Eisen7

I'm not having much luck with this ECU tuning caper... first my ETACS unit dies, then my front O2 sensor, now this

Basically all that happens is that I am driving along at normal load, AFR's all look nice at cruise and then all of a sudden, AFR's will shoot up into the 20's, the CEL comes on and stays on and the car splutters to a halt choking because it's way too lean. Not ideal in traffic! It's happened twice in the last few days after driving the car maybe 10 times.

To fix it, all I've been doing is switching the car off for a minute or two, then restarting. Then it all seems to behave again Perhaps my ECU is running Windows 95?

As said above, the only things I've changed recently are the ETACS ECU and my front O2 sensor (using a universal 4 wire one which needs to be joined to the existing wires). All I'm trying to do is sort out my injector scaling and latency and get the trims as close to 0 as I can. Then I can move on to some power runs.

Attached is a log where the issue happened. The problem happened at LogID 247. Note that I keyed off and rolled to a stop towards the end, so you will see some of the data stops logging, but the car is still moving.

Any ideas? The car is a 2001 JDM Evo 7 running a 90550001 ROM with TephraMod 5.1. I'm also using an LC-1 via serial cable for wideband logging. I can post the ROM if needed.

EvoScanDataLog_2010.09.09_Fault.xls.zip

tscompusa2

iTrader: (22)

Are your fuel trims back to normal yet? is it occuring when you go WOT 20 a/f ? Sounds like you might be out of injector or fuel pump.. be careful running that lean is not safe.

colonelfox

iTrader: (41)

What code is coming up? I would suspect a faulty fuel pump. Maybe the wiring harness isn't fully seated (this happened to me once before) or the pump itself is going bad. It doesn't sound tune related to me because it's not consistent.

Eisen7

No it's not happening when I go WOT or anything. The AFR's are always pretty close to what's on my inj map, so it's nice and safe normally. This problem has just occurred a couple of times when cruising on light load (where AFR's are usually showing about 14.7 on the WB02). It's like one of my spark leads has been pulled - you lose all power and the car coughs and splutters away going ultra lean. Then I stop and restart and it's fine again. It's really bizarre

That's the other thing, it's not actually logging an error code. I tried to check DTC's and it's not returning anything. But the CEL is lit up constantly while this issue occurs. Then it goes out again when I restart.

l2r99gst

iTrader: (2)

Looks like your car is dropping out of closed loop right when it happens. Are you sure the O2 sensor that you replaced is working correctly? I have heard a lot of bad things about cheap replacements.

Anyway, with your trims being so far off, that instant you drop out of closed loop, you're actually adding like 11% or fuel. Your car may be misfiring as soon as that happens. From your log, your STFT dtops out to 0 (well .1 in your logs, looks like you have the wrong equation).

That's the only thing that sticks out to me in the log.

Fast_Freddie

iTrader: (50)

If you used the Bosch universal 4 wire, then this can definitely cause some issues...

Eisen7

It looks like the new O2 is functioning correctly. As you can see in the logs the voltage is fluctuating nicely between 0.1 and 0.9, but when the issue happens and it goes lean, it goes to 0.0x

Might have to try a new O2 if it persists. This is the one I am using. It's my 2nd one now- the first one didn't have any issues like this, it just didn't last all that long.

http://cgi.ebay.com.au/OXYGEN-SENSOR...item20b322d003

colonelfox

iTrader: (41)

I've never heard of that kind of reaction due to a faulty O2 sensor. I run a universal sensor myself and have installed them on several Evos and none have had a problem. The worst that would happen (assuming you didn't wire it incorrectly and it's not grounding out somewhere against the engine) is that you get bad gas mileage, fouled plugs and a CE light. It won't lead to the car shutting down like that.

Eisen7

Yes that's what I would have thought. Surely the O2 can't make a car behave like this?

I've just realised there was one other change to the car at the same time I installed the new O2. I fixed up one of the wires on my loom where they had messed up the AVC-R boost controller install (years ago). For some reason they completely cut the speed signal wire and wired that into the AVC-R - so my ECU was never getting a road speed signal. The other night I joined the speed signal wire back to my ECU and spliced in the white AVC-R wire so it can share the signal (instead of completely taking it from the ECU )

I checked the wiring last night and there's no exposed wires or anything... Just wondering if it could be something to do with that?

Eisen7

Well I logged the car for hours over the weekend sorting out my idle/cruise trims and injector scaling, and the issue never occured. I also did a few WOT runs to check the tune and AFR's. I thought 'great it's all sorted!' Nope

Driving to work this morning in traffic, It happened again. I was on a 2 lane road full of traffic, so ended up keying off and then back to on while still rolling along and the car ran normally again. So frustrating!

Eisen7

I just found this thread and thought I should post that the issue is resolved. It turned out to be the infamous P0300 misfire issue.

I've disabled the P0300 test via periphery as per https://www.evolutionm.net/forums/ec...0300-test.html and the car has been fine ever since