If there isn't evidence of oil, then the lack of compression in those cylinders makes me think there could be some leaking valves/guides as an explanation. If it's valve related, possibly they could be creating a lean/rich condition at idle and when you give gas the leak might not be enough of an issue. Do you have the long term fuel trims from last time you threw the code? It's not as accurate as watching the trims graph out live, but I'd be curious to know what they were when the code was last thrown.

 

As a part-time family taxi driver, I have to take great granddaughter to dentist. Be back Monday afternoon.

 

So we are looking for any indication of a problem at this point. We are completely done with ignition system.

You haven't yet done the dead end fuel pressure tests, but I would not expect these additional tests could indicate any problem other than possibly a weak fuel pump (over what you have done already and based on it passing the pressure value and leak down test you reported at Key-on pressure). I wouldn't expect a weak fuel pump to cause a miss at idle only, unless you have a vehicle voltage problem at idle only. We may need to revisit and complete the dead end pressures later, but for now lets go on.

So we know the fuel pressure regulator in the spider is working and that the spider is not dribbling fuel out of the pressure regulator into the intake. If you have the MPFI injector spider, then we know none of the hoses to the injectors are dribbling fuel into the intake either. If you have the old-style spider, the solenoids are all in the center body of the spider so the hoses are not checked. But running rich from fuel dribbling out inside your intake is not your problem. I suppose an injector (or poppet nozzle) could be sticking closed at low RPM, or that battery voltage being lower at idle might not adequately operate a marginal solenoid or injector.

We also know that the compression is somewhat low on #3 cylinder in relationship to the highest ones. Usually it is said that all should be within 20% of each other. I do not personaly believe that rings could be the misfire problem with 120psi on the compression test, since only 60psi or so is required to fire. But valvetrain problems like worn valve guides can manifest in strange ways. if intake guide or intake valve is shot, exhaust gas could leak back into the intake (especially at low RPMs), diluting the intake charge and causing a misfire. Lets keep this in our hip pocket for now.

So still looming over us is a possible vacuum leak, valvetrain problem, injector problem, or voltage at idle problem causing low or no delivery of fuel.

So lets artificially enrichen the fuel mixture by spraying a shot of flammable MAF cleaner into the intake ahead of the MAF sensor (air filter area) and see if the monitored misfires disappear at idle and then come back after the MAF cleaner burns off.

Another thing to do is to temporarily disconnect some wires so the alternator does not charge and monitor misfires to see if it might be a voltage-related problem at injectors/solenoids or even possibly fuel pump. This might even tie in with your observed weak battery.

As far as the crank sensor relearn, it is specifically used for misfire detection only. Engine should never have a felt miss because of it. The factory manual states in the tables and diagnostic charts to do a relearn at certain points, but in reality it is done to eliminate a false misfire detection condition. You have felt misfires. We have effectively already eliminated false misfires as a problem by verifying that the existing crank learn is good enough to correctly detect misfires when you pull plug wires (even though 1 and 2 were switched in the software). Furthermore, I believe it is a mistake to calibrate the crank sensor on an engine that actually has a felt misfire (pretty difficult to do sometimes). It can destroy your currently working misfire detection, if you can get it to take. So I don't recommend it for you (at least at this point).

At least this is my story and I'm sticking to it. But know I have been wrong before!

Please try these two things in red and report back. I wouldn't hook up the steering column just yet (may want to eventually do a cylinder leak down test using your air compressor). If the misfire improves with additonal fuel to burn, we can definitely eliminate some more stuff and hopefully start zeroing in on the problem. I think you continue to make good progress.

I'm still wondering what precipitated the original diagnosis of a head gasket, and if by the right head you mean passenger side. Did it have a coolant leak? Does this engine have any ticks or knocks under any condition?

 

So lets artificially enrichen the fuel mixture by spraying a shot of flammable MAF cleaner into the intake ahead of the MAF sensor (air filter area) and see if the monitored misfires disappear at idle and then come back after the MAF cleaner burns off.
This had no effect. A little squirt or a larger one that almost killed the engine.
Another thing to do is to temporarily disconnect some wires so the alternator does not charge and monitor misfires to see if it might be a voltage-related problem at injectors/solenoids or even possibly fuel pump. This might even tie in with your observed weak battery. Removed the multi-wire plug...again no effect on the misfire.

"I'm still wondering what precipitated the original diagnosis of a head gasket, and if by the right head you mean passenger side. Did it have a coolant leak? Does this engine have any ticks or knocks under any condition?"


Answer #3: Originally the engine would "barely" run. On a whim, I did a compression check on the RH side and #4 was only 80psi, so I removed the RH head (passenger side). The head gasket was broken between #4 & #6. I reseated the valves while I had the head off. I did nothing to LH side. No coolant leaks and no ticks or knocks.


I wanted to report these facts, so I didn't do any fuel trim numbers. The Cam Retard* is still 0.


Today: I installed a new battery. The misfire was 0 on both #3 & #4 until it warmed up. Probably due to the higher rpm's. Again, I repeat, the misfires disappear @ 800rpm. I also did a leak down test on the cyls. My gauge limits are as follow:


Green = LOW = 0-40%
Yellow = MODERATE = 40-70%
Red = HIGH = 70-100%


#1 = 22% low
#2 = 23% low
#3 = 50% moderate
#4 = 46% moderate
#5 = 31% low
#6 = 33% low


I have not done any "serious" vacuum checking. Does anyone have a diagram, so I can be sure I check "every circuit"?

 

I'm thinking you have a leak that is introducing excess oxygen into the engine. Since your low COMPRESSION is on two cylinders opposite each other, I would first focus on the middle of the LIM on both sides and focus something (some people use propane, some use water, some use carb spray...whatever you're comfortable with) to see if you can make the engine change one way or the other. If you use water, put it in the stream mode and focus on the intake seams to see if you can hear it gurgle like something is sucking it. Do it cold before the engine warms up. I don't have a vac diagram for your truck unfortunately.

 

Hi Joe,

Leak down test indicates cyl 3&4 definitely have a problem as they are different from the rest. Same cylinders that are missing. Sorry to say but only thing you need to do is figure out rings, intake valve, or exhaust valve on those two cylinders. Its pretty definitive.

What is the model of your leak down tester and what pressure did you use? Avoid oil as you can use your ear to determine which it is. I'm betting intake valves.

 

Leak down tester is an el cheapo from Harbor Freight called U.S. General. Only abut 10psi input to zero the testing gauge part.
Will try the propane treatment.

 

Joe,

If propane deoes anything and MAF cleaner doesn't - it's because the MAF cleaner wasn't flammable type. But I think this test is really not necessary anymore.

Standard leak down testing on automotive cylinders is done on a warm engine while rocking piston about TDC on compression stroke. Air is regulated at 100psi by the leakdown tester, put through a 0.040 orifice, and then pressure is read again (on cylinder side of orifice). 90psi would be 10% leak down. 75 psi would be 25% leak down. Supposedly equivalent testers have been shown to vary 10psi from each other (100 psi input) on same engine and cylinder. This has to be due to inaccuracy of pressure gauges used or inaccuracy of the orifice size. Most important is using the same tester - cylinder to cylinder. Generally acceptable leak down is 25% in a used passenger engine but remember the sometimes 10psi variation from tester to tester. Race engines can get below 5% - even down to 1% leak down I am told.

Your Harbor Freight obviously does not have a 0.040 orifice and operates at a much lower non-standard pressure - therefore your second guage is in percent and not psi - hence the need to zero it. However, your guage should work similarly and be able to tell differences between cylinders. This you have accomplished without a doubt!

You have found differences in both compression and leak down on #3 and #4, the same cylinders that are giving the misfires at idle. It's a slam dunk that you have found the problem + 50% is simply too much leakdown with any tester used. I know this is not the problem that you wanted to find, but nevertheless you have found it. Congrats!! Very few people get this far in misfire diagnosis! Great job!!!

Now just pressurize the cylinder to around 100 psi at TDC of compression stroke for that cylinder with your air compressor (no leak-down guage necessary) and listen at exhaust pipe and throttle body for air past each respective valve. It will leak air at a higher rate with the higher pressure. If no audible leaks at exhaust or throttle body, figure rings. You can compare sounds to one of the known good cylinders if necessary. At 100 psi air pressure you might want to take off the serpentine belt and hold engine in place with a socket in case the piston decides to go down and turns the engine. Don't want to loose any fingers.

Best regards,

Les

 

How is a misfire detected?

 

Short answer is by detected slowing of the time for reluctor to pass crank sensor at the time when that cylinder should have fired. This is based on data obtained during the crank sensor learn procedure and timed to cylinder using the cam sensor signal.