EFIDynoTuning Forum

Agree there isn't an ITHBMA, just meant the FN818 (ITHBMA) function.

Regarding your suggestion to increase the ISC multiplier, Initially I did add a bunch to those tables in an attempt to increase the ISC air flow to try and reach DSDRPM during start-up. But it eventually became apparent it didn't matter how much I added to the tables because the ISCDTY would hit 60.2% and not go higher and never reach DSDRPM, and of course the IPSIBR would just start to climb while the ISCDTY was stuck at 60.2% and never reach DSDRPM until the start-up adders expired and DSDRPM came down to something less than 1100rpm. So then I just reduced the ISC multiplier tables back down a little to something typical/reasonable because it was not helping, and pulled RPM out of the adder tables to reduce DSDRPM during start-up. That is when I started looking for whatever is limiting the ISCDTY to no more than 60.2% in idle control. Can't seem to find a root cause.

The injector data, MAF, and engine size are scaled 50%. Is there a correlation possibly?

IPSIBR only climbs if I set the start-up RPM adders such that the total DSDRPM exceeds around 1200 RPM for the first few seconds of cold start. That seems to be because the ISC won't open far enough to reach DSDRPM. IPSIBR is always very close to 0.000 at all times otherwise. I reduced the ECT based start-up RPM adders to no more than 1100 RPM (700 NUBASE + F825A + FN826A + BZZRPM) total which is all I can get out of the ISC during cold starts.

This is really only an issue for cold starts. It doesn't have any issues when DSDRPM is 700 for Park/Neutral when ECT is at normal temps, the ISC duty is 29%, IPSIBR is very close to 0.000. The idle airmass of 0.38 at 700rpm is correct based on logging at normal temps with no fans or A/C. The ITHBMA of 0.32 is also correct. Throttle Body air was determined by capping off and unplugging the ISC, then let it idle at normal temp (no fans or A/C) and 0.32 PPM is the measured air flow with no ISC. The air flow values are based on logging 'AM_UNSCALED_CALC' Is this correct?

Bringing this one back as I haven't uncovered the root cause and continues to be problematic at start-up. The maximum ISC duty cycle it will command at cold start is always clipped at exactly 0.602 (60.2%) until required air mass drops enough for the ISC to begin to close. This is in park. However, when driving dashpot will rise and ISC will open accordingly but hesitates at 0.602 momentarily then shoots up seemingly as if to catch up. At which point the ISC duty will rise beyond 0.602 up to the 0.98 clip. So the ISC will exceed 0.602 at times, just never during cold start, which prevents hitting cold start desired RPM. Basically nothing greater than around 1100 RPM when engine is cold, so for now I just compromise and limit the cold start RPM adders to limit DSDRPM to nothing greater than 1100 RPM. Well anyway, just wanted to throw it back out there; open to ideas what the issue might be. Latest tune attached. Thanks.

I was hoping that maybe it was an issue with the older definition I was running, but the ISC clip persists even after downloading the latest 2023 version. Oh well, I will keep searching for the culprit. For now, I just reduced the cold start DRSRPM adders so that it didn't try to exceed 1100 RPM which seems to require just slightly less than 60% ISC on cold starts.

I got curious and downloaded your latest 2023 definition and sure enough the FN1052 table popped up. But that version also has the unscaled views of the fuel and MAF data, which didn't appear to be correct or even close. I prefer just leaving it scaled the way I'm used to seeing it over the years by using earlier definition FBGIO_200510. I tried changing the scaling percent to 1 as directed in the hi-slope comments, but that didn't work out right. So I took a random value from the MAF transfer FN036 and divided it by the previous value and then multiplied the scaling percentage value by that ratio and that got the MAF, injector slopes and breakpoint back to what I had previously using the 2020 version of the definition. Not sure the comments in the 2023 definition do a good job of explaining how to back out of the unscaled view. But I got the scaled view of injector data, MAF, manifold volume, SARCH correct again regardless. I haven't tried to load this into the QH yet to see if runs the same before I mess with FN1052.

No, sorry, not familiar with that table. Does it go by another name or number by chance?

Thanks for the suggestion. I will go ahead and try turning adaptives back on and lower the max allowed ACT for update if I can find it.

The MAF error is essential zero when ACT and ambient agree within a few degrees. But once they diverge by more than +10 F degrees or so it sure seems like the MAF error starts walking up proportionally. Fuel injector data is scaled FRPP data for their injectors so it should be correct. It sure seems correct when ACT and ambient are in close agreement.

It could potentially have something to do with vapor lock, but fuel pressure monitor shows normal 44psid and since hot restarts fire right off I don't think it is the root of the matter. Especially since the dead lean condition is persistent for several minutes after hot restart until the ACT comes down somewhere close to ambient.

Not sure. Good question for Decipha. But my understanding is that whatever KAMRF is already in memory is always going to be added on top of startup enrichment adding that extra % to lambse.

The problem with having the KAMRF enabled is once they reach +20-30% during hot idle or right after a hot restart is that the next morning those trims are added during cold start and warmup until the updates pull them back down to near zero. So it is way too rich for several minutes. Fueling is dead on when ACT is in agreement with true ambient air temps.

The reason I keep KAMRF disabled is because when at idle for any amount of time leading to high ACT readings (not necessarily true high ambient) the KAMRF will climb up +20-30%. After shutdown and overnight cooling the next restart with true ACT matching ambient, the KAMRF is now driving lambse way rich until enough time passes for KAMRF to update and settle down to something reasonable again.

My opinion is that the ACT sensor is hotter than the air flowing past it during hot restarts for the first couple minutes. If the ACT reading was true then the actual air density would be less than ambient density and the calculated air density would be correct so fueling would be correct. But that isn't what happens. The fueling is dead on when ACT is reasonably close to ambient, but lambse climbs as ACT begins to stray from ambient. If ACT was truely what it was reporting then air density would be less and fueling would be calculated correctly.

I did have to add a lot to the startup enrichment for +200F and above to get it stay running for the first minute after hot restart.