waterinjection.info  

Go Back   waterinjection.info > Injection Applications (making it work) > Gasoline Forced-Induction

Reply
 
Thread Tools Display Modes
  #1  
Old 21-09-2016, 04:21 AM
sambeeb sambeeb is offline
Junior Member
 
Join Date: Jan 2016
Location: sydney
Posts: 24
Default WI causing knock?

Hi,

I have a vw polo gti 9N3 with the audi 1.8T 20v turbo. It runs 1.4bar max, and a 0.4 "A" post intercooler/pre throttle body boost triggered (@0.85bar)nozzle.
I have been running generic flash tunes in my car but have facility to make changes to global timing changes across the timing map. In the past when setting up the water injection I would increase timing bit by bit until I could see timing pull generated by the knock sensors. I would stop there with my timing adjustments and then I would only continue on from there with the water injection active.
I recently changed to an updated flash tune and have tried to set up my timing in the same way. However I'm finding that when I run the water injection, the timing pull observed is actually increasing!
I have removed, cleaned and air tested the nozzle aok. I have moved the trigger point up from 0.85 bar to over 1 bar thinking that quenching might be at play and lower in case I didn't have enough early enough. I have gone slightly higher and lower around that critical timing point. This tune doesn't make more power than the last ie I'm not under watered compared to the last tune. However when I run with water injection I get slightly more timing pull!
My guess is that this tune is not fuel limited (its at 0.83'ish lambda) and so its going to be very hard to get any benefit from the water when I can't adjust fuel, but I'm struggling to understand what conditions would be at play for water injection to be not only not getting rid of the timing pull but enhancing it? any ideas

thanks
sam

Last edited by sambeeb; 21-09-2016 at 04:27 AM.
Reply With Quote
  #2  
Old 21-09-2016, 02:05 PM
rotrex rotrex is offline
Senior Member
 
Join Date: Jan 2014
Location: Germany
Posts: 187
Default Re: WI causing knock?

at that power level, those jets are rather small anyhow.
usually injecting pure water slows flame front speed and increases the ignition delay.
it has the same effect as retarded ignition timing.
As the water has to take steep turns though the intake plenum you will have rather uneven cylinder distribution.
Still this would not cause increased knock with water.
I could envision that the new "tune" applies more ignition advance than the old one. Timing is only pulled once the knock count exceeds a certain level. you might start closer to this knock threshold with the new tune compared to the old one and thus exceed it earlier causing the ECU to pull timing.

What about mounting 4x 0.4mm C jets like I use right before the main fuel injectors. it is tremendously effective. then run water/methanol, just feeling and increase boost to 1.6bar :-).
From my limited 1.8T experience, they take 1.4bar and a tad more even without any water with a bigger T4 turbo.
Reply With Quote
  #3  
Old 22-09-2016, 01:08 AM
sambeeb sambeeb is offline
Junior Member
 
Join Date: Jan 2016
Location: sydney
Posts: 24
Default Re: WI causing knock?

I guess I should get it out there that the primary reason that I installed water injection in the first place was to mange high inlet temps. That is why I sited the nozzle just after the intercooler so as to give it the maximum time to evaporate and do its thing and hopefully all be in suspension by the time it got into the plenum so that cylinder distribution would be good. The flash tunes I've run are always fuelled up and I didn't think they were ever running lean enough to be able to run a meth mix which introduces a second fuel.
Its very possible that I may have buggered up my calculations but at 230hp/320nm, I worked out that the 0.4"A" at 80psi was flowing the right amount of water at peak torque. I was always concerned that as boost trailed off towards redline and injector duty cycle dropped off that I would then be running too much water up high, but on the last tune I never lost power doing this. But like I said maybe I had under rated it slightly.
After a bit more investigation I'm also seeing a slight 'flat spot' around 5000rpm where the rate of air flow slows a little when I'm not running water. Then when I run water even with the timing advanced, this flat spot turns into an actual dip where the air I'm flowing g/s through the MAF will actually decline for a few hundred rpm, even though no timing has been pulled there. This leads me to think that at the fuelling the tune is running, that like you say, the water is slowing the flame front and decreasing the cylinder pressures too much.
Your last comment regarding the boost and cylinder pressures a 1.8T can take without needing water leads me to think that water is just not going to work on this tune unless the boost is increased. Fuelling is fixed, and it seems timing limited. 1.4bar sounds like a lot but this is being pushed by a very small borg warner KO3s turbo that isn't moving a lot of air in real terms so perhaps that's where i'll look next. A tad more preload on the wastegate should achieve that - I only have 2mm at the moment.
The only other alternative I might have is to switch the water on at say 1.2 bar. That way the water will kick in approaching peak torque and then switch out soon after because with my turbo the boost trails off quite quickly towards redline being so small. Then I might not kill off the top end with too much water.
Reply With Quote
  #4  
Old 22-09-2016, 07:19 AM
Richard L Richard L is offline
Manufacturer sponsor
 
Join Date: Oct 2003
Location: England
Posts: 4,936
Default Re: WI causing knock?

Concur with nozzle being too small. Especially the 0.4mm (A) will not work well at 80psi. You need 140psi+. Smaller nozzle requires higher pressure to atomise properly.

What system do you have?
__________________
Richard L
aquamist technical support
Reply With Quote
  #5  
Old 22-09-2016, 11:26 AM
sambeeb sambeeb is offline
Junior Member
 
Join Date: Jan 2016
Location: sydney
Posts: 24
Default Re: WI causing knock?

I'm running a 100psi ebay diaphragm pump. 5L/min 4.5A. I have a bypass line in place with an adjustable valve so that I can adjust pressure down from there. I was running it at 80'ish psi because I was worried I was flowing a tad too much water at the rated 100psi. By all means i'll put it back to 100psi if you think it'll help atomisation. I'm using a simple industrial pressure switch to activate a fixed flow which is controlled by a solenoid and the nozzle is a check valve aquamist 04 "A".
Reply With Quote
  #6  
Old 22-09-2016, 12:36 PM
rotrex rotrex is offline
Senior Member
 
Join Date: Jan 2014
Location: Germany
Posts: 187
Default Re: WI causing knock?

at this power level, you can use 300-500 ml/min, if you can fully map feeling and ignition.
Richard's rule of thumb of 1.5 ml/min per HP gets you an idea.
If your ignition system copes and you are knock limited and still cannot hit best timing, you can significantly exceed that flow rate.
Just look for some of the recent threads here.

running pure water limits you if you cannot properly retune. Water likes slightly leaner AFRs and more ignition timing. Water/methanol requires less timing change, still a fair bit, but needs feeling reduced to maintain AFRs. Mine likes 11.8 to 12.5 the best on water/methanol 50:50 by weight running 4x 0.4mm C Aquamist jets, direct port, at 8 bar pump pressure.

in the past I tried 13.5:1. While it gave excellent power and reduced fuel consumption on track, coolant temps were slightly higher, albeit not critically high. I did not fully optimised ignition timing with that mix. Could be I was not advancing ignition enough leading to higher coolant temps.


in "Pre Turbo Injection - How Much and When to Inject" fellow PARMAS injects 60% 50:50 mix to fuel in his nonchargecooled turbocharged Honda.

Last edited by rotrex; 22-09-2016 at 12:38 PM.
Reply With Quote
  #7  
Old 22-09-2016, 03:32 PM
sambeeb sambeeb is offline
Junior Member
 
Join Date: Jan 2016
Location: sydney
Posts: 24
Default Re: WI causing knock?

So since I can't adjust fuelling then meth mixes are pretty much ruled out or I'm going to be running really rich. It looks like using water to make substantial power gains is off the cards too because again I can't lean it out and it isn't accepting more timing.
Is it going to at least be possible to inject a small enough quantity of water that will give a significant amount of charge cooling, yet not be so great as to rob the engine of power?
Reply With Quote
  #8  
Old 22-09-2016, 04:21 PM
rotrex rotrex is offline
Senior Member
 
Join Date: Jan 2014
Location: Germany
Posts: 187
Default Re: WI causing knock?

in honest opinion and experience, spraying water though the turns of a plenum leads to streams of water running along the walls with little effect on combustion., hence it wonÄt accept more timing.
Seriously, try direct port injection with the jets located in the final straight section of the runners on the top side near the fuel injectors.
For the same flow, you get way more effect and it will accept a lot more timing.
But as you can only globally add timing, you will have too much advance at part throttle.
I would also seriously consider programable ECU. There are plenty of options for VAG ECUs to be remapped.
Reply With Quote
  #9  
Old 22-09-2016, 11:17 PM
sambeeb sambeeb is offline
Junior Member
 
Join Date: Jan 2016
Location: sydney
Posts: 24
Default Re: WI causing knock?

dam inflexible flash tunes! I think I might just have to use it for hillclimb events when its baking in summer (Oz) and inlet temps get risky. The car is still running in a road registered over 2.5L class. If I ever take it off the road then yes, I will be going programmable.
It interesting what you said regarding the water distribution. I always figured that the droplets would just evaporate and with the air speed at the time I never expected that water could come out of suspension at that temperature and wet down the walls. However having said that, I do notice that there is more commonly timing pull on cylinder 4 (my plenum is fed from the cyl 1 side). That could point to what you're referring to I suppose.
Reply With Quote
  #10  
Old 23-09-2016, 04:17 PM
rotrex rotrex is offline
Senior Member
 
Join Date: Jan 2014
Location: Germany
Posts: 187
Default Re: WI causing knock?

Methanol quickly evaporates, the water takes much longer. Especially as bigger droplets hits the walls of the plenum in the turns. Just for fun, calculate the g-forces in a turn of a plenum at full chat.
folks running water/meth mixes benefit from the methanol evaporating first and this generating finer droplets of water after the meth has evaporated.
very fine water mist makes it to the cylinders, hence why it can make sense to inject in the air filter housing pre compressor.
the air cooling effect is anyhow not that big compared to what happens in the cylinder. The main effect of water droplets suspended in the cylinder volume happens after ignition as the flame front develops.

Here you see a picture of my pistons using direct port injection with the jets in the straight sections right before the fuel injectors. Spray hits the intake valves both at the front and the rear of the valve. The bigger droplets move straight down and hit the front and the rear of the pistons. the finer mist stays in the volume and supresses knock.




here is a picture of one of my old pistons. it was also used with direct port injection, but with jets on top of curved runners. Most of the fluid was centrifuged out of the air stream and entered the cylinder as a fluid stream with little spray left. on closer inspection, you could see where streams of water run across the piston.



Here is how flow in a cylinder looks like. this gets you an idea on how the blanc spots develop with proper direct port injection.



this one nicely shows the path the fuel takes if sprayed properly. Any big droplets do not make the turn and hit the front and rear piston crown. This perfectly matches the blank spots on my pistons.
Image the turns the spray has to take being injected after the IC. lots of tight turns removing droplets. This is IMHO also why folks have better success with higher methanol concentrations. it evaporates even as a film or stream and being a high octane fuel. water as a stream does very little. It passes straight through the engine into the exhaust.
Look at old pictures of the WRC Subaru using Aquamist 2c systems with direct port. they mounted the jets in the very last turn pointing straight down into the inlet port. They were only allowed to use 100% due to regulations. without that straight shot into the inlet port of the cylinder head, most water would do very little as a big percentage would have been centrifuged out against the walls of the runner.


Here is a photo showing the water rail for cylinders 2 and 4. Water injectors are mounted at the "knee" of each intake manifold "leg" and Subaru called this direct water injection.






you can get software to remap the OEM ECU you are using very much like the flash tunes. Just search for VAG software and your ECU model. As legal or illegal for your hillclimbs as the tunes you are using now. No aftermarket ECU required.

Last edited by rotrex; 24-09-2016 at 09:27 AM.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


All times are GMT +1. The time now is 12:41 PM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.