89 Mustang 5.0, idle misfire

Hello All,
New here, but not new to cars.
I've had my current Mustang for better part of 20 years.

I've had this idle misfire that has been bugging me for years. It's been more of a nuisance than a real problem.

I just rebuilt the engine a few months back, but the misfire was there before rebuild.

Short summary of engine is aluminum heads, slightly upgraded cam, intake from 93 Cobra, aftermarket MAF sensor and injectors. I also have tuning software that I've used to tweak it a bit. No amount of tune changes have resolved it. Does it with and without MSD box installed.

Under medium cruising, highway driving, and WOT, no observed misfire. When I'm lightly on the throttle or sitting at idle, I get a noticeable misfire. It's almost like I lose multiple cylinders in a row because it rocks the car a bit, and then it'll settle down for a few seconds and then repeat, if that makes sense. I'd say it's more noticeable when engine is hot.

I've checked for vacuum leaks (smoke test), sensor signals (via tuning software), changed injectors, cap, rotor, wires, plugs, coil (not all to resolve the issue, just changes over the years).

I've played with the tune to see if making it rich or lean had any effect and no change.

I know the TFI module and pip are known to go bad, but I want to test to verify.

I have a hantek 1008c and my thought it to connect get some injector, primary and secondary waveforms along with pip and spout signals. Not sure I can do all at the same time.

Does that sound reasonable? Any other items to check?

Update: I was able to hook up my Hantek to the car and monitor a few signals.

PIP, SPOUT and injector 1 (only checked #1 for now) looked good all the time, no dropouts.

What I found is that the main spike is sometimes small and equal to the small spike after plug burn.
Other times, I get a normal coil waveform.

Primary and secondary match all the time.

I had issues getting the trigger to set off the HT lead, so at one point I had injector 1 as my trigger to get a steady graph.
I also had issues getting a good coil waveform from the wires off the distributor cap.
If the picture works below, it shows injector 1 trigger and then you can see 5 consecutive coil fires that have the odd spike and then the 6th is a good spike. Also include PIP (purple) and SPOUT (green).

Engine was hot and it was about 85F out today.
I'm thinking this could be a TFI issue or possibly cap/rotor?
I will say I noticed it more when engine is hot. I'll try to repeat it tomorrow morning when it's cooler.



Here's the primary and secondary.

Vacuum leak and valve train are my top suspects for an idle only misfire.
Since you have a scope, relative compression would be a valuable test.
Can you narrow it down to a specific cylinder(s) by killing spark to one cylinder at a time?
Are you positive the firing order is correct? I've seen good techs not only use the non-H.O. firing order, but actually change a perfectly good engine over it.
How do the plugs look?
And for my own information, what tuner are you using to look at data?
I love Fox Body Mustangs but hate EECIV. I finally pulled the trigger on a Microsquirt stand alone ECU, the only reason I didn't spike the A9P into the nearest trashcan is that they're only getting harder to find...

Based on your Primary and Secondary waveforms, my first guess is valve sealing issue. A cylinder with lower compression will have a lower Firing KV demand. I second Noah's suggestion of a Relative Compression test.

Thanks for the input.

I had this issue with the engine before it was rebuilt, so it's not new. Just a bit of engine background.
The only different major parts were the crank(reason for rebuild) and block. Obviously new gaskets, rings, bearings...etc.
I also had the heads resurfaced, new valve seals and also lapped the valves before reassembly. When I checked chamber volume on the heads with alcohol, they seemed to seal up nicely. Valve guides seemed to be tight as well.


I can try to pull wires and see if I can narrow down which cylinder, but I feel like it's random.
Based on my datalogs and O2 sensor responses, it bounces from one bank the other. That doesn't necessarily mean it's not one cylinder on each bank, but it leads to believe it's on both banks.

Maybe I need to make that clear, this isn't a dead miss, it's random. Car seems to idle decently when cold, but when it gets hot, it seems to misfire more.

I pulled driver side plugs and they all look good. A few weeks back I pulled all plugs and they all looked good.
Firing order is correct 13726548.

I might run a smoke test again to check for vacuum leaks, but it should be pretty sealed up.
I also thought about pulling valve covers to verify rockers are still adjusted correctly.

For relative compression with a scope, do I need an amp clamp? If so, I do not have one (yet).

I have a leakdown tester so I could run a leakdown test.

I'm running Tweecer with my A9S computer. This was a California car. A9S is almost the same as A9L, just a few small differences for CA I guess.
Also, I swapped the capacitors in the ECU (known to fail), but no change. ECU is original to the car as is all the wiring.
Speaking of wiring, I did do the salt and pepper shaker mod (better connection).


I plan to scope it again this morning when it's cold to see if I get the same waveforms.

I appreciate the input. Hopefully I can finally figure this out because other than the miss at idle, car seems to run well.

You do not need and amp clamp to do Relative Compression. You can get the same waveform by using Battery Voltage. As Current goes up, Voltage goes down. If you take a waveform of Battery Voltage, the waveform will be upside down, compared to a Current waveform. You can use an invert function to flip it. I use AC coupling, with a 2 volt scale.

Since the misfire is random, a very long cranking period may be necessary.

OK, sounds good.

Another update, I went out and ran some tests again with the car cold.
Traces seems to be the same, randomly good/bad.

I tried an old school relative compression test where you crank the engine with pedal to the floor which is flood clear mode so it doesn't fire the injectors. I didn't notice any obvious differences in cranking sound.

2 things popped up when scoping (engine running cold).
1. I could not get a good signal after the distributor with the HT clamp. I tried my current 9mm plug wires along with an 8mm set I had. It was very noisy and the trace rarely looked like a good fire (if ever). I had it clamped near the distributor. Maybe it was too close to next wire and getting mixed signal?

2. See below image of 2 firing events for cylinder 1. HT lead is on the main wire from coil. Trigger is #1 injector, so the first coil fire after that is #1.
First fire is not good, 2nd one appears to be a lean misfire (based on my minimal knowledge). If you look closely, you can see the injector is firing. The pintle hump is there. I wonder if my HT lead is junk and giving false readings? The coil charge seems to be OK though.







EDIT: One final update for this morning.
I was able to get the trace after the distributor, but I had to change the scale to 200v/div, so the spike was about 600v.
It was also noisy and inconsistent.

I ran a quick leakdown test, mainly to see if there were any valve seal issues.

Leakdown was about 8-10% on cold engine with Harbor Freight tester. My ring gaps are on the larger side because I plan to eventually put my supercharger back on.

All leakdown was past the rings. I could not hear/feel anything from intake or exhaust.

I also checked all injectors with scope and they all looked good. The only thing is I used the injector to trigger the scope.

If I had an injector not fire, would the signal drop out for split second?

I ask because I didn't see any dropped waveforms, it was consistent with all injectors.

Any updates? I have a similar problem with my 94 5.0.

Nope, nothing really new.
I did replace the cap/rotor just because the old ones were a little questionable.
Maybe idles a little smoother, but not a dramatic difference.

I don't think it's related to PIP or TFI since those signals were very consistent when scoped.

Based on the testing of the spark signal, that seems to be where the issue is, but nothing obvious. I had another coil sitting on the shelf, swapped it it and no change.

As been mentioned, it could be a vacuum leak or cylinder sealing issue, but I've done both smoke test for vacuum and also leak down for cylinder sealing and no 'ah-ha' moment.

One thing that was mentioned is the possibility of vacuum leaks past the rocker stud threads. The threads for the intake rocker studs go into the intake ports on my heads. This is not a big issue as long as they are sealed, but I have no idea if they are sealed. I've had these heads for ~20 years. I did have new valve guides put in about 15 years ago, but I don't know if the shop pulled the guide plates off or not and if they did, not sure if they sealed the threads. Guess it could be worth the hour to pull the studs out and seal them just to check that box.

I've done a lot of the mechanical checks and haven't found anything significant either.

I did use the wrong sealant on my rocker threads initially. It started pulling oil through the threads and into the intake port, causing smoke from the tailpipes at idle. That problem was solved by sealing them up with some high temp thread sealant (white paste type).

If I find anything helpful I'll share it.