Fuel Trims Wildly Different with Flash Pro

FlashPro questions & answers specific to the 2006-2009 (US, UK, Asia) S2000
Post Reply
imstimpy
Posts: 8
Joined: Mon Nov 04, 2013 2:04 pm

Fuel Trims Wildly Different with Flash Pro

Post by imstimpy »

- serial number of your FlashPro: <I don't have it at the moment>
- FlashPro Manager version (not 'the latest'): 1.6.3.0
- the calibration (as an attachment): attached
- a datalog of a problem (as an attachment): attached
- what sort of vehicle you have: 2006 US S2000
- anything about the car which has been modified from stock which may affect the ECU (don't include what sort of rims you have nor stereo etc): K&N FPIK, J's header, 70mm Berk high-flow cat pipe, T1-R 70EM cat-back exhaust

As far as I can tell, something about the FlashPro is affecting my car's fuel trims. With the FlashPro disconnected during key on and engine start, fuel trims tighten up quickly to what I expect. With FlashPro connected on start, fuel trims swing wildly.

Initially I thought the Live Tuning Tables were the culprit, but I've disabled them and the problem has persisted. I also thought it was a pre-O2 exhaust leak, but restarting the car seems to fix it. I've gone so far as to remove my header to inspect for exhaust leaks and found nothing.

These images were generated by logging the same strip of road, the same tune, the same gas, but different times of day.

The first trip was captured by starting the car without FlashPro connected.
Image

The second trip was captured by starting the car with FlashPro connected.
Image
Attachments
Fuel trim back clip.fpdl
(376.68 KiB) Downloaded 269 times
Fuel trim example.fpcal
(8.53 KiB) Downloaded 275 times
pepperdanky
Posts: 81
Joined: Mon Oct 14, 2013 6:14 am

Re: Fuel Trims Wildly Different with Flash Pro

Post by pepperdanky »

different times of the day: how different are the IAT's?
User avatar
Hondata
Site Admin
Posts: 10434
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: Fuel Trims Wildly Different with Flash Pro

Post by Hondata »

+- 3% is about as good as you'll get with the Honda fuel injection system. There are many factors which the PGMFI system either does not consider or the input variables to the ECU are not 100% accurate.

It is difficult to see how the FlashPro being connected to the vehicle would affect closed loop (unless there was a strange o2 sensor current issue) but the test you should make is to do a short drive under the same conditions with the FlashPro both connected and disconnected, and then compare the fuel trims.
Hondata
imstimpy
Posts: 8
Joined: Mon Nov 04, 2013 2:04 pm

Re: Fuel Trims Wildly Different with Flash Pro

Post by imstimpy »

Both logs were started after a period of driving, shutoff, then restarting the car; the morning I drove to Starbucks before starting the log, the afternoon I had been autocrossing.

Ambient temp was ~20*F different (50*F in the morning, 70*F in the afternoon). IAT on startup was also 20*F different, lower in the morning, and ECT was 5*F lower in the afternoon (car sat longer before startup). IAT during the logs was also ~20*F different.

For a narrow focus with cruise control, the trims seem similar. For the entire duration of the afternoon drive, long term fuel trim starts out to lunch and finishes out to lunch.

Morning:
Image

Afternoon:
Image

The only other suspect piece I can offer is the morning started with a reflash whereas the afternoon did not. I have seen the unusual fuel trims post reflash, as well as correct fuel trims post reflash. I reflashed the car first thing before starting the day, logged the drive to Starbucks, then flashed the changes. The second flash that morning, at Starbucks, the car lost the immobilizer after the flash and required me to reflash the car again to revive it.

While I have done your exact request in the past, log with FlashPro connected and disconnected, I will do it again and prepare data for viewing. It may take me a few days, since gather a log long enough to establish fuel trim trends takes some time. I will eliminate reflashing from the equation as well. I'm not 100% confident that reflashing has nothing to do with the problem, but I'm all for isolating the problem. I know it was a bit early to post this, but I wanted to start the discussion early because it does not seem normal to experience such fluctuations. +/- 3%, sure, but +/- 7% with no evident trends?
Post Reply