Hi all,

I've been a long time forum creeper, but dont think ive ever posted before.

I have an offroad only jeep that I swapped a chevy 350 motor into and installed a TPI setup onto the engine. It has a TH350 transmission, but a reverse manual valve body, so it doesn't need any input from the engine. I've had the motor/trans in the jeep for about 8 years now. It has always run great and given me low end power. I took it out the other weekend, unloaded it off the trailer, and went to drive away when the issue started. It will barely run after the motor runs for about 2 mins. DIrectly after a cold start though, it will run beautifully! It has a quick and snappy throttle and will slowly increase RPMs in given a steadily increasing throttle. The motor begins to surge uncontrollably and will eventually choke out. If I try to apply steady throttle, it will break up and backfire through the intake.

The TPi setup is from a donor 92 camaro with speed density. I have an external Bosch 044 fuel pump, Bosch III injectors, BBK twin 52mm throttle body, MSD pro billet Dist., MSD offroad ignition box,a Holley adjustable fuel pressure regulator, and EGR block off.

When I hook a scanner up to the jeep, all sensors give me a normal reading and there are no codes stored. I have a FP gauge on the rail and that shows 45psi w/ the vacuum line unhooked.
I ohmed the injectors and they read as follows:
1 - 15.6
2 - 16.1
3 - 15.7
4 - 15.7
5 - 15.7
6 - 15.6
7 - 15.7/15.8
8 - 15.5
I did a leak down test by clamping the return line and the pressure line (after pressurizing) and it dropped about 10 psi over a 20 min. period.

I did just change the o2 from a bosch to the ACDelco sensor, but that didn't have any effect.

I sprayed starting fluid all over the intake base, horns, and upper plenum checking for leaks, and found nothing.

There are only 2 vacuum lines on the motor. One goes to the FP regulator and the other goes to the MAP sensor and both appear to be fine.

I did log the jeep with TunerPro RT yesterday, but am not really sure what Im looking at. I tried to upload that file, but the system wont accept that file extension so I uploaded a screen shot of the graph instead.
Red = BLM
Green = IAC
Blue = RPM

Like I said, everything has been running great for the last few years, until the other day. I'm not sure what happened, but I can't find anything that would suggest the motor should not run properly, so any help or suggestions would be greatly appreciated.

Thanks!

 

Go ahead and send me the TP data. I can take a look at it.

ultm8z@yahoo.com

 

So the one thing that sticks out is that as soon as you enter closed loop operation (approx 2.5 min after start up), your BLM's start going pig rich. Bottoms out at 108 indicating the ECM is detecting a problem that it believes is causing a super rich mixture. When that happens, it'll start pulling fuel out of the mixture like crazy, which is what's likely causing your running problems.

All of the sensor data looks good, which means something must be occuring that the ECM has no visibility on, other than it's resulting impact on the O2 sensor data.

Assuming your O2 sensor is good and hooked up properly...

Probably want to go through the ignition system with a fine tooth comb. Looks like a lot of aftermarket ignition equipment. In terms of plugs and wires, focus on the side of the engine with the O2 sensor on it (ECM doesn't monitor AFR on the side w/o O2).

Check your ignition timing as well.

Was your fuel pressure always 45 psi? If it was originally lower, it could indicate something going wrong with the FPR. Increased fuel pressure would drive down your BLMs.

Lastly, your system voltage seems a little low at ~12.3-12.5V. I don't think that's the cause of your issue, but probably want to make sure your alternator is going south... I'd usually expect 13.5 to 14.5V out of a good running (and properly sized) alternator.

 

Ignition control module/spark control module inside the distributor. Once it heats up it craps out.

 

I have another module in my tool box that Ill swap in and try.
Thanks

 

see above in red

I just sent the injectors out to be cleaned/ leak checked so it will be a few days before I can make any further progress on this.
Thanks for all of your help so far ULTM8Z, I really appreciate it.

 

Yesterday I received my injectors from their cleaning/leak testing from SouthBay. (Great customer service, if you haven't used them before I would recommend them) There were slight differences in the flow rate, but not enough to cause any of these issues and there were no leaks. I have attached the report.

While i had everything apart, I also swapped the HEI module for a new AcDelco unit.

Unfortunately, it still runs like crap.

Tonights plan is to check timing, and plugs. Ill throw the HEI spark checker on each plug wire and make sure the spark is strong enough to jump out of the cone.

 

If you look at it terms of the engine needing 3 things... fuel, spark, and compression....

Fuel pressure and injectors are good. You're experiencing the issue at idle (lowest demand), so it's pretty much now ruled out as a fueling issue.

Spark is most likely the source of the problem IMO.

Compression? Unless you have a bad piston ring or valve seal on one cylinder... if ignition all checks out, maybe do a compression check and/or leak down test on the cylinders on the side of the engine with the O2 sensor.

 

I found some time yesterday to mess with the jeep some more. I wasn't getting a consistent reading from my FP gauge so I checked it again with my fuel pressure tester. I also pulled the AFPR off and checked the diaphragm for rips/tears and obstructions in the passages (something I should have done when I had the rails off the jeep). As usual, everything looked good. The fuel pressure is set at 45psi now, confirmed.

I pulled the plugs, but they all looked fine.

I ran the spark checker on the end of all plugs and have a very solid bright white/blue spark (actually 3 because of the MSD box) on each so that tells me the motor is getting spark. When the motor surges/ dies and I apply throttle, there is sometimes a backfire through the intake so I'm leaning more towards an ignition timing issue. I found a way to test the MSD magnetic pickup in the distributor and also found one in stock at the local Pep Boys to replace it with.

A friend of a friend also suggested checking the injectors with a noid light to make sure none of them are staying open. I have a set, so Ill check that out.

 

I spent about 5 hours messing with the motor over this past weekend and i'm ready to take a flame thrower to the whole thing!

I pulled the dist and checked the magnetic pickup, but it was within spec.
I tested the MSD box per MSD's instructions and it checked out fine.
I re-ran through every piece of the harness looking for any broken wires/ mouse chews but didn't find anything.
I checked alt. voltage @13.9 and battery voltage @12.5. It seems a little low, but I don't think enough to cause all of this.

I have the initial timing close, but cant confirm its perfect. the motor has an aftermarket harmonic balancer which sticks out too far from the timing cover to see the marks behind the water pump. Its as close to 6° as I can get it though. ( i bought a stock replacement balancer im going to swap on to ensure the timing is perfect.)


Amidst messing with everything, I got pissed at it and held the throttle wide open for a few seconds ready for it to blow. After doing so, the motor RAN PERFECTLY! It would idle, slowly throttle up, and not backfire/miss/surge. I had it running for 45mins to an hour. (I was able to log some of the run time if anyone is interested in the data collected.)

Twice while it was idling for the extended period of time, the motor would freak out and almost die. It was extremely sudden and had no warning signs when it was about to happen. It would be idling perfectly, then BAM! the rpms drop, the IAC opens, the motor would stumble, but recover and run for another 20 mins w/o issue.

After that runtime, I shut off the jeep and started picking up tools. I came back to it ~10 mins later and tried to start it, but it would only run for 5-10 seconds, surging bad enough it died every time.

At this point I honestly don't know what to do.
What is the possibility of a bad ECM?

 

I can look at the new data if you want... might be interesting to compare back to the original data and see what might be different.

 

What ECM are you running BTW in case i missed it

 

I just shot you an email, thanks!

 

Disregard the the AFI chip file number.

 

Have you backprobed the ECM terminals with a DMM? If you can show they the ECM is receiving all the correct inputs, it could come down to the ECM. This will help rule out things like bad grounds.

chevythunder.com has the info in terms of what voltages and inputs the ECM is supposed to be getting with key on and key on plus engine running.

 

The one thing that sticks out now is you're getting a code 42 error (EST Monitor Failure). That wasn't there on the previous data.

I will say though, the system voltage still kinda bugs me...

Do me a favor and back probe one of the injector plugs at the pink wire to see what voltage is actually getting to the injectors. Your system voltage is nearly ~14V at the alternator (per your earlier post) and your ECM is reporting that it's seeing ~12V. This discrepancy could be causing the ECM to be incorrectly using the Injector PW correction vs voltage table...

If I'm right, basically the ECM is running in the table range of 11.2 to 12.8V thinking the injectors are seeing voltage in that same range. But the injectors are actually in the 12.8 to 14.4V range based on your alternator voltage. So it's potentially incorrectly delivering a ~.3ms longer PW than necessary thinking your injectors are in the lower voltage range. The lower range gets a 1.2 ms "help" while the higher range only gets a ~0.9ms "help".

Your idle pulse width is currently 2 ms in the data, so .3 ms difference is a whopping ~15% increase in pulse width for injectors that actually don't need it. They're just dumping 15% more fuel in because the ECM is telling them to do that. ECM then looks at the O2 sensor and can't figure out why the engine is running so rich, so the BLM's start dropping.

By comparison, if I were to retune a VE table to correct a BLM of 108, I'd lean out the appropriate VE table cells by ~18% (128/108= 18%). So this voltage discrepancy could explain your issue...

Basically just take one injector plug off and probe the pink wire with a voltmeter with the engine running.

 

No, I have not.
Chevy thunder is an awesome site! I printed out the pinout chart and will check that.

 

Disregard the 42 EST error. I was trying to set base timing and forgot to reconnect the est plug.

I tried to check injector voltage first thing this morning before work, but it wouldn't run at all this morning w/o applied throttle and even then it was running terrible.

I installed the new balancer last night, so I can set the motor to a definitive 6 deg. provided I can get it running now..

 

Just throwing this out there. I had a TPS go bad this past summer. Car ran great, shut it off, and when i went to restart it, it Idled at 3000 rpm's! Shut it off and wouldn't start. No codes, read the TPS voltage and it was off the charts. I know you said the scanner shows all sensors were fine, but just maybe read the sensor with a digital multimeter to rule the TPS out.

 

Yeah his data suggests there's no issue with the TPS or any of his sensors for that matter. The ECM simply seems to be getting a bum input from somewhere else...

 

If you haven't already, the spark plugs need to be replaced as they are carbon fouled.

Get another ECM and try it (1227730). That one is likely going bad with cracked solder joints.

RBob.

 

Thanks for the idea, but the tps checked out with my MM.

 

I haven't, but easily can swap the plugs.

Is there any test I can do to the ecm to ensure that is whats failing? There aren't many junk yards around me to grab another to check with.

 

You can wrap on its case to see if it changes things and it’s able to run.

 

I fired up the torpedo heater and called over my friend who is a heavy diesel mechanic (so I wasn't freezing to death alone ) and we checked voltages on the ecm, injectors, and checked for voltage drops across major wires.

The injector plug voltage was 12.56v on the pink wire.

I back-probed the ECM per the instructions on chevythunder.com and have attached the results. Again, the overall voltage tends to trend low, so I'll pull the alternator today and have it load tested at the parts store.

 

Again out of the blue, it ran for about 20 mins last night. After recording all the necessary voltages from the ECM probing, we began to wiggle every piece of wire we could so see to try to find a break. Knocked on the ecm, wiggled all the connections going into the ecm, the point the engine harness goes through the firewall, dist harness, o2 harness, injectors, tps/iac harness/plugs, and alternator harness.

Nothing seemed to have an effect on the motor. It never repeatably stumbled/cutout when doing so.

 

You're getting battery voltage on all the grounds?

 

Sorry, no. The grounds acted as they should. They were just verified by using a known 12v source.

 

This probably can explain grounding better than I can, http://www.onallcylinders.com/2012/0...ground-issues/ Just something to ponder while your scratching your head. Not that this is your problem, but dirty/bad grounds can cause a whole host of problems.

 

Thanks for the link. I did check for phantom voltages across about everything I could think of as well as voltage drops across lengths of wiring.

I currently have a 2ga grounding wire on the engine block that goes to the frame. At that connection point, another 2ga wire jumps to the fender/firewall bolt and a 3rd 2ga ground wire carries onto the neg. battery terminal. I feel pretty confident in the grounds providing enough for the systems to work.

 

I took the alternator to the parts store and had it checked. They ran it 3x on the machine and it came back as good each time.

When I was reinstalling it I decided to trace the alt. wires as ULTM8Z keeps suggesting a lower than normal voltage across the entire system. Well, it turns out there was never a direct wire from the alternator stud to the battery + terminal. The alt. wiring goes into the vehicle harness and back out through the firewall before terminating on the starter and having to back feed into the battery. This would explain the lower overall voltage level in the system.

To solve this, I made a 6ga wire from the alt. stud to the battery + terminal. I even added a ground from the alt. case to the lower intake manifold. Now when the truck is running, i'm getting 14.6v across the battery! Im not sure how I missed this, or that its been running w/o issue for the past 8ish years.

With the system voltage up overall, I re-probed the pink injector wire and I an now reading 14.43v

After this, I decided to run the truck and see if the problem still existed. Unfortunately, after about 2 mins of running it began to surge uncontrollably again.

It has been said in this thread to check the ECM. I found a company (recommended on another posting on this site) Blue Streak Electronics. They remanufacture various automotive components, but they sell a re-maned ECM that is orderable through Advance Auto Parts. It also has a 3 year warranty and after coupon codes and some speed perks I had, it will cost me $64. So, I ordered it last night and it will hopefully be here by the end of the week.

 

Can you send me another datalog? I'm curious to see all the readings now that the voltage has been fixed. Send me your bin file too if you are able to...

Also, when it's warm and running like crap, unplug the O2 sensor to send it back into open loop (or rather, start it up with the O2 sensor disconnected and see if the it degrades into the surging condition.


If things improve and it runs nicely, that most likely means that it's not something necessarily wrong with the engine. Rather something is fooling the ECM into thinking something is wrong and causing it to think the engine is running overly rich (or it's just the ECM itself).

If it doesn't improve, then it could mean that there's soemthing temperature related (as you said earlier that on a cold start, it starts up and runs great).

 

I use my work laptop for logging so i'll bring it home this evening and capture a few sets.

 

I think I'm coming to the conclusion that there's something wrong with either your O2 sensor or the O2 sensor circuit. Though I think you said you replaced the sensor. So either it's in the harness or in the ECM...

But looking back over all your datalogs, whenever the car runs terrible, the O2 sensor is pegged at 1090 mV.

Now, one might say "well, it's because it's running super rich at 108 BLM and 80 Integrator, so the O2 sensor is pegged rich".

But, oddly enough, on the logs where you've outright disconnected the O2 sensor from the harness, I'm STILL seeing an O2 voltage of 1092 mV.

Something is pulling the O2 sensor voltage high.

The only time the engine runs normally is when in open loop and the O2 sensor sitting at .450 mV (and not being used by the ECM).

Your ECM backprobe data seems to indicate the O2 sensor wiring is good.

Maybe the ECM somehow has gone bad in this regard. It'll be interesting to see what your new ECM does.

 

Yes, the sensor is new as of 11/29.

I was able to log a cold run before work this morning with the o2 unplugged before starting the motor. Interestingly, the o2 starts at 450.84 until ~4 seconds in when it spikes to 1087.32 and dwells there for almost 6 mins. where it drops back down to 450 and the motor dies ~20 seconds later.

The o2 circuit is very short. I basically goes from the ecm (mounted high on the inside of the firewall) down the front of the firewall to the o2 plug. The total harness length must only be 3'.

Is the o2 voltage supposed to be variable between the min. and max.?

 

How is the engine behaving during the time the O2 voltage spikes to 1087 mV?

 

It ran ok, but not perfect. The IAC would fluctuate periodically and the idle would be mostly steady.

 

Hmmm... yeah that makes me think something may be wrong in the ECM then.

I don't know what the output reading should be in TP for a disconnected O2 sensor. But even if it's defaulting to .450mV initially, there wouldn't be any reason for TP to report a any change in O2 voltage since the sensor is physically disconnected from the ECM.

Also, being in forced open loop (with o2 disconnected), there's no reason the ECM should even be reacting to changes in O2 voltage (whether small or large change) in the first place.

If you had a short in the O2 wire in the ECM harness, that would pull the voltage down, not up.

So yeah, it'll be interesting to see what happens with the new ECM.

So AFIK, this 1V spike on the O2 voltage has no other explanation (unless I"m just not thinking of something).

If you were really curious, you could put a volt meter on the O2 sensor and see what it's doing... though I think that'd be academic at this point.... whatever it says, I don't think it would change my diagnosis based on what I'm seeing/hearing.

 

91 Pontiac Firebird 305 5.0 V8 TBI sorry for spamming you guys no idea how to post stuf here am new but need help ASAP ������

Hi there for the last 3 days I have the exact same problem my battery is 2 years old and I drove the car no more than 5 times I did start it up every month for 30 min since it’s a project car but my point is 3 days ago I wanted to move my car to the sec garage with heat in it so I can clean ntinue working on it

and the battery was completely dead so I hooked up a charger and charged it for about 90minm

it looked all good and I started it right up let it run for 40min went shopping and got gas I shut it of filled with new gas and went home next day I tried to start car was dead so I jump started it but cranked started and died than it wouldn’t even crank anymore only the security lights that stays on and when I try to crank it the red park and brake light comes on fuel gage goes all the way but no crank some klick under the hood and steering column

no blown fuses vat system my key is a 10 with 3.750

but when I checked the connector under the dash the wires read the key resistance from the ignition key

but anyway I did get resistors and got it almos perfect to that what the key is

nothing security light still solid on

when I open up the starter relay on the side by the foot wall and press on it and turn the key it will crank and start but die after 2 sec and when I jump it with wires no relay same story dies after 3 sec I did notice my battery now after being charged up all 100% disconnected from the car it will drop down to about 70 to 60% of charge wierd the bat is basically new fuel pump is working to I even tried to jump the safety switch but no results pls help I have no idea what the problem is the car is a 91 Pontiac Firebird 305v8TBI with 183xxx it cool st me well over 10K by now since I got it for 2.7K it’s a Knight Right det Replica that’s why such costs thanks guys for the help I n advance ����

 

Remove the cover from the ECM and start the engine. Gently press on each chip/area of the board, many times can narrow down exactly which chip has the cracked solder joint(s). Chip U5, about center of board, is the analog to digital convertor (ADC). I would start with that one.

Then the larger chips to the left & right of the MEMCAL socket.

Be sure to isolate the ECM both electrically and vibration wise from the chassis. The ECM case is already grounded via the harness to the engine block.

Not really, but should oscillate when in closed loop.

RBob.

 

Wow some of you guys are really overthinking this lol...

Stare at the first datalog you posted. Way before the engine enters Closed Loop, way before the BLM drops, and way before the idle becomes erratic, what is it that deviates from it's normal range during Open Loop and stays pegged in turn causing the other two in the graph to adjust for it when Closed Loop is enabled? Take a good look...

- Rob

 

Ok, I give up...

I do know that in some of his datalogs he sent me, I saw the BLM pegged at 108 during open loop (with the integrator at 128) which was quite odd as well...

 

From the first datalog we see a cold start, and this assumption is based on the data showing his engine entering Closed Loop a tad above 100 seconds, whereas a warm start would be closer to 30 seconds. Anyway, his IAC steps are approximately 20 steps during cold start which is already way off target, should initially be closer to 145 steps.

During Open Loop the IAC steps slowly increase and will inevitable peg in Closed Loop. As Closed Loop enables after the 100 second mark, O2 correction is activated, and BLM's drop to compensate for the increased air from the IAC, and the idle becomes erratic. The way the OP explained it was that the BLM and O2 fault first than the IAC, but it is the other way around...

- Rob

 

why does the IAC have to be 145? It depends on where the throttle blade are set.

also the IAC isn't "increased" or extra air... it' part of the overall airflow into the engine.

 

Why would the throttle blades be that far open running an IAC? It would defeat the purpose. As for 145 steps, this is why I mentioned it was a cold start, as a warm start would have the IAC steps closer to the 20-30 steps as the target and CLosed Loop would have been entered closer to 30 seconds. His plugs do not look like they were running rich despite the low BLM, the O2 correction has that spectrum covered anyway and will correct up to a 6% variance in fueling, and he stated that he didn't touch anything and that the engine suddenly started acting up. His injectors were checked, his fuel pressure was confirmed, and I am assuming his fuel pressure regulator has no tear in the diaphragm. It simply points to that area.

- Rob

 

Umm, no. It indicates a problem. BLM is locked at 128 and deviates to 108 after the IAC is already pegged. I am forced to disagree with your infinite wisdom...

- Rob

 

That's one thing I don't understand around here sometimes... the snarkyness in what is a purely technical discussion.

people can't even carry on a purely technical discussion without it devolving into a pissing contest.

After having taken the time to review several of his datalogs, I was deliberately taking the discussion with the OP into here as opposed to privately over email since (even if my diagnosis is wrong) the diagnostic procedure may help other people with searches who may be experiencing a similar issue.

But no worries... since i dont want to offend you with my "infinite wisdom" I'll just go back to discussing my opinions over email with the OP.

 

Excuse me? If I read correctly, you quoted me and your response employed "quotes" in which sarcasm on your part was more than prevalent. I did not post in here to argue with you, or to debate the characteristics of an OBD1 ECM. I simply posted in here to help this member because he seems very frustrated and it's been going on for almost a month. Again, assuming no tear in the regulator's diaphragm, a pegged IAC points to not enough air to compensate for the fueling at idle, or the IAC is simply defective and is stuck open. If it were the throttle blades being too far open causing the IAC to skew, the IAC would close down to zero. Anyways, I'm done. This should have been resolved almost a month ago...

- Rob

 

Hmmm.... I guess I'm surprised that putting a word in quotes somehow implies sarcasm...

Oh well, I guess stuff gets written down one way by one person and gets taken another way by someone else...

All I'm saying is the MAP sensor (which is what the ECM relies on for inferring airflow) doesn't know or care where the air is coming from (i.e., throttle blades, IAC, a vacuum leak, etc). With the total air flow combined from all these sources, you're going to end up at an RPM and a MAP value. Which is why I'm saying I don't see the BLM reacting differently if more air is coming from the IAC or the throttle blades... air is air.

but I know for sure that it'l react to a faulty o2 sensor reading once in closed loop. The reason the plugs don't look rich is because the engine is not running rich. It' running too lean because the ECM is pulling fuel out for no reason except a faulty O2 signal IMO.

 

Is the log file even attached on this thread ? Maybe im missing it