Our Community is 940,000 Strong. Join Us.


3800 cc Misfire Valves or Lifters?


bucchb
01-29-2009, 01:33 PM
Ok, here is a real head scratcher. Misfire on Cylinder 4 only, PO304 code.

I have a 1997 Buick Lesabre with the 3800 engine. Bought this car at auction and suspected misfire to be a coil pack, wire sparkplug. Changed everything, didn't fix.

I just drove the car 2000 miles at 75 -80 MPH. Consistantly this car will misfire when you accelerate, and get this when you applied the break????
Rolling down the interstate I would put the car in neutral, shut the engine, put ket to on, and restart. 50% of the time I was running on 6 cyclinders.

So I suspected a vacuum leak and checked every hose. Nothing.
Suspected an intermittent injector, changed. Nothing.

Ok, maybe a lifter or valve spring or stuck valve. So I screwed in the compression gage, and started engine. Right to 150 psi. Induced the "misfire" by holding break and accelerating and next try went to 90 psi.
Shut engine again and check and got a zero psi reading. 2 more trys and 0 psi.

Next step will be to run engine with valve cover off. I guess I will be looking for valves not opening = bad lifter.
Valves not closing=bad spring or sticking in guide.

I have already run Gunk Engine flush and Marvel Mytery oil. The Gunk did nothing, but the Marvel did seem to help.
I am trying to avoid a major overhaul, as this car just aint worth it.

I am hoping for someone to tell me:

1) If valve springs, are the valve guides removable from the top when I change the springs?
(I am planning to pump the cylinder with 120 psi from my compressor.)

2) has anyone seen a lifter be this intermittent? (ps I get no lifter noise when misfiring)

3) ANYONE EVER RUN INTO THIS PROBLEM AND FIX IT?

Thanks
Bill

maxwedge
01-29-2009, 03:50 PM
Welcome to AF. I would suspect a injector based on what you say has been done,.The odds are very high against it being internal.

Scrapper
01-29-2009, 04:24 PM
i agree with max on the injector or injectors since you say coil pack has been replaced. i would try some sea foam running it threw your gas. those 3.8 or 3800 are good motors if taken care of they can get up to 400000 mls out of them and plus over 400000...good luck...

stuzman
01-29-2009, 06:18 PM
Ok, here is a real head scratcher. Misfire on Cylinder 4 only, PO304 code.

I have a 1997 Buick Lesabre with the 3800 engine. Bought this car at auction and suspected misfire to be a coil pack, wire sparkplug. Changed everything, didn't fix.

I just drove the car 2000 miles at 75 -80 MPH. Consistantly this car will misfire when you accelerate, and get this when you applied the break????
Rolling down the interstate I would put the car in neutral, shut the engine, put ket to on, and restart. 50% of the time I was running on 6 cyclinders.

So I suspected a vacuum leak and checked every hose. Nothing.
Suspected an intermittent injector, changed. Nothing.

Ok, maybe a lifter or valve spring or stuck valve. So I screwed in the compression gage, and started engine. Right to 150 psi. Induced the "misfire" by holding break and accelerating and next try went to 90 psi.
Shut engine again and check and got a zero psi reading. 2 more trys and 0 psi.

Next step will be to run engine with valve cover off. I guess I will be looking for valves not opening = bad lifter.
Valves not closing=bad spring or sticking in guide.

I have already run Gunk Engine flush and Marvel Mytery oil. The Gunk did nothing, but the Marvel did seem to help.
I am trying to avoid a major overhaul, as this car just aint worth it.

I am hoping for someone to tell me:

1) If valve springs, are the valve guides removable from the top when I change the springs?
(I am planning to pump the cylinder with 120 psi from my compressor.)

2) has anyone seen a lifter be this intermittent? (ps I get no lifter noise when misfiring)

3) ANYONE EVER RUN INTO THIS PROBLEM AND FIX IT?

Thanks
Bill

Since you've changed everything for the ignition on cylinder 4 and the fuel injector, the only thing left is the internal for that cylinder. Seems odd that your pressure varies to 90 and then to 0. You may want to do a leak-down test on that cylinder since you've got an air compressor and see if that gives you an indication of where your problem may be.

HotZ28
01-29-2009, 06:57 PM
Suspected an intermittent injector, changed. Nothing.Obviously, you have changed the injector, but have you checked the wiring to the injector? Use a noid light for testing injector firing.
1) If valve springs, are the valve guides removable from the top when I change the springs?(I am planning to pump the cylinder with 120 psi from my compressor.)No, valve guides are not removable from the top! If you have a problem with the spring, lifter, guide, or valve sticking, you would have play in the rocker arm and rocker noise would be very noticeable.
2) has anyone seen a lifter be this intermittent? (ps I get no lifter noise when misfiring)If you do not have lifter noise, you do not have a valve train problem! None of the valve train possibilities you mentioned, seem to match up with your problem. Did you do a pressure test on the cooling system? Have you removed the #4 plug since you replaced it, to see if it is clean and has the appearance of good combustion & firing?
You may want to do a leakdown test. BTW, your procedure for checking compression is totally flawed! Follow the procedure below:
Make sure the engine is warm before testing, this allows for thermal expansion and for the oil to be warmed up. A cold engine will not test correctly.
Disable the ignition module, or coils and then remove all spark plugs.
Insert the compression tester into one cylinder at a time.
Hold the throttle to full open position to ensure the engine gets adequate air intake.
Rotate the engine for at least four to six revolutions to obtain an accurate reading on the compression tester.
Record the reading for each cylinder. If any of them vary more than 10% from each other, a problem may exist in one or more cylinders. If the variance is greater than 10%, specialized testing equipment may be required to fully diagnose the problem.
If all cylinder readings are within 10% of each other, no further testing is required and compression is typical, based on mileage & wear.

Add your comment to this topic!