Lambda Trim not showing up?

[Michael Smith]

Hello... I noticed that when recording my 1994 RRC that Lambda Trim is showing up as zero... I can see short term trim moving with the software, and last run I did, long term was a little negative as I recall...  But it doesn't show up in the log.  Attached is a file.

This is great software and it's been super useful.  Thanks!  I also have a suggestion if it is possible... Is there a way to reset all adaptive values?  That would be great to make a change then have the car re-adapt.  I recently adapted a Toyota TPS to the 4.2, and so far, so good.  The vehicle still hesitated on throttle up this morning, but later in the day didn't at all, so it may still be adapting after having a bad TPS (non-linear, with gaps) for so long.  I suspect is was fueling to a wrong map offset due to the TPS, if that is possible, though that would make it pretty smart and perhaps more adaptive than it really is.  It would be great to reset it all and see what happens.  I did disconnect the battery for 1/2 hour, which may also reset the ECU...

Any insights you may have to hesitation on mild acceleration would be very helpful!

Thanks for the help,

Mike S.

[colinb...@gmail.com]

Hi Mike,

The 14CUX isn't as sophisticated as more modern ECUs, so it doesn't store many values that it learns while running. The throttle position sensor is indeed assumed to be linear and monotonic, and as far as I know, the firmware does not attempt to compensate for any noise/cutoff in the TPS profile.

Removing battery power for a short period of time is sufficient to reset any stored data in the ECU.

I'm afraid I can't speak to the long term lambda issue that you see; the lambda feedback loop is one of the more complex subsystems in the code. You might want to try asking on PistonHeads. There's been a lot of discussion about 14CUX lambda in this thread and others:

http://www.pistonheads.com/GASSING/topic.asp?h=0&t=1325176&mid=0

--Colin

[Michael Smith]

Thanks Colin,

I tried a smaller recording set (only the things I suspect may be important variables to a hesitation problem, like RPM, AFM, throttle, and lambda) and now I get the lambda readings recorded, and the recording interval is much shorter, and this has indicated a rich condition on the left bank... More digging tomorrow, but it appears that some values are not written as text if perhaps other priorities or too many fields are selected.  Strangely, with the smaller selection set, a value that used to record, PulseWidthMS, is now always 32.768 (2^15/1000), when before it was around 3ms at idle...  Maybe it's a bug?

Here is some data (one with small dataset, the other with full data, some of which doesn't record).  You can specify what variables and a factor on the right, and use scroll bars to zoom in...

Thanks again,

Mike Smith

[Michael Smith]

I've also noticed several instances where with rovergauge connected, I will get a check engine light, and it can't reset it.  This happens when there was no MIL to start with (I was checking other things or recording a sequence).  A few tries, restarts, changing options, and trying again, usually with engine off, it will reset it.  There are no codes present.  I haven't figured out why sometimes it will reset, and others it won't.  Without the computer connected, I don't get the check engine light.  I wonder whether sometime during communication, something gets garbled and sets the MIL without a code?  I think this usually happens after I turn off a recording, though sometimes I've made multiple recordings with no issues. 

I think I solved the hesitation issue... The coil had a big eroded area near the HT lead, and it looked expanded.  I was seeing a large deviation in left bank trim, but it wasn't there today.  I think it was misfiring.  Replaced the coil.  So far no issues after a few miles...  What a great tool this is to see the engine operating details (much needed by geeks like me)...  Thanks so much again.

Mike S.