Flex Fuel Tuning - Trim Swings with Ethanol Increase

[LaTechGTO]

Sooooo I will start off with saying I have never tried tuning a car on flex fuel with a flex fuel sensor. When I have tuned in the past, it was very straight forward to tune for 1 fuel type. I searched the web for a few days now, and I can't seem to find what I am looking for. There are a ton of threads on spark advance adjustments in relation to flex fuel, but I can't seem to find anything on supplemental fueling.

My STFTs (LTFTs are disabled) were almost spot-on before installing the sensor and activating flex fuel in the tune as the car [09 G8 GT; e38] tuned on the "normal" 93 pump gas (e10). Yesterday evening on my way to work I decided to put a little bit of E85 (probably more like e70), which brought the content to high 3X% then settled in the high 2X%. I noticed the STFTs had increased to ~5-6% while cruising and 20%+ at part throttle acceleration while remaining in CL where they are normally right at 0%.

This might sound st00pid, but is there a table to supplement CL fueling for various alcohol contents, or is my only option to jack with the AFR table to force fueling? I am not sure if the latter even works... I reckon I might try when I leave work in a couple of hours. I was really hopeful that the injectors fueling would account for the alcohol content.

Sorry for the long post! I've attached the tune below for reference.

345s_and_E_V1.2.hpt

Edit: Does transient fueling need adjusting to account for fueling differences? The fuel mass evaporation and gain doesn't seem to logically make sense since it is all still reaching the cylinder in some form or another.

[07GTS]

u still have cat over temp on that can skew trims, same as desoot when entering PE should disable it, other then that dont touch the stoich numbers they are correct if its not staying within a few percent after it settles then something is off, maby change default percentage to 80% then if it fails u go rich not lean no matter what fuel u have in, the transients will be a little different cos of the difference with ethanol but wont be a huge shift like 20%

[LaTechGTO]

u still have cat over temp on that can skew trims, same as desoot when entering PE should disable it, other then that dont touch the stoich numbers they are correct if its not staying within a few percent after it settles then something is off, maby change default percentage to 80% then if it fails u go rich not lean no matter what fuel u have in, the transients will be a little different cos of the difference with ethanol but wont be a huge shift like 20%

Thanks for taking a look! Sorry for the delay; it didn't automatically subscribe me to the thread.

Correct - I left it active as the car has cats and lowered the max enrichment to reduce its effect. COT, as I understand it, will only be a periodic swing. The fueling discrepancy I noted is constant across the board.

Yeah, I definitely didn't want to go butchering the AFR table. Thanks for the reassurance.

Good call on default percentage! I'll adjust it accordingly as I haven't reached greater than 50% E yet while tuning. Taking small E steps to prove my issue...

I filled up this morning right before I got home from work. 10 gallons of e85 (e70 minimum) and 4 gallons 93oct (e10 maximum). The E stabilized around 45%, and I noticed the STFTs start swinging right away. It's like the tune isn't automatically adjusting for the E content change. So weird. I keep having to tune the fuel delivery after every fill-up, which completely negates the need for a flex fuel sensor.

[GHuggins]

Change transition and delay volumes to 0 - this should allow it to update as ethanol content is changing...

[LaTechGTO]

Change transition and delay volumes to 0 - this should allow it to update as ethanol content is changing...

Thank you! I made the changes and hope to try it out after the hurricane passes. I looked back to see where I got that data, and I copied it over from a 2012 Holden Sedan tune in the repository. I also noticed that tune has a virtual FF, not a sensor. It totally makes sense as to why this should be 0 for my application as the sensor is right at the fuel rails.

I also changed the Comp Change Thresh to 0.5%.

Hopefully those changes will prevent me from chasing this tune down every tank of fuel.

[LaTechGTO]

I've been meaning to follow-up since I loaded the tune then drove in to work Sunday night. Adjusting those tables definitely helped with the responsiveness of fuel parameters being adjusted within the tune so that was a "win".

I still had to adjust the MAF table slightly again, and this time it required another percent or two removed. I'm not sure why I have to keep chasing my tail. It's still about .5%-1% richer than I'd like as it sits, but I'm about to add a little more E in the morning when I leave and want to see what it does.

I'm too lazy to remove the WB off of my GTO so I'm just going to order another then I'll properly tune WOT (just fattened it up temporarily).

Overall, I'm very impressed with the increased E, but I can't tell if it's just a placebo effect or not haha.

I've read quite a few topics where the starting of the engine seems to get sluggish. That's my next issue to try to correct as I've been noticing the difference more and more as the E goes up. I'm going to try to increase the cranking fuel tables first.

Thanks again for all of the help!

[LaTechGTO]

Quick update:

I left work this morning and went to the gas station to get some e85 (seems to be closer to e70). I logged from the gas station to my house in the first file. You will see that the trims are pretty negative, which seems counter intuitive to me as I would think the high concentration of ethanol would swing the STFTs more positive.

In the 2 and 3 files, the trims are still negative outside of my acceptance, but there is a pretty decent correction that took place from the time I stopped and saved the log then restarted. Maybe 1 minute in between. On log 4, there is even a better correction when I restarted the log, which was less than a minute from the previous too.

What gives? Why is there such a long period for the STFTs to line out? Also, why would it swing negative to remove fuel instead of positive to add fuel? Again, the sensor is located right at the fuel rails.

Another small item that keeps bothering me. I keep getting a trace knock (<1deg), and I have shifted my timing tables to try to get rid of it. The knock showed itself in the 2nd log, but it might be in the others - rushing to get to work right now and wanted to post this as early as possible in hope for some feedback. Is it safe to assume this isn't real? It's always as part throttle and relatively low load conditions.

See attached logs.

[GHuggins]

Log injector tip temp and purge. Just sounds like something else is off.

[LaTechGTO]

Log injector tip temp and purge. Just sounds like something else is off.

Thanks! I'll do that when I leave work in the morning.

[LaTechGTO]

Log injector tip temp and purge. Just sounds like something else is off.

Here are some of the latest logs with the requested parameters being logged.

1-2 are from this morning when I left work, and the last 3-5 are from my drive back in. 1 has a short city drive toward the interstate and a short bit on the interstate, and the second is on the interstate. 3 is a short highway back to the interstate from the house with a short portion of the interstate. 4 is a pretty much all interstate, while 5 is stop and go.

I also remembered that I never loaded the newest tune so I loaded it as well.

I keep forgetting to ask... is it necessary for me to still attach the .xml files? I haven't messed much with sending logs since the new version of HPT was released, but I do remember having to send config files for the logs to be properly displayed. Is the XML the same thing now? Didn't have time to search... gotta run to the field.

[LaTechGTO]

Any ideas? I'm all ears and trying to learn about flex fuel tuning. Thanks!

Oh, and here is the latest tune file I am running for reference. Drives really great at part throttle.... very crisp. My WB should be here tomorrow, and hopefully I will finally get a day off work so I can dial in WOT as well as touch up some of the part throttle cells now that I won't have to strictly rely on NBs/STFTs.

V5.0 Normal.hpt

[openwheel25]

I would think if everything is spot on with gasoline. Than the only thing you change is the fuel and now it is rich or if it is now lean it would seem the only thing that can make that happen is the computer is targeting the wrong afr for the fuel that is now in the tank. If the sensor is working and it is changing the afr for the fuel it is either sensing the wrong amount of ethanol or the table for the afr for the percent of ethanol is wrong. I don't see any reason for not changing that table to fit what your sensor is telling you. Since your car is spot on before the fuel change there should be no reason to change your air flow tables. If you change your air flow tables every time you change percentage of ethanol it would be like running in circles.

[Lakegoat]

I drove to the local station that has E85--or whatever it is today. I logged going there, filled up, and logged when I left. The difference was only about 3 percent rich according to stft's and a little less on EQ error(wideband). Idle fuel remained the same, it was the low rpm cells--no wot or jerky throttle. I had a policeman behind me and my car is loud.

[LaTechGTO]

I would think if everything is spot on with gasoline. Than the only thing you change is the fuel and now it is rich or if it is now lean it would seem the only thing that can make that happen is the computer is targeting the wrong afr for the fuel that is now in the tank. If the sensor is working and it is changing the afr for the fuel it is either sensing the wrong amount of ethanol or the table for the afr for the percent of ethanol is wrong. I don't see any reason for not changing that table to fit what your sensor is telling you. Since your car is spot on before the fuel change there should be no reason to change your air flow tables. If you change your air flow tables every time you change percentage of ethanol it would be like running in circles.

Yeah, it does seem pretty odd. The largest swings are only for the first ~30min for whatever reason, and they swing the opposite direction than what I would expect; when increasing E, the STFTs go negative where I would expect they would swing positive if something was amiss. Agreed here... this is what I have been doing. I am going to fill up with 93 oct (e10) when I leave work in a few hours then see what happens. I would expect the STFTs swing positive even though they should logically swing negative.

I drove to the local station that has E85--or whatever it is today. I logged going there, filled up, and logged when I left. The difference was only about 3 percent rich according to stft's and a little less on EQ error(wideband). Idle fuel remained the same, it was the low rpm cells--no wot or jerky throttle. I had a policeman behind me and my car is loud.

I'm interested to see what happens once I get my WB installed, which should be arriving later today. Yeah, I am noticing that the car still drives completely fine and doesn't show any signs of being rich or lean outside of the STFTs.

At the end of the day, I am just going to send it

I'll report back with my WB findings in case someone else runs into this "issue".

[marksrig]

Why do threads like this just die?

[LaTechGTO]

Why do threads like this just die?

My bad. I got busy with life and forgot to follow up. I built a stroker motor for this car and stopped even looking at the swings when I would make big jumps in E. Car still drove great.

[marksrig]

My bad. I got busy with life and forgot to follow up. I built a stroker motor for this car and stopped even looking at the swings when I would make big jumps in E. Car still drove great.

Thanks for coming back to answer. I finally just trimmed up the AFR table and interpolated down. My injector data is properly configured according to the manufacturer which I sent my tune to and the sensor output is correct for the amount of alcohol in my fuel. So far this seems to work well. I still think it's injector issues though.

[LaTechGTO]

Thanks for coming back to answer. I finally just trimmed up the AFR table and interpolated down. My injector data is properly configured according to the manufacturer which I sent my tune to and the sensor output is correct for the amount of alcohol in my fuel. So far this seems to work well. I still think it's injector issues though.

Just for clarification - mine would only do that on a fresh fill-up with big changes in E. Either going from e10 93oct to E85 (actual e70 here) or vise versa. After 1-2 miles it would be spot-on where it should be. I played with the transition settings and such with no luck. Haven't checked recently to see if it's still doing it.