Engine stumbling

[drobinson]

My 89 has developed a misfire or stumble when it goes into closed loop. Hve replaced plugs and wires, there are no fault codes. The stumble occurs at highway speeds and quick accerlation seems to clear the problem. Where should I look?

[ronvb]

Check plugs, loose or broken vacuum lines.If there is pinging check EGR valve.

ronvb

[Guest Greg Ross]

You say this is occurring when you go into closed loop, what else is going on. What is the driving state, speed and gear it's in specifically? And how and where does it even out?

[drobinson]

I thought I had the stumbling, skip, miss or whatever isolated to going into closed loop however it has since occured with engine in the 120F range. If this were a carburated engine I would think of water or dirt particles in the gas. I am able to clear the roughness ( sometimes) by rapid acceleration (flooring it).

I have replaced the fuel filter in the gas line, run 5 cans of injector cleaner, and replaced the plugs and wires. There are no faults shown and the inputs are in the normal range although the engine goes to rich (155) when it stumbles. The EGR test ( override )seemed to function as decribed.

Don't want to start changing parts without a plan.

drr

[2seater]

Just a thought, but a leaking injector maybe? It will tolerate additional fuel when cool and my '90 will go closed loop at less than 130 deg.F, so that is not a hard and fast temperature. Have you tried going into the overides in the diagnostics? Go to the injector overides and you can toggle each one on and off to see if there is a difference in the effect on the engine as you do so. It may help isolate if this is confined to a single cylinder. I have never seen the fuel integrator or BLM go over 150, but if it is doing so, it is trying desperately to add fuel for some reason. What do ED19 and 20 read when, and if, it ever runs well? One other common reason for the fuel to go to the high range is a vacuum leak somewhere. If you run your fingers over the vacuum hoses, and they come away with a black residue like soot, they should be replaced. They can be porous even if they look to be intact. Pay particular attention to the hoses behind the throttle body area near the firewall. If you try disconnecting the lines from the plastic vacuum block on top of the throttle body adapter to the intake, be very careful as this is generally very brittle and can be broken easily. One other prime spot for a vacuum leak is the rubber grommets at the PCV valve connection. There have been reports of these getting hard and deteriorated.

[Guest]

Thanks to input from Nic Walker I think the problem is solved. I replaced the ECM with a remanufactured ECM from Advance Auto. Not only has the stumble gone away the whole drive train seems to be smoother. Too bad the ECM can't diagnosis itself.

Thanks again for the help

drr

[2seater]

Thank you for posting the fix. There seem to be more and more of the bad ECM's showing up, which appears to be more age, rather than mileage. You are right about the lack of codes for the ECM.

[Guest imported_dantm4]

What is a normal number for the BLM and fuel integrator (ED19 and 20)? Just curious since you say 150 and above is rich and my fuel mileage always seems low.

Dan

90' Black/Tan Coupe

[Guest wally888]

DATA FROM OLD AND NEW O. S.

On the way back from making my semi-annual deposit in Biloxi, I recorded these readings from old O.S. and......(Cruise on, flat terrain, 65 mph)

Averages of two readings:

ED-18 43 (42/44)

ED-19 131 (130/133)

ED-20 130 (129/131)

ED-22 71 (71/71)

Readings w/ new Bosch O.S.

May have gained about 1 1/2 MPG mostly city driving , 21 MPG , 91 Reatta.

ED-18=24

Ed-19=124

ED-21=30

ED-22=131

Ed-26=14.7

[2seater]

The scale for both is 0-255 with 128 being the center point and is the desired reading under ideal conditions. Padgett can possibly clarify, but I do not believe the entire scale to 255 is used, and generally 150 is the maximum rich adjustment that is available. The BLM is the long term fuel trim and the Integrator is the short term, or instant reading. It is normal, and actually desired, that the integrator bounces a little above and below the ideal 128. It will change as your acceleration and speed change. If the short term adjustment is consistantly above or below the ideal 128, and it stays there, the ECM will make an adjustment to the long term fuel trim by adjusting in the same direction as the integrator. If the long term adjustment corrects the problem, the integrator should return to the area of the ideal. If both the Block Learn and Integrator are reading way high, the ECM has run out of adjustment that it can make. Typical reasons for the adjustments being out of whack: vacuum leaks, restricted or leaking injectors, high or low fuel pressure, or an O2 sensor that is defective or biased rich or lean. Try this for a basic explanation:

http://www.gnttype.org/techarea/faq/BLMINT.html

[Guest imported_dantm4]

I found that page really cool, is this true for all cars? (at least from computer standpoint, I know the cells and calculations by rpm and maf might change for each automaker)

-dan

[padgett]

Wellllll, about all US GM cars, they tend to use the same subroutines over and over just like they have been using essentially the same CPU for over 20 years. 128 just happens to be the midpoint of an 8 bit integer so easy match to the O2 reading and generate a correction, much easier than trying to go around zero.