patch for 88590015: control front and rear O2 with wideband
Thread Starter
EvoM Guru
iTrader: (50)
Joined: Mar 2006
Posts: 9,675
Likes: 132
From: Tri-Cities, WA // Portland, OR
How the heck did I miss this thread, I thought this patch had kinda "died off". This is also great because, if I understand correctly, the ECU will now be looking for the AFR placed in the fuel table (with default fuel scaling) and will allow running 16.5:1 on e85 on my alt-map and something more conservative like 15.5:1 cruise on my main map.
awesome! ...time to find a wideband! (uhg, and take off that pipe again to get a new bung in it)
awesome! ...time to find a wideband! (uhg, and take off that pipe again to get a new bung in it)
what perfect timing, I just got my car running after being down for 2 months with clutch issues. Can't wait to test this out. I woner how well it will work with my wideband being after my testpipe, its pretty far back. I may need to get a bung welded in closer.
Ok, really dumb question....
I know there are differences in the narrowband switchover between gas and E85. I also know that you want a slightly richer AFR at peak boost with E85. I did not realize that the output voltages were so different in the AEM sensor between the two fuels.
According to this doc (page 6): http://www.aempower.com/images/produ...%2030-4100.pdf
It looks like the max AFR for pure ethanol (I know, add 15+%) that a normal wideband (at least the AEM) can read is 12.28.
When talking about going 16.5:1 or 17:1 with E85 are we talking about the gasoline equivalent which, according to that cart, would be ~3.43v or will I not be able to accomplish what I'm thinking with that module (or with an LC-1)?
I know there are differences in the narrowband switchover between gas and E85. I also know that you want a slightly richer AFR at peak boost with E85. I did not realize that the output voltages were so different in the AEM sensor between the two fuels.
According to this doc (page 6): http://www.aempower.com/images/produ...%2030-4100.pdf
It looks like the max AFR for pure ethanol (I know, add 15+%) that a normal wideband (at least the AEM) can read is 12.28.
When talking about going 16.5:1 or 17:1 with E85 are we talking about the gasoline equivalent which, according to that cart, would be ~3.43v or will I not be able to accomplish what I'm thinking with that module (or with an LC-1)?
That said, would we be able to effectively use the UEGO with this mod or would we need to scale your modifications differently? I'm not sure if the default range of the LC-1 (which I think you are using) is the same as this or if that even matters (but I would think it would).
Thanks MrFred!
Oh, it does! You got me spendin' all my money like fergie mrfred 
If you get a minute sometime I'd love a quick rundown of psuedocode/laymans terms for how the feedback system works (and how this mod works overall)
If you get a minute sometime I'd love a quick rundown of psuedocode/laymans terms for how the feedback system works (and how this mod works overall)
Thread Starter
EvoM Guru
iTrader: (50)
Joined: Mar 2006
Posts: 9,675
Likes: 132
From: Tri-Cities, WA // Portland, OR
I'll post the source code this evening, but in very basic terms there are two things that get done:
1) I tell the ECU to use the fuel table even when operating in closed loop. This sets the base fuel pulse from the fuel table.
2) I lookup the octane-averaged AFR, and use this target AFR to convert the 0-5V WB signal into a 0-1V NB signal where the crossover occurs at the target AFR. If the WB AFR is richer than the target AFR, then the I set the simulated NB voltage to ~1 V. If the WB AFR is leaner than the target AFR, then I set the simulated NB voltage to ~0 V. In practice, I don't use a step change in voltage between 0 and 1 V. I have a steep slope that transitions between 0 and 1 V over a range of approx 0.3 AFR. This NB signal then replaces the front O2 signal, and the ECU uses it to adjust the fuel trims to ensure that the AFR hits the target values in the fuel table.
dudical - I recommend moving the WB sensor as far up the DP as you can. The closer to the engine, the sooner exhaust gases from the engine will hit the sensor and allow a faster feedback loop for closed loop AFR control. Mine is at the end of the DP, and I got a bit of surging on pump gas due to the somewhat slow response. (I don't get any surging with E85 though because it seems to make the same power over a wide range of AFRs.)
1) I tell the ECU to use the fuel table even when operating in closed loop. This sets the base fuel pulse from the fuel table.
2) I lookup the octane-averaged AFR, and use this target AFR to convert the 0-5V WB signal into a 0-1V NB signal where the crossover occurs at the target AFR. If the WB AFR is richer than the target AFR, then the I set the simulated NB voltage to ~1 V. If the WB AFR is leaner than the target AFR, then I set the simulated NB voltage to ~0 V. In practice, I don't use a step change in voltage between 0 and 1 V. I have a steep slope that transitions between 0 and 1 V over a range of approx 0.3 AFR. This NB signal then replaces the front O2 signal, and the ECU uses it to adjust the fuel trims to ensure that the AFR hits the target values in the fuel table.
dudical - I recommend moving the WB sensor as far up the DP as you can. The closer to the engine, the sooner exhaust gases from the engine will hit the sensor and allow a faster feedback loop for closed loop AFR control. Mine is at the end of the DP, and I got a bit of surging on pump gas due to the somewhat slow response. (I don't get any surging with E85 though because it seems to make the same power over a wide range of AFRs.)
Thread Starter
EvoM Guru
iTrader: (50)
Joined: Mar 2006
Posts: 9,675
Likes: 132
From: Tri-Cities, WA // Portland, OR