Poor idle and misfire at high speeds

[metallic95]

Hello everyone, new here.

I'm working on a family members 1995 cutlass supreme with the 3.1 L.

At idle, it's got a rough idle and engine vibration. Not like a crazy vibration but enough you can feel it while sitting at a stoplight. In addition, if you suddenly hit the gas pedal to accelerate at higher speeds(around 50 mph) the engine shudders and it feels like the car has no power. If you push the pedal slowly, it tends to do it less.

 

So far, I've changed the spark plugs, cleaned out the EGR and EGR passageways, cleaned the throttle body, and checked all the vacuum lines for leaks or cracks (found none). I inspected the spark plug wires for breaks or cuts, but found none. I used a multi meter to check the resistance of each wire, and I believe one was out of range from the service manual spec. Could that be causing the issue? It was far past 30,000 ohms. I'll check again.

 

I'm pulling my hair out with this one as they need the car soon, and I'm not too keen on taking it to a mechanic. What do you guys recommend? I know its hard to diagnose car problems over the interwebs, but I'm at the end of my rope here. There's no check engine light.

 

Should I start checking fuel pressure? Replace wires? I've got a scanner that can check fuel trim if that helps.

 

Thanks for any help you can give me.

[Imp558]

Fuel pressure and then make sure it has spark on all 3 coils. How did the plugs look? anything suspicious?

[metallic95]

Well none of them looked particularly bad, just worn out. Some of them had a whitish deposit near the tip area. The car has just over 75,000 miles on it. Not to sound like a newbie or anything, but is there a better way of checking for spark on all 3 coils without pulling the plugs while it's running?

[Imp558]

It may be a good idea to do a good tune up, plug wires and all and see where you're actually starting.

if you put an old plug on a wire and lay it on a good ground like an engine mount you should see a spark. I usually pop the fuel pump relay or fuse out and have someone else crank it while I watch.

[Schurkey]

Test fuel pressure, cranking compression, and spark power.  I prefer a spark tester like these--you'll want the HEI-style with the recessed center electrode.  There are other designs of spark testers that work OK, too.

 

 

[Galaxie500XL]

Well none of them looked particularly bad, just worn out. Some of them had a whitish deposit near the tip area. The car has just over 75,000 miles on it. Not to sound like a newbie or anything, but is there a better way of checking for spark on all 3 coils without pulling the plugs while it's running?

One disadvantage of the DIS system is one bad coil means two cylinders are affected...over the years,I've replaced several coils on DIS engines that looked fine externally.

 

A spark tester is an excellent suggestion--I've found most of my bad coils that way.

[metallic95]

I'll check fuel pressure as soon as I can. I checked for spark from all the plugs, and each one checked out.  I also changed the wires out just for good measure, but still no joy.

 

Thanks for the replies guys, its much appreciated. One thing I did notice is that there is a very strong smell of fuel coming from the vacuum line that attaches to the fuel pressure regulator - is that normal? I know that if fuel comes out the regulator is bad, but I'm not sure about the smell.

[cutty collector]

I used to argue with my shop fore man at a gm dealership I worked at for years the test plugs posted are identical to gm's set up for testing spark but I found the crank out style of tester much better for diagnosing a gm ignition system you should be able to open the gap on one of these testers about an inch and still get spark .this test will show you weaker coils vs stronger ones hereby you can stress test the ignition system to the point where you can rule out a bad secondary .As for the rest of the advise given in this tread it is dead on.If you feel is missing in gear at idle duplicate it in the driveway at night have a buddy hold it in gear and spray it down with a garden hose watch for sparks flying 30 k ohm resistance is too high for a 8mm oe type plug wire .we used to ground a test light as though we would be using it to probe for power then slide it down the plug wire between the boot and wire towards the plug until the rpm dropped and the wire was shorted .If engine rpm didn't change when this occurred you found the affected cylinder .A fuel pressure gauge on lthe car stuck under a wiper blade and then a test drive would be the way to verify good fuel pressure. check the vacuum harness from the map sensor it was known to snap the cheap plastic lines

[cutty collector]

I just remembered some of those 3100 engines with multec injector were bad for poor spray patterns. It was one of the few cases where gm allowed us to perform an injector flush under warranty .If you can find a shop with a motorvac injector flushing system they work wonders better than the stuff gm offered

[cutty collector]

If there is a fuel leak there the fuel will be drawn into to engine and it will really affect idle quality .If you have acces to a hand vacuum pump when you have the fuel pressure gauge on it you should see about a 5 psi difference in pressure from top to bottom . There were some recalls for leaking regulators but I believe only some 3.8 liters were involved and I think maybe later nineties

[metallic95]

I checked fuel pressure. No leak down, the fuel pressure hits 45, drops down to 38, and stays there when the ignition is on without the car running. When the car is running, it drops about 2 psi. I went ahead and applied vacuum to the regulator and it's working (when I applied vacuum, the pressure dropped about 4 psi). I changed the wires out so I know they're not the problem. All the fuel pressure specs match what's in the service manual. I checked the vacuum hose tee coming from the MAP and they look good. Where are the vacuum lines for the charcoal canister and EVAP system? Behind the front bumper?

 

I'm really getting frustrated, don't know where else to go. Load test the ignition coils?

My scan tool gives a long term fuel trim of 145 - 148 in several frames - that means the engine is running lean, right? Also, for some reason it also says "quad driver 4 bad". There's no DTC.   

 

I'd test fuel pressure while driving, but the service manual doesn't indicate what the fuel pressure should be while driving. 

[55trucker]

Something not quite right there....

 

the LTFT (Block Learn) staying at 145-148  suggests to me that the engine is somewhat short on fuel, now 148 is nowhere near the max of 255 but the reading is not getting down to the proper 128 and I take it that the LTFT never drops into the - column, the LTFT should be closer to 128, the Integrator or (STFT) will shift moreso depending upon the engines immediate fuel requirements.

 

The LTFT is stored in the pcm's KAM (keep alive memory) and will remain there until that value is reset with a scanner or the power to the pcm is cut.

 

What is the reading on your scanner where the air/fuel mixture is concerned?  14.7?

 

 

see this simple link on the quad driver situation

 

 

https://answers.yahoo.com/question/index?qid=20080620190708AAaLq6G

[Imp558]

Could be a PCM, shorted connection, or a short/open circuit at a device. Lean would make me think injector, injector wiring, or PCM. My experience is all OBD-II, hopefully Saar chimes in, If he does that will be accurate information - very accurate.

[metallic95]

Just noticed something. There was an electrical connector I spotted that's disconnected. I assumed it was one of those plugs that don't get used during assembly, but I decided to check the service manual, and it's definitely not listed as a disconnected plug. Here's a video of the connector. I tried looking for where it connects to, but I just can't find anything. The ignition module is plugged in, and so is the o2 sensor. Anyone can chime in if it's supposed to be connected or not? Sorry for the dark video.

 

 

Air fuel ratio was not listed in the live data for some reason. It doesn't stay at 145-148 long term fuel trim, it fluctuates. It will be 148 in one frame, 122 the next. Then back to 144, holding at 148 for a few frames, then it drops back down to 129. This was measured as I test drove it in city and highway driving. 

 

Thanks again for the help guys.

[Imp558]

Evap, not that it matters, the car was running fine without it presumably so it's irrelevant to the problem you have now.

[metallic95]

If the EVAP is the small box with two hoses connected to it under the coils, then it's plugged in. Anywhere else it could go? 

[Imp558]

looks like the purge solenoid, that's usually a red two pin plug and it has a green wire with a white stripe on it. I didn't look it up in the schematics but I've seen a bunch of them. is it a green wire with a white tracer?

[metallic95]

Yeah, actually it does have a green wire with a white line on it. I'm not sure it has anything to do with the original issue anyways, but I only have the bad quad driver fault as a lead.  

[RobertISaar]

LTFT bouncing the way it does it is likely the PCM just  switching between the BLM(LTFT) cell matrix. the matrix can be as large as 16(or 17? been a while) cells, divided by RPM and MAP. IIRC, GM only enabled  4 of them  on the 94-95 3100 calibrations(the 94-95 3.4 used in the F-bodies used the same PCM with a different calibration, I think all cells were enabled in those cals). there is an option in the calibration for LTFT  to reset to 128 upon cell changes, which is  probably what you're seeing. hold it steady-state and you'll likely see it hold at a single value before too long. normally, LTFT is limited to  about 156 in most OBD1 calibrations, but  I seem to remember it being different here.

 

in any case, your LTFT is trending  high,  so the PCM is adding additional fuel, so the engine is running  lean or you have a misfire occurring(the uncombusted air will be read by the O2 sensor  as excess air). if I were you... I'd start disabling cylinders to see which one(s) are weak. depending on your  scantool, you may be able to do it that way. I setup the tunerpro definition for these  cars to be able to disable any cylinder on-command. also setup  the ability to disable an entire  bank(1/3/5 or 2/4/6), which oddly enough runs smoother than just a single cylinder being disabled.

 

I  THINK you have a coil/ICM/injector  that has failed, but you would need to be able to disable individual cylinders to find out  which cylinder to focus on and determine the cause from there.

[metallic95]

Thanks for the reply. 

I'm using an actron elite autoscanner. It doesn't give me the option to disable anything. So I just get an ALDL cable and download tunerpro? If you could give me some pointers that'd be great. 

 

Also my connector is one of those weird ones that doesn't have a B terminal. I think it was referred to as OBD 1.5? 

 

Maybe there's a way I can remove and test the ICM?

[RobertISaar]

lack of an ALDL B terminal is just an evolution of OBD1... GM started implementing 3(and 4)  digit error codes for a lot of platforms and  rather than make the blinkout procedure more complex, they just dropped the capability while retaining the same comm protocol that had been used starting in 86. there are even some cars that have the B terminal that don't blink codes. Cadillac started doing that before most other GM marques.

 

an ALDL cable + tunerpro RT will do quite a bit for what the combo costs. it does require someone to have built a definition for any given application to use it, but I did that years ago for almost every common GM OBD1 application. I've spent quite a bit of time developing the P66 V6 definition in particular, you could download tunerpro and the P66 V6 ADX to get an idea of  what you would have access to ahead of time, but I've got most of the functionality of a Tech1 replicated.

 

honestly.... every 60V6(and 3800) owner should have at least one spare ICM and coilpack(preferably a set) hidden away somewhere for when(not if) they fail.

[Schurkey]

I wouldn't be happy about a quad-driver failure.  You "probably" need to address that before you go any farther. 

 

However:  I spent two years--on and off--screwing around with a K1500 that had high fuel trim and a lean code.  NOTHING helped.  Fuel pressure was OK, engine ran OK.  O2 crosscounts at low speed were fine.  NO crosscounts at highway speed, fuel trim went high, then the dash light came on and the "lean" code set...again.

 

Amazing what a replacement O2 sensor will fix. 

 

How old is your O2 sensor?

[primergray]

Yeah, actually it does have a green wire with a white line on it. I'm not sure it has anything to do with the original issue anyways, but I only have the bad quad driver fault as a lead.  

 

I had a quad driver fault, I don't remember which quads control what, but I do remember that the quad fault I saw was for either TCC or purge valve. TCC seemed fine, replaced purge valve. I had the engine light on for that issue, though...and the new purge valve took care of it.

 

 

[metallic95]

Well, I decided to check for spark again from each plug, and found the plug going to the #1 cylinder to be dead. Changed the coil, issue is gone and car runs great again. I think when I checked last I may have pulled one of the rear plugs twice. Oh well, if I had a quarter for every time-and-frustration costing error I made while working on cars I'd be a billionaire.  

 

Thanks for the help again guys, great community and I'll definitely be back again for any other roadblocks I hit. 

Now onto fixing that pesky transmission fluid leak....