT to the red stays perfect as the previous full time pre-compressor jet setting. It's running in a stretched compressor map. What a treat!

It's a good mod, worths a try.

Next, I'll try fine-tuning the stage 1-2 switch point & several different jet sizes to play with a two stage w/f ratio map....

[Steve VXR]

I know this is digging up a quiet topic, but I was just wondering what spec the Z20LET is running in your car please Gelf? I've just fitted a 1S system to a a Z20LET and that is running a hybrid turbo and about 24 PSI peak boost.

[Gelf]

I know this is digging up a quiet topic, but I was just wondering what spec the Z20LET is running in your car please Gelf? I've just fitted a 1S system to a a Z20LET and that is running a hybrid turbo and about 24 PSI peak boost.

Hi Steve, standard turbo and exhaust, remapped with 19 psi peak boost

[Richard L]

No matter it's a table or graphic, you may cut & paste onto a PowerPoint template. By that, you can "save as" a .jpg file, that would be much more convenient on this forum.

Done it, 200k jpeg.
http://www.imagehdd.com/d1y2004/9473Slide1.jpg

Interesting that temps exiting the IC have cooled to below ambient by the time they reach the throttle plate. Evapouration is continuing down the 20" pipe from the IC to the throttle. Thats after the water travelled 10" to and through the turbo, and 18" to and through the IC.

Gelf, what happen to the image? I was jsut going to take another careful look.

[ctischmick]

No matter it's a table or graphic, you may cut & paste onto a PowerPoint template. By that, you can "save as" a .jpg file, that would be much more convenient on this forum.

Done it, 200k jpeg.
http://www.imagehdd.com/d1y2004/9473Slide1.jpg

Interesting that temps exiting the IC have cooled to below ambient by the time they reach the throttle plate. Evapouration is continuing down the 20" pipe from the IC to the throttle. Thats after the water travelled 10" to and through the turbo, and 18" to and through the IC.

Gelf, what happen to the image? I was jsut going to take another careful look.
If we can't get the pic up, you should unsticky this thread.. its kind of useless.

[Richard L]

Thread is un-sticked.

[Tjabo]

No one copied the image(s) from the beginning of this thread? It seems like there's some good information here, but it's hard to make sense of it without the pictures. . . Does that mean I'm illiterate? :lol: