Water injection or Intercooler Spray?
- Thread starter Simon L
- Start date
-
- Tags
- injection intercooler spray water
I agree and I don't blame you, the 1S is not for me either, I just wonder that injector cycle mapping is plenty of control in the real world.
Agree that the 1S system isn't that great (Although results have been good) and I would prefer one of the 2X systems.
The 2S can be mapped to match your torque curve.
The 2D system maintains a water to fuel ratio using a injector duty cycle input signal. The injector duty cycle takes into account the revs, the boost pressure etc. As boost pressure increases the injector duty cycle must also increase to maintain the correct A/F ratio, else it'll go lean. Also, as the revs increase the time available for injection decreases and so the duty cycle increases (If measured as a percentage of available time).
Whilst a multi-dimensional mapping for water injection seems like a good idea on the face of it (eg. dimensions for boost, revs, charge temrperature, EGT, etc), you have to remember that you also have to map the ECU to take advantage of this. Mapping the ECU to always expect a constant flow rate of water after say 1bar boost must be easier than mapping the ECU to take into account the multiple variables that control the water flow rate.
IMO, I think its back to the law of diminishing returns. I think the simple boost triggered constant flow system (1S) gives a big advantage over no water injection, then a mapable system (2S) or a fuel/water ratio maintaining systems (2D) probably give a small improvement in results over the 1S system, and then a fully multi-dimensional mapped system probably gives a very small improvement over the 2D/2S systems.
Of course the above is only my opinion, and I have no evidense to support it.
I too, as a software engineer have considered developing my own system and I can think of some useful features that would be nice to add (eg. set an EGT threshold, and if that threshold is broken increase the flow rate. The more the threshold is broken the more you increase the flow rate. Could help prevent things melting). But to be honest, i think its probably more hastle than its worth.
Good luck to you though.
Cheers
Ben
The 2S can be mapped to match your torque curve.
The 2D system maintains a water to fuel ratio using a injector duty cycle input signal. The injector duty cycle takes into account the revs, the boost pressure etc. As boost pressure increases the injector duty cycle must also increase to maintain the correct A/F ratio, else it'll go lean. Also, as the revs increase the time available for injection decreases and so the duty cycle increases (If measured as a percentage of available time).
Whilst a multi-dimensional mapping for water injection seems like a good idea on the face of it (eg. dimensions for boost, revs, charge temrperature, EGT, etc), you have to remember that you also have to map the ECU to take advantage of this. Mapping the ECU to always expect a constant flow rate of water after say 1bar boost must be easier than mapping the ECU to take into account the multiple variables that control the water flow rate.
IMO, I think its back to the law of diminishing returns. I think the simple boost triggered constant flow system (1S) gives a big advantage over no water injection, then a mapable system (2S) or a fuel/water ratio maintaining systems (2D) probably give a small improvement in results over the 1S system, and then a fully multi-dimensional mapped system probably gives a very small improvement over the 2D/2S systems.
Of course the above is only my opinion, and I have no evidense to support it.
I too, as a software engineer have considered developing my own system and I can think of some useful features that would be nice to add (eg. set an EGT threshold, and if that threshold is broken increase the flow rate. The more the threshold is broken the more you increase the flow rate. Could help prevent things melting). But to be honest, i think its probably more hastle than its worth.
Good luck to you though.
Cheers
Ben
Fen said:
Yes, nossle blockage and running out of water can be terminal if the ECU mapping expects the water to be there. If the car was mapped without WI and the WI was added later then you should be ok (Although at the instant the water stops being injected, the ignition would probably be more advanced and you'd possibly have detonation until the knock sensor pulls it back.).
The 1S system has NO failsafes against a blockage or running out of water. The 2X systems do. This is one of the main reasons wy I'm not happy about having the 1S system in mine ATM.
The blockage/out-of-water signal from the 2D/2S systems can be used for a variety of things. You can simply get it to light and LED and/or warning buzzer, or you could do something more fancy such as use it to lower your maximum boost (This could be done if you have the Jabba boost controller by having 2 circuits, one that uses the Jabba controller and the other that has a resistor giving the same resistance as the Jabba controller does when set to min. The warning singal could trigger a relay that switches from the Jabba controller to the fixed resistor circuit.).
As for how much water you need.... I have a 45 litre fuel tank and a 6.6 litre water/methanol reservour (Washer bottle). I'm using the largest size nossle to give the largest flow rate. I probably have to fill the washer bottle once for every 5 tanks of fuel... and thats mostly due to washing my windscreen!
Max flow rate should be approx 10% water, so thats 4.5 litre of water for a 45litre tank of fuel.... but remember that the Aquamist is only active on my car at over 1bar boost, and in the real world thats not long at all. Even on track thats probably less than 50% of the time, so thats 2-3 tanks of fuel for each refill of water/methanol.
Cheers
Ben
ryan_s3 said:
I have the 1S. I REALLY wanted the 2D or 2S but Jabba tld me I had to have the 1S or nothing as they didn't have time to fit/map the 2X systems back then.
They also told me that Mike (Jabba) is working on his own controller, which can be fitted to the 1S system at a later date. No disrespect to Mike, but I'm not that confident that the project will ever see the light of day, simply because there are not enough hours in the day for him to do every project he wants to. For example, Mike originally told me that he was making a new version of their boost controller that would allow you to run different boost limit in 1st and 2nd gear... but when I spoke to them a year or so later they said that they were no longer going to do that (Although I'm thinking about doing it myself).
They also said that my 1S system can be upgraded to a 2X system if Mike doesn' t complete his own controller.
Cheers
Ben
[ allow you to run different boost limit in 1st and 2nd gear... Ben[/QUOTE]
similar to the Apexi AVC-R?
similar to the Apexi AVC-R?
cuprabaz said:
similar to the Apexi AVC-R?[/QUOTE]
Dunno, I'm not familiar with that. Does it work on our cars?
To be honest, its not very difficult at all. For purposes of this example, lets say that the Jabba boost controller pot gives 0ohms resistance what on max boost and 100ohms resistance on min boost (Not realistics numbers, just for illistration purposes)....
You could wire up a very simple circuit that has a 3 postion switch to select between 3 circuits. The 1st has a fixed 100ohm resistor, the 2nd has a fixed 50ohm resistor and the 3rd has no resistor (Or the Jabba controller pot). This switch will the allow you to select from min boost, 50% boost or 100% (Or pot position) boost. Now all you have to do is find a mechanical way of getting the switch in position 1 when in 1st gear, position 2 when in 2nd gear and position 3 when in 3rd or above.
Obviously you could use relays & microswitches etc instead of a 3 way switch.
How does the Apexi unit know what gear you are in? I don't think the ECU knows does it (Is here a signal from the gearbox?). The only other way to tell would be to compare revs with road speed, and from that work out the gear ratio (And therefore the gear).... but this wont work well ith things like wheelspin etc.
Cheers
Ben
Not too sure mate but the guys i've spoken to who have used them say they are the best.
seen them on ebay for £250ish.
My local tuner has offered to fit and set one up on my car and if it works then i'll buy it. but it won't adjust fueling so i'm thinking no at the moment.
seen them on ebay for £250ish.
My local tuner has offered to fit and set one up on my car and if it works then i'll buy it. but it won't adjust fueling so i'm thinking no at the moment.
cuprabaz said:
Hmmm, not adjusting the fueling would be a big NO for me!
Trouble is, the 1.8T as simple as older turbo cars. On older turbo cars you could just have bleed valves etc to adjust boost (These dont adjust fueling either), but the 1.8T ECU will notice this and say "Eh! That boosts not right!? LIMP!!!".
Chances are, the Apexi unit is just an electronic bleed valve. If it knows what gear you are in then it can adjust the bleed valve accordingly, but like you say it wont adjust the fueling. If this is what the Apexi does then I'd be very surprised if it works on a 1.8T car at all.
If someone can come up with a RELIABLE way of detecting what gear you are in using switches or similar, then I might have a go at moding my boost controller to make it gear dependent boost. Reliable is the key though. I don't want something that detects what gear you are in 80% of the time, or only works if you change gear in a certain way. ideas?
Can the same mechnaism thats used for turning the reverse light on be used for all gears?
Ben
See in the attached Excel sheet (using real data from Bill's Oettinger chip) - if you plot duration, it looks like a torque curve. Likewise duty cycle looks like the power curve.
Mapping to injector duty is the correct way to map for boost and torque as peak injector duration (not duty) is at peak torque which is at peak V.E.
Mapping to injector duty is the correct way to map for boost and torque as peak injector duration (not duty) is at peak torque which is at peak V.E.
Apex-I do a boost controller and a fuel management system as two separate components I believe.
I have no idea if they work on the 18.T. They DO work on the 944 Turbo but it has a traditional standalone wastegate rather than an turbo/actuator and I have always assumed you need a dual port type and they manipulate that. They're a bit "ricer" for my taste though and in any case I'm going to go with a Link2 full management system when it's time to replace the ECU in mine.
I have no idea if they work on the 18.T. They DO work on the 944 Turbo but it has a traditional standalone wastegate rather than an turbo/actuator and I have always assumed you need a dual port type and they manipulate that. They're a bit "ricer" for my taste though and in any case I'm going to go with a Link2 full management system when it's time to replace the ECU in mine.
Glenn said:
Yep, useful info that. I'm not sure if the 2D system uses the duty cycle percentage time or the injector on time to control water injection.
I note that from the raw data the injectors are running at over 100% duty cycle!!! Scarey. I wouldn't want to push mine much beyond 80%!
Ben
Just did some vag-com tests earlier just too see if the water/methanol injection had any effect on intake temps.
I did three runs with and three without, but i tested it on then off, on then off like that to make it fair.
Full throttle right the way through 2nd, 3rd and 4th.
Outside temp was 0.5 Degrees C,
Max Temp without WMI was 33 Degrees C
Max Temp with WMI was 24 degrees C
So i'd say that's a good result as far as i'm concerned. And not a waste of cash as many people think, that together with A&M's FMIC things should be nice and cool come summer.
![[B)]](/forums/styles/seatcupra/oldforumsmileys/shades.gif "Cool [B)]")
I did three runs with and three without, but i tested it on then off, on then off like that to make it fair.
Full throttle right the way through 2nd, 3rd and 4th.
Outside temp was 0.5 Degrees C,
Max Temp without WMI was 33 Degrees C
Max Temp with WMI was 24 degrees C
So i'd say that's a good result as far as i'm concerned. And not a waste of cash as many people think, that together with A&M's FMIC things should be nice and cool come summer.
cuprabaz said:Just did some vag-com tests earlier just too see if the water/methanol injection had any effect on intake temps.
I did three runs with and three without, but i tested it on then off, on then off like that to make it fair.
Full throttle right the way through 2nd, 3rd and 4th.
Outside temp was 0.5 Degrees C,
Max Temp without WMI was 33 Degrees C
Max Temp with WMI was 24 degrees C
So i'd say that's a good result as far as i'm concerned. And not a waste of cash as many people think, that together with A&M's FMIC things should be nice and cool come summer.
Interestingly, my Aquamist has contributed nothing since fitting larger FMIC... so in my opinion it is good for intercooler system which is'nt working well enough to beghin with... and my 1st test results & gains were with the std FMIC AM core. Not any more tho. Zero gain.
It would be a very good add on to std SMIC for people not wanting to hack things around to fit FMIC's, as it does work very well when it has a hot air charge to cool.
Its cold weather currently, so results will probably be better again when and if we get a summer.
To add, go onto the www.aquamist.co.uk forum and have a good read.. There's guys mixing nitromethane to their water/meth mixes and claiming big power gains.
Shock_Xe
Guest
ibizacupra said:Interestingly, my Aquamist has contributed nothing since fitting larger FMIC... so in my opinion it is good for intercooler system which is'nt working well enough to beghin with... and my 1st test results & gains were with the std FMIC AM core. Not any more tho. Zero gain.
It would be a very good add on to std SMIC for people not wanting to hack things around to fit FMIC's, as it does work very well when it has a hot air charge to cool.
Its cold weather currently, so results will probably be better again when and if we get a summer.
To add, go onto the www.aquamist.co.uk forum and have a good read.. There's guys mixing nitromethane to their water/meth mixes and claiming big power gains.
yep, You did have good results over the standard 10core with wmi, Would of been nice if you had had it with CC also as a comparison base. For IHI cars WMI or Big FMIC is essential really. However for chipped cars, a 10 core or wmi is suitable,
My train of thought is Aquamist will cost you around £375 by the time you chucked on VAT/Delivery. Then you have to have a resoviour as its not included (for those who dont want to use washer bottle) which will be £x... then there is constant maintencance (refilling etc). Plus the Water intejection is not constantly working as with a fmic, so there is alot of hassle.... So for an extra few hundered quid is it worth the hassle long term for WMI on k0x equipped cars?
Similar threads
