I would think it has to be the length of the loop. The distance between the injector and the O2 sensor. By the time the exhaust gets to the sensor you are already in a different place so the correction becomes meaningless as it it is history rather than what is happening right now. With OL the ECM will not make corrections for current conditions based on historical data that is not in sync.. I would supect if you get the OL tune perfect and then slow down the response rate of O2 corrections it might work a lot better.. as you would get a correction for the average condition rather than each little wiggle. A thought anyway I do know on aftermarket ECMs this works well..
Dave

 

Hi Dave!

good point, stock, the O2 is just a few inches from the head, with full length headers and O2 in the collector, its far away from its stock location, therefore, there must be a different response time...

which tables do you have in mind to change?

i will try the open loop thing, and then decide, if i go back to C/L with some slowing down, or keep O/L..

thank you for the suggestion!

 

If a SD 90-91 deal I would be looking at these..

Fast O2 Rich/Lean Threshold at Idle
Fast O2 R/L Threshold Hysteresis at Idle
Upper Zero Error Ref Slow O2 R/L at Idle
Lower Zero Error Ref Slow O2 R/L at Idle

I think if you lossened these up a bit by allowing a bit more error and more hysteresis it would not try to chase it down so quickly. I tune aftermarket ECMs so don't claim to be a GM ECM tuner but they all work the same more or less and that is where I am coming from.. Any time you osscilatein a closed loop system it is usually from over control or the is shorter than the delay time.. if you can get it to make smaller changes per step that would help as well.

I general terms it comes to this tell the system to not be concerned with minor errors and not try to chase down every wiggle and you have to give it a very good starting point so there is little to change as well. You might as well try changing the target air fuel ratio for closed loop also, I find it funny that in the SD ECM table I looked at that it even mentions air fuel ratio as it is not what the senso measures in the first place it measures oxygen so Lambda is the appropriate unit. Air fuel ratio is inffered from lambda with a typical blend of gasoline. Unfortuately the blend changes even worse when the car was built and originally programed there was no ethenol used so while lambda 1 can be tuned in both cases and in both cases provide good engine operation the actual air fuel ratio will be different..
Dave

 

I had PCMFORLESS email me a tune and it is near perfect in regards to drivability. I havent had a chance to datalog it yet. I have zero vacuum issues with the cam, no surge, no bucking, no hesitation, no killing, etc.. The only thing it may need is the idle bumped up just a tad when the AC is on. My cam specs are 226/234 565/565 109LSA on the stock bottom end. If I was in your situation I'd get it set on a dyno first then tweak it if needed for other driving conditions.

 

Hi all and thank you!

@ddahlgren

i already lowered the values, and it seems to help the idle a bit...but, i don´t think, that is helping for part throttle driveability..
anyways, when you look at the typical NbO2 curve, AFR vs mV, just a small change in AFR will make a huge change in output voltage...so´, it´s near to impossible, tu use the stock O2 for a real measuring device..as its stated in some other forums..it just bouncing between rich/lean....
how much vacuum do you pull at idle with your cam?



@rickreeves1

sounds great, problem here in Austria is..there are no dynos(at least for the public)..yes, a few, but austria is a different world..you drive to a dyno shop, pay a lot of money, run the car and drive home, they don´t give you the time to tune your car... ...the few people, that are in professionell motorsports here, are mostly owners of such shops with dynos, so, they are not limited in time...