The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Fuel Injector Clinic
Please Support STM Tuned

ECUflash Not Lean In Closed Loop But Fuel Trims Always High

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

randman2011

10+ Year Contributor
552
282
Feb 26, 2012
Indianapolis, Indiana
Maybe I'm not searching well enough, but I haven't found anyone else describing the issue that I have. I'm tuning the black box ECU in my 98 GSX for Deatschwerks 1000cc injectors after an engine build. Background:

It was running well but would suddenly go insanely lean (35-100:1 AFR on the gauge) and I had a hell of a time finding the problem. It turns out that the Minitimer to USCAR adapters took just enough slack out of the injector harness that the wires were breaking at the injector connector, causing it to intermittently run on three cylinders. I replaced cylinder 3's connector with an USCAR connector without an adapter and it ran perfectly for five miles before I lost cylinder 2. I have now replaced all four connectors and the car runs perfectly. For about 25 minutes.

Deatschwerks provides a flow value of 945cc/min for my four injectors. I put 941 in ECUFlash and it runs fantastically. Until the fuel trims start updating. They are always positive. The longer I drive, the higher they go, maxing out at 12.5%. If I adjust the injector scaling factor down to compensate, the fuel trims will still max out. The car is not running lean and the O2 feedback trim is not overly positive at any given time, but the fuel trims keep rising. What makes it especially confusing is that at WOT (or any open loop operation) the AFR is now rich by whatever percent the scaling factor has been enriched plus the long term trims, pointing the finger at closed loop software. This means that my rock solid cruise is interrupted by bouts of 7.35:1 WOT awfulness. Additionally, the throttle tip-in additions seem to be negatively affected by the adjusted scaling factor, meaning that the longer I drive the car the worse driveability gets.

I can set the scaling factor back to 941 and reset the fuel trims and it runs well again until the trims max out at 12.5%, so I'm now disconnecting the battery twice a week to reset things. I'm completely baffled by this issue. Does anyone know where I can start looking?

Additional info:
Deatschwerks 1000cc High Z injectors (resistor box removed)
LC-2 wideband, narrowband output replaces front O2 signal to ECU
Latest Ceddy ROM. Default settings except injector calibration and periphery bits to disable O2 CELs.
 
Did you touch the deadtimes / latency's? I only see you mention adjusting the injector scaling.

http://ceddy.us/tuning/injectors/

I use the steps listed above for injector tuning.

The deadtimes from Deatschwerks were used originally but they were so far off that the car had trouble idling. Their numbers assume peak and hold drivers, not saturated drivers like we have so I found some of their injectors that have specs for both drivers listed and did some adjustments to to my latency data. It is very significantly improved, but I haven't touched the values since due to the inconsistent feedback that the car was providing. I wouldn't think that latency would play into this issue since all three trims rise to 12.5% at roughly the same rate and they aren't handled differently between closed and open loop, but I have certainly been wrong before.

For the record, I am VERY happy with the response that these latency values provide. Tuning to the fuel trims says that it's lean everywhere and tuning to open loop says that it's rich everywhere, so tuning has been put on hold until I get to the bottom of this.
 
Last edited:
1st let me say i know nothing of evoscan. I was cruising their site in an effort to educate myself and came across a Tech Info page with the title Mitsubishi EvoX - Stopping the fuel trims affecting WOT fueling
(link HERE).
Seems to be the same problem your having. Hopefully it has some info you can make use of.
 
@dustyboner, thanks for that link. That is something that I had considered separately from this issue and it's good to know that the EVO guys are doing it. I have been searching around Evolutionm as well but I haven't seen that thread before. If I recall correctly, all three of my trims are being updated during cruise. I have never messed with the trim changeover thresholds but I wonder if Ceddy enabled the high trim on the ROM that he provided. I certainly remember back when I used ECMLink that it was disabled in there by default. I'll have to check that table when I get home. Assuming that the address exists in the xml. Only a handful of tables were mapped by Ceddy before he abandoned development on this hardware.

This would be passable bandaid for getting WOT to lineup with closed loop temporarily, but it doesn't address why the fuel trims always increment to their maximum values and require injector resizing, which is the underlying issue.

Reading into the mod more, it's mostly a way to prevent fuel trims from making dyno runs inconsistent since they change every 4 minutes. This is a road car, so I am more interested in the seasonal safety provided by fuel trims than in being incredibly consistent on a day-to-day basis. It's interesting nonetheless and makes me want to upgrade to the EVO 8 ECU that much more.
 
Well this is embarrassing. The good news is that I found the reason behind this issue. Or A reason, at least. The bad news is that I fell for the oldest trick in the DSM book: boost leaks. I assumed that because my entire charge pipe path was brand new there couldn't be any issue with it. The ugly is that I had a massive boost leak at the J pipe flange and smaller leaks at the BOV flange, the throttle body flange, the BISS, and a coupler. It turns out that on my brand new Punishment Racing intercooler pipe kit, not a single one of my three flanges is flat.

I finally got around to building a boost leak tester today after work and the results were grim. I got everything except the throttle body flange leak sorted out, and that will have to wait until tomorrow. WOT AFR is up from 7.35 to 9.5 and idle fuel trim is down from ~40% to 11%. Closer, but there is still quite a ways to go. I'll keep at it and report back if this completely solves the problem.

I did confirm that the Ceddy defaults had the switch from the Low to Med fuel trim at 88 and the Med -> High switch point at 188, meaning that all three trims were in use by default. I have set it up according to the linked article above where High is enabled at 300 Hz which is above the open loop threshold so the fuel trim at WOT is ALWAYS 0%. I won't leave it this way, but it is immensely helpful for tuning. Thanks again for pointing me in that direction, @dustyboner!
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top