bpaspi

The LTFT values are generated by the STFT values. It's a statistical calculation. That's fact.
The STFT values are generated by the O2 sensor current jumps, keep in mind that an ordinary O2 Sensor cannot read AFR.
The reason for these corrections is that changes to the engine, or slight malfunctions from the sensors (e.g. a higher connector resistance at a plug) must be corrected to keep the air fuel mixture at Lambda 1 in closed loop.
With Wideband sensors it's possible to keep different AFR even in closed loop, or to keep the closed loop even at high loads. Some engine management computer can handle this, but a wideband sensor is a must for this operation.
Again: a narrowband sensor can only detect to lean or to rich around Lambda 1

The reason why engines need richer AFR's at high load or high engine speed is cooling the inner parts with fuel. This is why open loop is needed.
Because the narrowband lambda doesnt "jump" in richer regions, it's worthless to read its value in open loop.
And because the narrowband O2 sensor doesn't work at open loop, the STFT values are not modified. But the switch to "open loop" is also a signal for the computer not to feed any data into the STFT tables anymore. Otherwise it will only read "to rich" all the time and correct the injector tables into large negative correction values.

The LTFT tables are fed by the STFT values over a longer period and these correction values are for general adjustments to the injection system to adapt an engine to new environmetal circumstances - not Temp and Air-Pressure - these are measured by separate sensors! But to maybe a clogged Injector, a clogged catalyst or exhaust, a dirty Airfilter, a bad sensor reading, a bad valve adjustment, a.s.o.
So it makes absolutly sense to use these long term readed Adjustments also in open loop to keep the engine from running to rich or to lean.

Last thing: I don't know for sure if the Honda ECU uses LTFT in open mode. I thought it would, but if even Hondata makes unclear statements about this....

But for BOSCH engine managements I have trusted proof that they use the LTFT tables even in open loop.

H22toF20

Wow great stuff. Now if we could figure out if Honda's us LTFT in open loop.

spectacle

It doesn't. I don't know why I'm arguing this but it just doesn't. Its ignored. Nothing o2 related is involved with open loop in a Honda ECU. I've seen this with my own two eyes.

spectacle

Not to be a dick (even though this will seem like it so whatever), but you're on a Honda board talking about Honda ECU's...if you don't know about Honda ECU's don't come in here talking about other manufacturers "might's" and "could be's". Lets stick to proven knowledge and the topic at hand.

bpaspi

If it's so sure that LTFT is not used in open loop (what I doubt) shouldn't there a warning for other members not to use the AP1 ECU for AP2 engines?
Some are driving around with this.
In the engine management section is actually a question if it's possible and some stated it is OK....

rog06s2k

This is extracted from an EPA study on the effect of mid-blend ethanol on emissions...
" By segregating the vehicle fleet according to power-enrichment fueling strategy, a better understanding of ethanol fuel-effect on emissions was realized. Vehicles found to apply long-term fuel trim (LTFT) to power-enrichment fueling showed no statistically significant fuel effect on NMOG, NMHC, CO or NOX. For vehicles found to not apply LTFT to power-enrichment, statistically significant reductions in NMHC and CO were observed, as was a statistically significant increase in NOX emissions. "
link: http://www.nrel.gov/docs/fy10osti/46570.pdf
So, in some cases the LTFT's are applied, which ones? my guess is that it started with late OBDII, all of them (manufacturer)? No clue...
Looking around the web and trying to decifer the BS from the truth is daunted task, this subject is a hot topic that is argued everywhere...
A guest mentioned that the Mitsu ECU does, from what I read the GM ECU does, do all Mitsu & GM ECU do? No clue...
What I know for sure is the STFT are definitely frozen to "0" during power enrichment, if LTFT datalog shows them lock at anything else than "0" then my bet is that it is being used during PE map, this would be a logical computing assumption...Otherwise the ECU would freeze the LTFT at "0" as well...
Gnerby has been doing a lot of datalogging, as a matter of fact where is he? It would be nice to see a before & after part throttle tuning LTFT value. Are they negative or positive value?
Until last weekend my knowledge was dead set on no outside influence on the PE map, now I'm definitely keeping an open mind about this...
Sorry for the long post
Cheers

spectacle

Early AP1 ECU's are rich by design in the WOT map which works well enough for the AP2's extra displacement. Has nothing to do with LTFT

s2konroids

On the face of it 'initially' you would think that the LTFT is applied during WOT for certain criteria that could alter the tune/effect the tune, a duff injector etc as bpasi said, but it isn't.

I agree with spectacle here.

Why do i think/know this.

a) I've read books recently as i tuned my car myself on a dyno using a greddy EMU, i didn't specifically read the fact about LTFT on honda ECU's but read it elsewhere and the fact that i read about LTFT's etc.
b) Prior to a, i used a vafc2 (piggyback) to tune my car myself there was a lot of wishwashy comments about 'the tune will get learnt over', fast forward in time and several dyno visits after the tune hasn't altered, well it did 'minutely' due to being a hot day once (not unexpected).
c) Again the o2 sensor is ignored in WOT, see spectacles point.
d) If it was it would be constantly fluctuating (not good for different countries).
e) Get a obd2 logging unit, make the car go in to 'open loop' by satisfying all requirements of open loop, notice that the LTFT and STFT drop to 0 and stay at 0 until you get back into closed loop.

Gernby

LTFT and STFT are just 2 values. There are no LTFT tables or STFT tables. As everyone agrees, they are only calculated while in closed loop.

I always thought LTFT was applied during open loop, since it seems like that would be its primary value. It would be able to compensate for aging fuel systems and air filters to prevent damage to the motor. It would also make it safer to run fuels with varying amounts of ethanol.

However, it seems clear that LTFT is NOT used in open loop in my '08 AP2. When I study my FlashPro datalogs, it is clear that it doesn't matter how large or small my LTFT is when I go full throttle, my AFR curve at full throttle stays consistent.

After doing a full E85 test and tune for a couple weeks, it was also very clear that LTFT was not applied, since my AFR was very inconsistent at WOT. If LTFT had been applied, E85 would have worked much better. In fact, I think that is why r06s2k's quote is not applicable to this discussion. The functionality of a Flex fuel ECU is very different.

spectacle

Thanks gernby. Hopefully this put an end to the debate.