A thread on the Facts about running ITBs on a non ITB ECU
Posted: Mon May 06, 2013 11:07 am
The other thread turned into a $#!!&!$!fest so I am starting this thread in hopes that we can put down simple facts.
I am looking for people who have run ITBs on the 16v ECUs with widebands and any other sensors to monitor effects and or people who have dyno tuned ITBs on any of these ECUs.
I am not looking for opinions or theories unless you have specifically studied induction theory, engine management and engine tuning to the point of knowing enough that you are extremely confidant your theories are accurate and scientifically/mathematically supportable.
I started this thread because many people say the 16v ECUs will not run ITBs yet not one person has shown this to be true. No one has posted dynos or showed AFR logs or even just explained the effects the mod had on AFRs.
As importantly there are too many variables and unless someone plays with all of them you can't make a blanket statement like ITBs don't work with x ECU.
First lets get the performance part out of the way. ITBs are not a performance mod until your motor is very heavily built. It is not enough of a performance mod to justify the time and cost until you are getting close to Formula Atlantic territory. This thread is not about gaining performance through running ITBs because by the time you can you will have already been on aftermarket engine management for some time.
This thread is about running ITBs on a motor and having it function as well or better than it did with the single TB.
Now for a second of opinion. I think this is a silly mod. I do not suggest support or condone putting ITBs on your 16V. I think it's a waste of time and money. However I just want to try to straighten out facts. Saying you can't run ITBs on a stock ECU is a lot different than saying you shouldn't run ITBs on a stock ECU. Even then it's more just that it's silly from a performance standpoint.
Let's look at the basic science first. There is nothing that makes ITBs drastically different from a single TB setup. The throttle plates in the runners allow for quicker throttle response although we are talking in the range of milleseconds. So quicker response is a good thing. The plates in the runner cause a disruption for the intake pulses that allow for resonant induction. It's interference becomes much greater as the throttle plate closes so anything off WOT ITBs are actually much worse for resonant induction. A single TB will be better for street cars and overall efficiency across the spectrum of throttle travel. This will effect VE but it shouldn't be extremely drastic. An AFM ECU should have no problem compensating. A MAP ECU might have a little more trouble.
Now aside from that you really just have runners of x diameter and x length. Diameter is important at high level tuning but we have seen plenty of variations in runner diameter or people running SPs on LP ECUs, vice versa and such to know that it's not a make it or break it factor for this discussion.
Next is length. Length will effect the VE curve of the engine. It may not be possible to run all ITB variations on all 16V ECUs. Just because you try super shorty stacks on ITBs with bad results does not instantly mean ITBs will not work altogether.
When trying to build a custom system around a set engine management your ideal setup would likely be as close as possible to the original setup from there some tweaking may actually make it better.
So from a VE point of view ideal would probably be to try to match the runner length as close as possible to stock.
I still think there would be a lot of wiggle room on length especially on an AFM ECU.
Now for my personal experience.
I started out with a 4AGZE AW11. At one point the supercharger seized so I converted it to NA while I put together my turbo build. I went from the GZE intake to the NA intake. It was slow without boost but it drove just fine. Going from 2 in runners to 12 in runners seemed to have no effect on how it ran or drove.
I installed the DSM T25 turbo and started driving it at stock boost. Now that I had a turbo I wanted to up the boost but I wanted to do it safely so I bought a wideband. The new setup with the NA intake mani behaved just as others described the GZE behaving as far as AFRs. From there I had a baseline. At just 10 PSI I started to get low end leanout. People said the 440cc injectors could just be dropped in although I couldn't find one person who could back that claim with a wideband or a dyno.
I decided to give them a shot. I learned that the GZE ECU could not handle the 440s and that it ran too rich no matter what. If you reset the ECU it would run way too rich across the board while it started to relearn. After that it leaned out the closed loop back to where it was supposed to be but the open loop stayed int he 10s. The 20% increase did allow me to up the boost to 12 PSI before the low end leanout got scary but that was not enough to lean it out above that.
We theorized the larger 3SGTE AFM would lean it out since the larger flapper should open less per unit of air so I decided to try that. For whatever reason the 3SGTE AFM actually makes it run richer. I was able to tweak the AFM to get closed loop AFRs back down but it did not help WOT AFRs.
After running that setup for maybe 5k miles I got rod knock. I decided I wanted a BT instead of rebuilding the 16V.
I did a lot of research and found the dizzy worked the same, the TPS output was the same, all key components between the GZE ECU and the BT should be able to work together so I decided to try to run the BT on the GZE ECU. I found 380 injectors which after my experience with the 440s I knew should work fine. I also decided I was going to run my DSM T25 on the stock BT.
At that point most people told me it wouldn't work. No way you can run 11:1 compression on a GZE ECU. No way you could boost stock BT pistons. No way you can run ITBs on a GZE ECU.
At that point I realized these were valid concerns but decided to push forward. I actually initially planned on removing the ITBs and running a plenum with a single TB but I decided I didn't want to bother with all the extra work. Plus I'll admit the ITBs are kinda cool. I decided I'd just have to try to get it to work as it was.
Initially I left the wastegate disconnected making it essentially an NA. In the very beginning it had a bad tip in stumble. If you gave it gas quickly it would often stall. My theory is that the GZE has such a large volume of air between the TB and the intake that it probably had some delay in the ECU. The ITBs might respond so quickly the ECU doesn't have time to adapt.
I adjusted the TPS on the very tightest side of it's adjustment range so it would register coming of idle as early as possible.
I played with the AFM adjustments for a while balancing AFRs and trying to reduce the tip in stumble.
Finally I got it to where the tip in stumble was only there when it was cold and even then I could drive around it easy enough. Once it warmed up the stumble was gone and it was fully responsive.
The AFRs behaved just as they had previously. Possibly a little richer than the originally 365s but much better than the 440s. At idle and cruise it found stoich immediately. At WOT it would go from about 13.5:1 around 3k RPM to about 12:1 at redline. AFRs behaved exactly as well as I could ever hope for a stock ECU.
At this point I also had a knock light and a pyrometer.
I took the timing from 6 deg to 8 deg with no noticeable change. I took it back to 6 and hooked up the turbo and slowly took the boost up to about 6-7 PSI.
From there I took the timing back to about 8 deg and it seemed happy. After that I took the timing to 10 deg and I got some knock at low RPM pulling away from a stop so I took it back to 8 deg.
I have driven the car with the VVT turned on and with it turned off and there is no noticeable change in the AFRs, EGTs or knock.
I do not believe that ITBs change anything enough for an AFM ECU to have any trouble whatsoever with them.
I am also quite confidant that with the right tweaking a MAP ECU could be made to run them just as well as the original setup. That however is only theory.
I do not have personal experience with the NA ECU on ITBs but I have seen people run 8 PSI boost on a stock AFM ECU with minor tweaks. If it can do that then I can see no possible way ITBs could effect VE enough to cause any sort of running/driveability problems.
So that is my story and my reasoning.
If anyone has any datalogs, dynos, or personal experience with making ITBs work or with issues that they had with trying to get them to work then post them up.
I am looking for people who have run ITBs on the 16v ECUs with widebands and any other sensors to monitor effects and or people who have dyno tuned ITBs on any of these ECUs.
I am not looking for opinions or theories unless you have specifically studied induction theory, engine management and engine tuning to the point of knowing enough that you are extremely confidant your theories are accurate and scientifically/mathematically supportable.
I started this thread because many people say the 16v ECUs will not run ITBs yet not one person has shown this to be true. No one has posted dynos or showed AFR logs or even just explained the effects the mod had on AFRs.
As importantly there are too many variables and unless someone plays with all of them you can't make a blanket statement like ITBs don't work with x ECU.
First lets get the performance part out of the way. ITBs are not a performance mod until your motor is very heavily built. It is not enough of a performance mod to justify the time and cost until you are getting close to Formula Atlantic territory. This thread is not about gaining performance through running ITBs because by the time you can you will have already been on aftermarket engine management for some time.
This thread is about running ITBs on a motor and having it function as well or better than it did with the single TB.
Now for a second of opinion. I think this is a silly mod. I do not suggest support or condone putting ITBs on your 16V. I think it's a waste of time and money. However I just want to try to straighten out facts. Saying you can't run ITBs on a stock ECU is a lot different than saying you shouldn't run ITBs on a stock ECU. Even then it's more just that it's silly from a performance standpoint.
Let's look at the basic science first. There is nothing that makes ITBs drastically different from a single TB setup. The throttle plates in the runners allow for quicker throttle response although we are talking in the range of milleseconds. So quicker response is a good thing. The plates in the runner cause a disruption for the intake pulses that allow for resonant induction. It's interference becomes much greater as the throttle plate closes so anything off WOT ITBs are actually much worse for resonant induction. A single TB will be better for street cars and overall efficiency across the spectrum of throttle travel. This will effect VE but it shouldn't be extremely drastic. An AFM ECU should have no problem compensating. A MAP ECU might have a little more trouble.
Now aside from that you really just have runners of x diameter and x length. Diameter is important at high level tuning but we have seen plenty of variations in runner diameter or people running SPs on LP ECUs, vice versa and such to know that it's not a make it or break it factor for this discussion.
Next is length. Length will effect the VE curve of the engine. It may not be possible to run all ITB variations on all 16V ECUs. Just because you try super shorty stacks on ITBs with bad results does not instantly mean ITBs will not work altogether.
When trying to build a custom system around a set engine management your ideal setup would likely be as close as possible to the original setup from there some tweaking may actually make it better.
So from a VE point of view ideal would probably be to try to match the runner length as close as possible to stock.
I still think there would be a lot of wiggle room on length especially on an AFM ECU.
Now for my personal experience.
I started out with a 4AGZE AW11. At one point the supercharger seized so I converted it to NA while I put together my turbo build. I went from the GZE intake to the NA intake. It was slow without boost but it drove just fine. Going from 2 in runners to 12 in runners seemed to have no effect on how it ran or drove.
I installed the DSM T25 turbo and started driving it at stock boost. Now that I had a turbo I wanted to up the boost but I wanted to do it safely so I bought a wideband. The new setup with the NA intake mani behaved just as others described the GZE behaving as far as AFRs. From there I had a baseline. At just 10 PSI I started to get low end leanout. People said the 440cc injectors could just be dropped in although I couldn't find one person who could back that claim with a wideband or a dyno.
I decided to give them a shot. I learned that the GZE ECU could not handle the 440s and that it ran too rich no matter what. If you reset the ECU it would run way too rich across the board while it started to relearn. After that it leaned out the closed loop back to where it was supposed to be but the open loop stayed int he 10s. The 20% increase did allow me to up the boost to 12 PSI before the low end leanout got scary but that was not enough to lean it out above that.
We theorized the larger 3SGTE AFM would lean it out since the larger flapper should open less per unit of air so I decided to try that. For whatever reason the 3SGTE AFM actually makes it run richer. I was able to tweak the AFM to get closed loop AFRs back down but it did not help WOT AFRs.
After running that setup for maybe 5k miles I got rod knock. I decided I wanted a BT instead of rebuilding the 16V.
I did a lot of research and found the dizzy worked the same, the TPS output was the same, all key components between the GZE ECU and the BT should be able to work together so I decided to try to run the BT on the GZE ECU. I found 380 injectors which after my experience with the 440s I knew should work fine. I also decided I was going to run my DSM T25 on the stock BT.
At that point most people told me it wouldn't work. No way you can run 11:1 compression on a GZE ECU. No way you could boost stock BT pistons. No way you can run ITBs on a GZE ECU.
At that point I realized these were valid concerns but decided to push forward. I actually initially planned on removing the ITBs and running a plenum with a single TB but I decided I didn't want to bother with all the extra work. Plus I'll admit the ITBs are kinda cool. I decided I'd just have to try to get it to work as it was.
Initially I left the wastegate disconnected making it essentially an NA. In the very beginning it had a bad tip in stumble. If you gave it gas quickly it would often stall. My theory is that the GZE has such a large volume of air between the TB and the intake that it probably had some delay in the ECU. The ITBs might respond so quickly the ECU doesn't have time to adapt.
I adjusted the TPS on the very tightest side of it's adjustment range so it would register coming of idle as early as possible.
I played with the AFM adjustments for a while balancing AFRs and trying to reduce the tip in stumble.
Finally I got it to where the tip in stumble was only there when it was cold and even then I could drive around it easy enough. Once it warmed up the stumble was gone and it was fully responsive.
The AFRs behaved just as they had previously. Possibly a little richer than the originally 365s but much better than the 440s. At idle and cruise it found stoich immediately. At WOT it would go from about 13.5:1 around 3k RPM to about 12:1 at redline. AFRs behaved exactly as well as I could ever hope for a stock ECU.
At this point I also had a knock light and a pyrometer.
I took the timing from 6 deg to 8 deg with no noticeable change. I took it back to 6 and hooked up the turbo and slowly took the boost up to about 6-7 PSI.
From there I took the timing back to about 8 deg and it seemed happy. After that I took the timing to 10 deg and I got some knock at low RPM pulling away from a stop so I took it back to 8 deg.
I have driven the car with the VVT turned on and with it turned off and there is no noticeable change in the AFRs, EGTs or knock.
I do not believe that ITBs change anything enough for an AFM ECU to have any trouble whatsoever with them.
I am also quite confidant that with the right tweaking a MAP ECU could be made to run them just as well as the original setup. That however is only theory.
I do not have personal experience with the NA ECU on ITBs but I have seen people run 8 PSI boost on a stock AFM ECU with minor tweaks. If it can do that then I can see no possible way ITBs could effect VE enough to cause any sort of running/driveability problems.
So that is my story and my reasoning.
If anyone has any datalogs, dynos, or personal experience with making ITBs work or with issues that they had with trying to get them to work then post them up.