WTF? A truly great idle mystery?!?
 

Ok, here's your guys opportunity to solve a really good riddle. 🤔

My car is a '92 with a fully built 1.8 swap from a '96. Garrett 2560r, running MS2 built by braineack. Car has been dyno tuned and made 290/291 last year. Ran pretty well all fall, about 1500 miles on it to date.

Anything listed as brand new has been replaced in the last week trying to nail this down:
-sequential injection ecu and harness mod
-Fuel Injector Clinic 650's (1500 miles, went in with build)
-Walbro 190lph hp (brand new)
-Wix fuel filter (brand new) - this was actually done days before the issue arose.
-Ngk bkr7e gapped at .031 (brand new)
-Ngk blue wires (brand new)
-1.6 coils
-Beck arnley oem FPR (brand new)
-valve cover gasket (brand new) - this was actually done days before the issue arose.
-Bosch 30-2001 LUS 4.2 wideband sensor for the AEM UEGO wideband (SENSOR is brand new, gauge is original).
-DIY mxp4250 2.5 bar MAP
-DIY electronic boost control

Issue:
Coming home from Warren a few weekends ago just cruising in 5th at 70, uego suddenly with no reason started reading SUPER lean. Pulled over, car idling like crap. Limped it a few miles to next exit, car died waiting to turn onto road. Fired it up, limped to station a few hundred feet away, and it started idling beautifully. I figure bad gas or something, because i was down to 1/8th tank, so I top it off while it's running and put some Injector cleaner in. Car drives beautifully all the way home including several pulls back in Mexico.

Next day, runs great including several pulls. Get about 1/2 mile from my house, gauge goes super lean again and I limp it home. No power, feels like it's cutting out with throttle. Trying to accelerate there's just no guts.

I start replacing ignition and fueling items, see above. Take it over to my buddies shop and the whole way there it's running showing lean and drivability is crap. Every once in a while the gauge will read normal AFR's, car will run great, and then it'll go lean again.

I've replaced all the parts listed above that read new within the last week.

Things I've looked at in tuner studio and/or troubleshooting:
1. Coolant temp sensor is reading correctly.
2. MAP reads near 100kpa with car keyed on but not running. I've noticed that at idle the target kpa fuel load in the table is 52 and I'm seeing around 32 in tuner studio gauges.
3. I've tried a second set of oem coils, no change.
4. Afr in tuner studio matches what my AEM gauge is reading (essentially pegged lean). If i blip throttle i see gauge and TS matching.
5. My tuner said that the ecu shows it's adding 15% fueling in the idle logs, which should be substantial but I'm not seeing anything reflect that in my AFR's. Smells rich though...
6. Someone else pointed out that im seeing LOST SYNC REASON 38 show up in several of the idle logs...which is 4g63 CAS.
7. I have 40psi on the feed line and pressure increases immediately with blipping throttle input. Tank is spotless inside.
8. Pulled plug wires one by one while the car was idling, and cylinder one wire pulled out of the tube caused NO IDLE CHANGE. Again, i've tested this with a second coil pack.
9. Turned EGO control completely off, idle was lower rpms and didn't improve in quality at all.
10. Plugs are already showing rich characteristics, and the cylinders smell strongly of unburnt fuel.
11. spark plug tubes are dry, no oil leakage

I'm completely stumped. All my vac lines look good, grounds in the engine are still connected and i even cleaned them up again. Boost gauge is showing -20 vacuum solid at idle. Literally something decided to fail on the highway in a split second and I cannot understand or find what it is. Nothing with my tune has changed. I can't even take the car out to drive and get a log at this point, trying to back it out of the shop or pull it up onto blocks to get the lift arms underneath it requires tremendous throttle input with broken/non existent power in response.

CAS issue? I don't think my board has the second opto circuit for the CAS but the car ran fine before so....?

Attaching some recent idle logs. Glad to capture/post anything that will be of further help.

Thanks in advance guys!!

Over 70 views so far and no guesses haha. See, I told you guys this was a tough one

Post a tune also please.

My '90 did that with a bad wideband sensor. And something kind of similar in my '994 when the CAS harness was loose at the connector.

How about an intermittent injector connection. Next time it stumbles and runs lean play with the injector connectors, and harness. Maybe you have a bad crimp in your harness.

Also when I had a bad wire in my ignition harness it manifested itself with a misfire and the gauge going lean. The unburnt fuel wasn't registering and it would go lean.

So next time it goes bad try and play with connectors, wiggle harnesses, maybe unplug each coil, and see if unplugging one coil at a time changes anything.

Will post it when i get home after work.

I did just replace the sensor itself a few days ago, no change. The gauge itself is unchanged though...perhaps bad gauge ground or something? When I blip throttle, i can get it to start to read richer and it matches gauge reading to what I see in my gauge on tuner studio.

That's a good idea, I'll try that with wigging the harness for both the Injectors and the CAS. Both of them have been modified with the swap/build. The CAS wiring obviously had to be extended since the CAS is on the other side of the head, and the injector harness had to be modified to accept the two new signal wires. So odd that it would have functioned without hiccup for this long and then just BAM, failure.

My vote is for the CAS wiring. Friend's car did exact same thing with intermittent connection.

Would intermittent connection cause a cylinder to not even be firing? Like i said, pulling ngk wire from cylinder 1 had zero impact on idle or causing stumbling. The other three cylinders caused drop off.

Am I right in understanding that you're saying you pulled the spark plug wire for cyl. 1, and it appeared to have no effect on a rough idle? Are you running batch ignition? If so, I'm inclined to think that you're actually having a fuel problem. You can run injector test mode to see if the cyl 1 injector is getting signal (clicks like a mofo). If it's getting signal, try pulling it to see if it's actually squirting fuel. A clogged injector seems like the type of thing that could suddenly happen while cruising.

Lol i totally missed that. Ok so use megasquirt test mode to individually test the spark and fuel on that cylinder.

Car is no longer on batch injection, It's sequential injection. It's still on batch ignition.

Exactly. Batch spark combined with only one cylinder acting up. You already changed plug wires, which makes me think it's actually the injector. Injector test mode to see if it clicks, then pull it to see if it squirts.

I can test tonight. Can you share the Injector Test Mode settings I should use for testing just cylinder 1? I obviously don't want to soak the cylinder in fuel more than i have to.

To pull, I'm assuming you mean the entire rail with all injectors in it...finding a way to keep injectors 2-4 in place and not getting pushed out of the rail. Then run test on just cylinder one again to watch spray patter/flow into a container?

Sorry, I only used it once to make sure I'd wired for sequential correctly, but there is a bunch of info out there including videos. I see the googles in your future...

Of course, you could always go old school & swap the injector between cyl 1 & another cyl to see if that cylinder then stops running. If the problem moves, then it's a bad injector. If the problem doesn't move, then it MAY be bad injector wiring.

I'll take a look. Could a faulty CAS signal as mentioned cause cylinder one alone to not be firing? If it's not registering 1 at TDC or there's a wiring issue leading to the CAS itself, are the symptoms im seeing on #1 possible?

Oxygen sensors sense oxygen, not fuel. Miss = unused air = too much air.

i think your AEM is bad...

fueling looks perfect on those logs at idle, so if nothing has changed, i think the sensor is going crazy.


https://cimg2.ibsrv.net/gimg/www.mia...3f2e6a8bb1.png

that's pegged full lean, but you're fueling exactly like your map demands. if it's not actually shitty idle, id think the aem is false reporting.

compared to the first log:

https://cimg7.ibsrv.net/gimg/www.mia...0896eb76b5.png




I'd check for an exhaust leak...

actually read a little, sounds like it's driving/idling like shit.


and yeah i see the lost sync:

https://cimg5.ibsrv.net/gimg/www.mia...1333a4c53a.png



can you check to see if your timing marks still line up to TDC?