wendy-lad Posted February 6 Report Share Posted February 6 Hey guys, Got a turbocharged Holden 304 and haven't had a problem for over 2 years (5 to 10 000kms) with the AFR doing anything weird. It's the longest I've had a wideband sensor last so far. The tune has been fine that whole time running around 11.3 at full boost. Basically overnight the AFR dropped to 10.0 or below at the same full boost/throttle, so dud sensor giving a false reading I thought. There's no misfiring or massive soot cloud out the back, it pulls just as cleanly as it always did. Swapped a new sensor in and get exactly the same 9.85 flatline AFR. The car would've idled and I gently drove it for probably 2 minutes before the first full throttle pull, so surely didn't foul the new sensor already... I did check and clean the wiring connectors/plugs coming out of the can lambda module, nothing looks out of place anywhere. The power and ground for the module are still firmly connected to the terminal posts they've always been. I've made sure the sensor is screwed in fully and tightened the v-band coming off the turbo. Does anyone have any other ideas? Log file: https://www.dropbox.com/scl/fi/j1s1yc66obphl1u1okvit/rich-afr-9.8.llgx?rlkey=as0w5rfva4il0v55b8hl5lvu6&dl=0 Tune file: https://www.dropbox.com/s/p0e4bhadao0thz9/holden 304 crank trigger.pclx?dl=0 Thanks Quote Link to comment Share on other sites More sharing options...
Adamw Posted February 6 Report Share Posted February 6 How confident are you that the new sensor is a genuine bosch part? Initially nothing in your log gave me the impression that there was any issue with the lambda reading, it sits on target 90% of the time and only drifts rich at higher loads so I was leaning more towards it being something like a MAP sensor or fuel press issue. But I do notice about halfway through your log the lambda reports error 54, which although is something that could occur from a very rich mixture, I have only personally ever seen that error from a broken or fake sensor. So I do have a little suspicion. Do you have any older logs from a time when lambda looked more normal so we can compare PW's etc? Quote Link to comment Share on other sites More sharing options...
wendy-lad Posted February 7 Author Report Share Posted February 7 Hey Adam, This is a log from when it's been running fine for that 2+ year period: https://www.dropbox.com/scl/fi/rpd2skziw34bsr7a6rdab/888.llgx?rlkey=apag883gifyerwiyyzebjmydg&dl=0 Tunewise change between those 2 log files are the knock sensors were added, and I set the 4D added fuel table back to basically zero cause I initially believed the sensor was reading true and very rich, so started pulling fuel out of the 4D cause that was easiest way to test and it didn't change the AFR reading one bit. It's a fuel temp compensation so only ever added up to 7%ish anyway. The sensor was from the local EFI Automotive store here for $140 so pretty confident it's genuine, both the new and 2 year old sensor. Thanks mate Quote Link to comment Share on other sites More sharing options...
Adamw Posted February 7 Report Share Posted February 7 That is the same log file as your previous post. Quote Link to comment Share on other sites More sharing options...
wendy-lad Posted February 7 Author Report Share Posted February 7 5 hours ago, Adamw said: That is the same log file as your previous post. Whoops sorry mate, I edited the post above for the correct file. Thanks Quote Link to comment Share on other sites More sharing options...
Adamw Posted February 7 Report Share Posted February 7 I compared a couple of points with similar operating conditions in the before and after logs and the Injector PW's look pretty similar so I tend to agree that it seems the lambda values in at least one of the logs doesnt seem right. There are some fuel trims that dont show in the Inj PW runtime such as async fuel, knock trims, CLL, etc, so I have possibly missed something, or something has since been turned off in the recent map, but I dont see anything obvious. One further thought is you are on quite old firmware and there have since been a couple of fixes which may be having an influence on the lambda measurement. The exhaust press compensation value in the notes below is the main one that springs to mind, although I dont remember it having a big effect when testing and its hard to imagine how it would cause a change 2 years later (BAP is a little different), so it would be worth updating the firmware to rule those possibilities out. Quote Link to comment Share on other sites More sharing options...
wendy-lad Posted February 15 Author Report Share Posted February 15 Hey mate, I updated the firmware and it didn't change the AFR. Though I did end up borrowing the local tuners old analogue wideband controller setup he uses to road tune cars with, where the sensor clamps up inside the tailpipe. And it showed a 10.0-ish AFR under full load too, so that's his sensor, my new sensor, and my 2 year old sensor all displaying the same AFR, so I'm 99% sure it's just a rich condition that's come on for some unknown reason. I'm gonna put new plugs in, maybe it's just a hint of misfire under full load that I can't feel, but is enough to produce a rich AFR reading? Failing that, possibly an injector or 2 leaking more fuel in than it should, but only under full load? I'll get them flow tested if the new plugs don't change anything. Will report back with results. Thanks Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.