O2 feedback at idle and general O2 questions

[foliage]

Is it possible to turn O2 feedback off at idle? I read in this thread that if you have 14.7 in any cell (eg 0 like where the car idles) it can't be turned off. I have a problem where my idle just keeps getting lean no matter what I do, my injector latency is at 1100us, if I change it to 1250-1300 my idle ms goes from 2.15ms to 2.3 and the idle is perfect, then I give the car a rev and it goes back to 2.15ms, I then have to change it to 1600us latency to get an injector pulse width of 2.3ms and so on even with the O2 flags turned off on the cells, obviously I returned it back to 1100 but I'm just trying to figure out why my idle keeps getting lean.

If I went and put a raw hex value of 1 in all of these cells around idle (so that O2 feedback can be turned off) and then turned the O2 feedback flag off, would increasing the injector latency have a permanent increase at idle, eg reaching my target 2.3ms at for a nice smooth idle? I tried putting ridiculous values of 5-6 at low load at idle however the idle ms still stayed at 2.0-2.15ms which was lean so even though O2 was turned off on these cells (with 0 in them) it still seemed to be using O2 feedback.

I also get lots of popping holding revs in neutral around 2-3k as they get down to such a low injector pulse that it is obviously lean as well so would be nice to add just a tiny bit of fuel here. I still want O2 feedback working at cruise though otherwise a simple solution would be to just unplug the O2 sensor.

Saw some people mentioning to fiddle with TTPMIN, if I adjusted this at idle would that cap ALL maps so that they can't make the pulse width lower than this value? If so that might be a good solution to my problem as it seems it is lean and will pop at anything less than 2.15ms

Am I thinking correctly with this, or does idle completely ignore the fuel maps/O2 flags and I have gotten that wrong? I am fairly new to this so any help would be great.

Cheers.

edit: When looking at the standard rb20det fuel maps I notice that there are raw values of 187 (eg negative numbers) etc in some cells, eg they have O2 feedback enabled and it can't be turned off. Is the concept of the raw values that any AFR > 14.7 requires O2 feedback to achieve this AFR hence it can't be turned off? I thought o2 feedback just tried to achieve 14:7 and that was it, I'm guessing that this isn't the case and O2 feedback is used to achieve anything from 14:7 to say 15-16?

If say you wanted to tune with no O2 feedback with standard everything (eg the stock rb20 map) and achieve AFR of 15 at cruise how can you get the pulsewidth low enough if putting 0 = 14:7 estimated, are you then required to change the K factor to get the pulsewidth low enough?

[foliage]

Changed the TTPMIN map from 8 to 11 around idle rpms, changed my injector ms up to 2.25-2.3 which reduced the popping a lot, think this might be an easy solution, can't stop it popping in neutral even if I increase the ms massively, I'm guessing these injectors just have a shitty spray pattern (740cc deatschwerks)

[Matt]

Cannot disable O2 at idle without disabling it all together

You are increasing the latency too high and it will cause the extra fuel to be added to your closed loop (and everywhere else) which will cause the pops

How are the AFRs on the rest of the tune compared to idle? If they are okay, need to work out why the vehicle is lean at idle and fine elsewhere. What is the condition of the MAF?

Take a log of where you are getting the pops to see the MAF voltage, injector time and AFRs

Yes any AFRs > 14.7 have to have O2 feedback enabled due to the limitation of number ranges in the ECU

[foliage]

Rest of the tune is fine, idle however randomly gets as lean as 17:1, I don't currently have a wideband but this is what I have seen it get to. I was just adjusting the latency to see how it affected the idle I was never going to use that as a solution, it was however still lean with that huge increase.

I will take a log however my dodgy fix of changing tpmin at idle revs seems to have fixed it for now.

[foliage]

https://dl.dropboxusercontent.com/u/142 ... 37_30.xlsx

The red section is idling with 8s in the TPMIN scale where it pops badly
The non red section is idling with 10s in the TPMIN scale at idle where it idles pretty good
The 2nd red section is after I give it a big rev (pops like hell when Im revving it as well) and change nothing but the O2 trims the idle lean again and it pops

Basically the O2 trims whatever I do unless I put big numbers in the TPMIN scale, however in some circumstances that makes the idle very rich where it still won't idle nice but puffs black smoke at idle, anything less than 2.35-2.30ms injector time seems result in annoying popping at idle, any suggestions for another work around? Is it worth trimming the O2 lean voltage to see if I can make it trim less off the idle?

I spoke to Trent from status tuning as he has used these injectors before (740cc), he said he had similar problems but he was using a z32 ECU, he ended up using a powerFC and it idled fine suggesting there is something going wrong here but not sure what, are powerFC batch fired injection as well, or are they fully sequential? if the latter then that could just mean it has better injector control, if it is batch fire then it points that something is going wrong but I'm not sure what.

Also you can see the O2 voltage just oscillating badly when it is popping, I'm guessing when it misses it makes the O2 reading completely inaccurate and this just makes it uncontrollable as the O2 thinks it is rich due to the miss but in reality it is still lean in the cylinder?

[Torque]

Hi!

When you get a 'miss' the O2 will report lean not rich.

Also AFAIK the ECU does not run in closed loop at idle (unless you are not in neutral)

What exactly is the issue with the 'popping'?

[foliage]

Of course your will have excess o2 in the cylinder... My bad

[Torque]

No worries,

Sometimes people forget that the O2 sensor measures O2, not fuel.

Anyhow ..
I have that popping at 2000 rpm as well.
And I use 555cc injectors.

I suspect it might be my spark plugs or the coils or ..

My TPMIN is at 3, but this just cosmetics.
The original RB25 tune has a tPMIN of 7 @ 370CC injectors.

Will need to investigate further

[foliage]

I am finding the O2 is learning my settings to make idle richer and then its causing other problems like if I idle for more than 1 minute it randomly goes down to 300rpm idle and pops lots then comes good again so my fix is not a fix at all and I had to revert to a linear TPMIN array. I really want to get an aftermarket plug & play ECU and set it up to just idle/free rev, this will prove if it is a problem with my injectors or a problem with the ECU (or a limitation of batch fire with big injectors)

Does anyone in Adelaide have an aftermarket ECU that is plug & play for an R32 rb20det loom they could loan for a week?

Cheers

[Torque]

Hi

It is my understanding that at idle the O2 sensor is NOT USED.
From my observation it seems that closed loop is only ON when the car is in gear.


Many engines don't idle well at Lambda 1.


As said I have a similar issue and am about to investigate.

I have missed ignition events at 2000 to 3000
Unstable idle
High fuel consumption

I suspect an ignition issue and will check the plugs.

[foliage]

Hi

It is my understanding that at idle the O2 sensor is NOT USED.
From my observation it seems that closed loop is only ON when the car is in gear.


Many engines don't idle well at Lambda 1.


As said I have a similar issue and am about to investigate.

I have missed ignition events at 2000 to 3000
Unstable idle
High fuel consumption

I suspect an ignition issue and will check the plugs.

Matt seemed to think it is definitely used at idle?

If it is not used how can you explain my injector ms changing at idle after I modify the tpmin array? eg I change it from 8s to 10s at idle rpm, it sits at 2.3ms then after a rev and after time it changes to 2.1 and pops again? If I change it to say 15 and make idle super super rich and leave it at that for a few minutes, then change it back to 8 idle goes down to 1.8ms as if the O2 has learned that it is being very rich and pulled fuel?? As originally my tpmin array was at 8 and the idle was at 2.1ms, it seems the O2 will learn 2.1 no matter what value is put in there, any adjustments only change it in the short term.

Wonder what would happen if we all chipped in and started coding a new "main loop" that interpreted these maps ourselves, it seems most of the work has been done with a nistune figuring out these maps, writing the C code which compiles to asm that essentially replaces the original code whilst time consuming wouldn't be impossible. Also surely some crazy engineer in japan has the original tuning software that bosch/nissan used, I'm sure purchasing a license wouldn't be that out of this world in terms of cost due to the age of these ECUs, has anyone investigated this avenue?

Otherwise when problems like these occur it makes sense that tuners just go to the next aftermarket ECU that is available.

[Torque]

Hi!

From memory and when I did testing with my wide-band I only saw Lambda 1 when the car was in gear.
It sounds a bit odd but I am quite sure that was the ways it worked.
I will double check this since it's been a while when I did this.

Which engine/ECU do you have again?

With 1000cc shouldn't your TPMIN be about 1 or 2?
370CC=8 on RB25

I think there is no general issue with NT and idle.

I have attached a screenshot where I hold throttle fixed and you can see how RPM fluctuates due to missed ignition events.
Also the O2 sensor fluctuates quite a bit ..

[foliage]

If I set tpmin at 2 the car will stall, it pops like crazy at anything less than 8 and only idles nice if I raise it to 10, injectors are 740cc and ecu is r32 rb20det running rb25

[Matt]

TTPmin controls the minimum TP which the ECU will use. TP is based from MAF voltage

If you are increasing it, then it raises the floor on the minimum TP (load) the ECU uses. So can potentially stop the vehicle running leaner than the MAF is reading. What is causing the 'pops' is this from the AFRs going too rich initiially? Log should show this if you have your wideband hooked into it

Normally O2 used during idle (watch o2 voltage rise and fall in your log to confirm)

[foliage]

Check the logs I posted, I don't have a wideband to check actual afrs but the O2 seems to need to register always rich for idle to not pop, not sure why, I have smoke tested the intake and have no air leaks, though when I did have air leaks the idle was fine with 440cc GTR injectors so I think the issue is some where else.

Example is the car will onyl not pop if the O2 volts stays at almost 1v non stop, as soon as it gets anywhere below that it will pop.