So I'm helping my buddy resurrect subj Blazer after it's been sitting in a field for two years. I've been working on it for several weeks, teaching him basic mechanics as we go (while I'm continuing to learn also). We've replaced the fuel pump/sender, water pump, radiator, shocks, brakes and fluid, both ABS sensors, alternator, battery, dist. cap, rotor, plugs and wirs. Replaced multiple gaskets, multiple vacuum lines, several sensors and both door bushings on driver/pass doors. It is almost done, but I can't figure out this darn P0304 code. It did not run when we pulled it in to my garage, but we now have it running and driving decent, other than it's registering 11,000+ misfires on cyl 4 during a 30 min drive. It does not set P0304 code every time we drive it; sometimes it does, sometimes it doesn't. Some of the other cylinders register single digit misfires, but nothing serious. We used AC Delco/Delphi parts for the tune-up, I've swapped plugs from #2 to #4, tried a new wire, different cap and fuel trim (STFT) is reading single digits while driving around. I'm at a loss here. Could this be caused by CMP Retard being off? I haven't checked that because my Auto X-ray scanner does not show CMP Retard. This Blazer has 130k, with a perfect paint job, zero leaks, starts and runs decent, absolutely no body damage and everything else works (now...thanks to the info on this website!).

I'd really like to get this fixed and would appreciate some rudder to point me in the right direction to fix this. THX!
~Magman

 

is it a real felt misfire? Have you jacked with the crank sensor?

 

No, we haven't touched the Crank Sensor and it does seem to be a real misfire. The weird thing is, according to my scanner, once up to speed (above 35 mph) it doesn't register any misfires. All misfires are registered when we are driving under approx. 35 mph. I've witnessed this on the scanner while my buddy is driving. Once under 35 mph, the scanner misfire counts go up by the hundreds every refresh (approx. 10 secs. on the scanner), once over 35 mph they stop accumulating altogether. I'm stumped...but we can definitely feel the misfire.

 

OK - so the misfire seems to be real. Sounds like you've done a lot of the preliminary testing already. I would monitor and correct the Cam sensor retard if not 0 /- 2 degrees, keeping in mind that distributor gear wear can cause it to be off. If nothing found there, try the running compression tests per https://support.alldata.com/article/...pression-tests to see volumetric efficiency on each cylinder and see if anything becomes apparent.

 

I didn't see if u replaced the coil . If u didnt , replace it. They have a habit of arcing straight down to the valve cover where u can't see it.

 

I won't be able to get to that until this weekend with work and all. I ran and recorded fuel trim tonight though. Not sure if it helps or not...
In the garage and warmed up, car was in closed loop before I recorded
At Idle:
STFT Bank 1 - Ranged from 0.0% to 1.5%
STFT Bank 2 - Ranged from 0.0% to -3.2%
LTFT Bank 1 - Registered 8.5%
LTFT Bank 2 - Registered -1.6% but did show -3.2%

At 2,000 RPMS
STFT Bank 1 - Ranged from 0.7to -5.5%
STFT Bank 2 - Ranged from 2.3% to -4.7%
LTFT Bank 1 - Registered 0.8%
LTFT Bank 2 - Registered -7.9%
Misfires accumulated during this time:
Cyl 1 = 3; Cyl 2 = 0, Cyl 3 = 0, Cyl 4 = 6,013, Cyl 5 = 0, Cyl 6 = 4.

My scanner showed another reading of: LTFT = 132 and also RCH/LN LN/RCH = 0
Total time was about 6-8 mins and I reset codes before recording fuel psi. Before that, I disconnected #4 wire and grounded it, ran the car and witnessed approx. twice the misfire count on the scanner and did not set the SES light. After that, I put it back together and ran the tests without setting the SES light.

Even though it registers misfires, it runs smooth with plenty of power. The drawback is, it occasionally sets the SES light. Thanks for any insight.

 

So which is it? Felt misfire or not? My previous instructions were in the case of a felt misfire.

The following is a KEY concept. Felt misfires are real. Misfires that can't be felt (false positives) are not necessarily real - even though the system may tell you they are occurring. If the system is telling you a misfire is there when it is not, then something is likely wrong with the misfire detection. Make sense?

So lets look at how misfire detection works. When a cylinder fires there is a very brief acceleration of the crankshaft over and above it's average rotational speed. So there are 8 accelerations occurring every 2 revs of the crank. The PCM knows which cylinder is firing at any time through the cam sensor in the distributor. During a crank sensor relearn (done only with pricey scanners and software), the PCM records the electronic waveform from the crank sensor (which includes the specific accelerations during each cylinder firing) and uses this data for all misfire calculations until another crank sensor relearn is performed. If, when running, the PCM does not see exactly the waveform that it expects - then it flags a misfire count. If the PCM sees the crank sensor waveform that it expects within some programmed values, then all is fine. This is why we always say to NOT disturb the crank sensor or replace it *****-nilly to see if it will help performance. Who can say if it will then generate a waveform that will screw up the misfire detection and give false positives until a crank sensor relearn is done.

Unfortunately there are other things that can mess with the crank sensor waveform, and this is per the factory manual. Anything attached to the crankshaft (even through the serpentine belt) can cause false misfires through balance problems. So Flex plates, flywheels, harmonic balancers, alternators, etc can cause problems with misfire detection too. My favorite is what happened to me on my pretty perfect Blazer right after I got it four years ago. Excessive clearance in engine bearings was causing false misfires, but found this only after I fixed the SES light so it would work (PO had removed the wire from the cluster connector and taped it - apparently when trading it in). So silly me for thinking the engine was as perfect as the body. My only clue was a single lifter ticking loudly after driving home 120 miles. With the problems these Blazers had with lower intake gaskets leaking coolant into the oil, I can see how that happened.

From my experience, I found that misfire detection was completely goofed on mine. It couldn't detect a misfire that I actually induced on a previously known good cylinder. Neither could I get the PCM to conduct the crank sensor relearn until the bearings were fixed. That is why I'm going to suggest that you disconnect the plug wire for one of the known good cylinders (currently zero misfires) and see if misfire detection works as it should on that cylinder.

Don't think you're going to go very far with the fuel trim at this point.

 

Thanks Les, I appreciate the insight. I was working on it yesterday and noticed my scanner registered EGR Fail so I was looking it over and found the line coming from the left manifold completely snapped in half. I'm headed to a junkyard now to pull the parts and fix that first. I'm wondering if this could be a causal factor. I'll get back to you on the compression tests, my buddy and I are working on it this weekend between football games! Thanks again for your assistance!
~Tim

 

We fixed the EGR tube and re-ran the vehicle yesterday with the same misfire problems. I finished the compression test this morning and was rewarding with disappointing results. First was KOEO fuel PSI: solid at 62 and after a 5-min leakdown it was 56. So, no problem there.
Next was Static Compression, followed by 1200 RPM Compression, and Snap Compression (1200 to 2500 RPM). The results are as follows, written in the same respective order for the right cylinder bank.
CYL 2 - 160, 60, 100
CYL 4 - 92, 32, 90
CYL 6 - 132, 45, 100

So, from what I gather CYL 4 has a significant loss of sealing capability and CYL 6 has a minor loss of sealing capability. This leads me to believe I'm dealing with a bad head gasket even though I have no coolant/oil loss or mixing. I believe my next step is to pull the heads and send them off to be checked over, pressure tested, and resurfaced. These results coincide with the history of the vehicle. Two years ago it was parked due to overheating from loss of coolant. We found and replaced the leaking water pump thinking that was the problem. However, it appears the previous owner probably overheated it and damaged the head gasket when he lost coolant (water pump leak).
Thanks for guiding me to the compression test Les, this definitely showed me where to focus.

I think I'm on the right path now instead of chasing faulty ignition components or fuel. Any other insight? Thanks!
~Tim

 

It's not going to be a head gasket, so don't jump to conclusions and take it apart just yet. Probably slightly stuck or rusty intake valve from sitting with it open for the two years. Do the Sea-foam treatment and see if it helps. Might just need to be driven some as well.

Seafoam treatment is with completely hot engine 1/2 can down throttle body slowly with engine still running (you'll have to find a place where you can let'r smoke), then dump second half all in at once to kill engine. Let sit a couple of hours and then start and let smoke clear. Rev engine quickly full throttle several times, but don't get too high RPM (keep below 4000). Take out and make that engine pull by riding the brake or other means - drive the P*SS out of it for a bit, but again keep those RPMs down so you don't take out the bearings n an old engine. If you have a freeway convenient, hitting passing gear a few times from 50 mph up to your speed limit will do the trick. Then one more can in 1/2 tank of gas, and run that out in normal driving.

Good luck, and let us know how it turns out.