3.4 TDC trials and tribulations (Page 1/1)
tjingers JAN 30, 05:47 PM
So i think i've done enough homework to warrant a "help me please" post...

I have an 87 that used to have a 2.5, that's gone now. I bought a 92 3.4 TDC from someone else's 88 with the wiring, ecm, etc since it was available and i was going to do a 3.4 swap anyway. 5sp 282 is the trans that was attached, and that i'm using.
Fast forward a few months, i've got it installed with WCF mounts, a new 4th mount, etc, wired in and everything was going well. Engine ran, idled, revved, and i had heat cycled it a few times after doing the coolant plumbing. Engine ran well and fired right up as I expected every time.
Speedo works, tach works, coolant temp works, oil pressure works, etc. Have to press the clutch to start so i'm assuming that wiring is good too.

First drive: made it about 1/4 of a mile, and it stopped running. Found out i had forgotten to tie down part of the loom and rubbed through the knock sensor wire on the crank pulley. The crank sensor wires also got dusted, but not severed. I repaired all these wires shortly after.
This drive was at most 30 mph, maybe i got up to about 40 on the road just outside my neighborhood before it died. Was already going to take it easy due to new clutch.

After the "incident", the engine no longer runs at all below 1000 rpm, usually 1300. It starts very hard and yet will run very well above 1300 rpm. When cold it idles nicely at the highest "step" for about 15 seconds, but once it tries to drop it just wheezes and dies. I can usually keep it alive with throttle but every time it dips back down it does the same gasp and dies.
It almost seemed like a vacuum leak, but i've already gone through and replaced the upper intake gasket, plugged all manifold ports (both near TB, and the brake booster one). The map sensor manifold is in good condition and has all o-rings and seals. I have sprayed around everything with carb clean (it's a struggle, with the engine not idling on its own) and i can't find any kind of vac leak. The throttle body is set just barely before it "parks" on the blade itself.

What I have verified:
knock sensor: appropriate resistance at ~3.9k, scoped the wire to ecm and it pulls down to 2.5v as expected with key on, from 5v if i unplug the harness.
crank sensor: scoped AC waveform and it looks fine. pulse from ignition module to ECM looks fine.
coolant temp sensor: verified via tunerpro that temps are all plausible, reflects ambient temp after sitting all night.
injectors: these are about 6 months old and only have a few hours on them. They're brand new from the previous owner replacing for the classic 3.4 dying at idle under light throttle *sometimes* issue (only in reverse, for him). I've not had that issue.
TPS: 0-100%, confirmed with tunerpro and my foot
IAC: performed the reset procedure several times (short aldl, let run, unplug connector, key off, plug back in, key on, etc), it shows what gmtuners says are normal counts for start/warmup (110-143 is what i see). Very clean and seems to work fine.
injector pulses: scope shows a nice pull to ground as it runs at high rpm. Plus, it runs at high rpm, so the drivers don't seem shot.
MAP sensor: looks fine, tunerpro shows 4.5 ish volts at atmosphere. It drops to whatever other voltage when vacuum starts (i have logs...)
ignition module: verified distributor pulse to ecm, verified good spark, verified spark at cranking speed, verified ECM is "taking over" with bypass signal, replaced entire ignition module with brand new one. Verified that ECM is sending a pwm signal once it kicks on the bypass.
Spark in general: looks fine. I can pull a plug and get a bright spark when cranking. Tried all 3 coils this way.
o2 sensor: verified normal cold voltage, i haven't been able to get into closed loop yet though since the incident, though, so not sure this matters at all.
All sensors have been swapped with spares i have from my spare engine which came with... every time i try one they behave the same.
ECM:
This ecm did have a tune from sinister on it. I ripped the .bin and compared to the stock BCFA bin, mostly just EGR delete changes. I can upload both as examples. Seems unlikely to be an issue since it ran great with this tune before the "incident".
Verified all sensors wires are continuous and that tunerpro logs will show the data (seems good enough if the ECU can report what it thinks the sensors are...)
Verified no burned traces, etc, but only visually.
Emulated the stock bin using moates auto prom. No change in behavior.
Reset ECM multiple times
I have been getting a code 43 (ESC) regularly, but i don't ever see knock count increment except while cranking. Once the engine is running, it never increments. I also never see knock retard > 0.00. I have a good strong dizzy pulse signal as verified by the scope and the "engine rpm" showing up just fine in my monitors in tunerpro.
Knock filter appears to be intact with no burned traces. I don't claim to know how to test these, but there is plausible resistance across the terminals that people measure to compare tuned frequency (pcmhacking forum posts is all i have to go off here).
Target idle speed is normal (1300 or higher usually, since it's not exactly having a good time at the moment).


Unfortunately i didn't have any logging / etc before the incident, i bought all this gear to help me figure this out.

Does anyone here have any ideas? I'm pretty well losing my mind here...
The only thing i can think to do from here is buy a reman ECM but i'll have to wait until payday (turns out buying all these nice things from moates and a handheld scope, parts, etc is ruining my project car budget).
The one i have is the normal 9396 that came in the 92 lumina.

I appreciate any insight. I can provide log files (not sure how... can you upload files here?) of what the behavior is like, if you have tunerpro and are interested.
tjingers FEB 06, 05:35 AM
Well i'll throw the answer in here for any time travelers having the same issue.
I finally decided i needed numbers for my fuel pressure and it was the best thing I could have done... rookie mistake, i guess. I had 5 psi of fuel pressure at the rail with the pump jumpered on.
The piece of rubber hose from the fuel pump to the sender was blown out, so all the pressure was being lost before it even left the tank. This was causing literally all of my issues and everything runs fine now. I measured 38 psi at the rail with the pump jumpered on, car not running.
The audible difference between before and after replacing that piece of hose was pretty big (you could hear the fuel pump making a "gushing' noise before, now it just sounds like any other electric motor).
no2pencil FEB 06, 08:15 AM
Glad you got it figure out. Got any photos of your final install?
Dennis LaGrua FEB 06, 08:58 AM

quote
Originally posted by tjingers:


The piece of rubber hose from the fuel pump to the sender was blown out, so all the pressure was being lost before it even left the tank. This was causing literally all of my issues and everything runs fine now. I measured 38 psi at the rail with the pump jumpered on, car not running.
The audible difference between before and after replacing that piece of hose was pretty big (you could hear the fuel pump making a "gushing' noise before, now it just sounds like any other electric motor).


I hope you replaced that in tank hose with SAE J30R10 "submersible fuel injection line" or you will be doing that job again.

------------------
" THE BLACK PARALYZER" -87GT 3800SC Series III engine, custom ZZP /Frozen Boost Intercooler setup, 3.4" Pulley, Northstar TB, LS1 MAF, 3" Spintech/Hedman Exhaust, P-log Manifold, Autolite 104's, MSD wires, Custom CAI, 4T65eHD w. custom axles, Champion Radiator, S10 Brake Booster, HP Tuners VCM Suite.
"THE COLUSSUS"
87GT - ALL OUT 3.4L Turbocharged engine, Garrett Hybrid Turbo, MSD ign., modified TH125H
" ON THE LOOSE WITHOUT THE JUICE "

Raydar FEB 06, 09:10 PM
Thanks for the update. I'm about to start an LQ1 swap, sometime this spring, and I like to hear success stories.

------------------
Raydar
88 Formula IMSA Fastback. 4.9, NVG T550

Praise the Lowered!

tjingers FEB 06, 10:26 PM

quote
Originally posted by Dennis LaGrua:

I hope you replaced that in tank hose with SAE J30R10 "submersible fuel injection line" or you will be doing that job again.





Yeah... i'm a genius so i just didn't replace the hose when i put the correct pump in (i figure the TBI pump wouldn't have cut it... don't have any reason to believe that but i had the correct pump so why not)
Anyway, the new delphi pump had the replacement piece included and i installed that when i fixed it just now. Should be good for at least a few years.

I'll get some pictures soon enough.. there's still some trimming i have to do to the trunk lid to make it all close properly and i'm just SLAMMED with day job stuff right now.
Notorio FEB 11, 04:20 PM
This is another fine example of what seems to have caused the problem (i.e. wire loom damage) that in the end actually had nothing to do with it (i.e. blown hose). Congratulations on figuring out the real solution!