CarTalk.com Blogs Car Info Our Show Deals Mechanics Files Vehicle Donation

Fuel Pump or Injectors?

My 2001 mustang v6 has a bad misfire upon starting but only if the engine has been warmed up, turned off and then restarted while still warm. I also smell strong fumes/exhaust when it misfires. It’s definitely NOT gasoline though.

If I let it cool down first or do a cold start there is no misfire and it will continue to drive perfectly even when it warms up to operating temperature.

So far I have only got a bank 2 lean code (P0174) and a P0193 code which indicates the fuel rail pressure sensor is malfunctioning.

First thing I tried was replacing the IAC (own idea) and also the fuel pressure regulator and fuel filter after looking up the code but none of these fixed the problem.

I reset the code after replacing these and nothing came back yet other than the CEL blinking 3 times and then shutting off (which indicates a misfire.)

My questions are:

1.) would a regular OBD 2 scanner tell me which cylinder misfire(s) are stored in the computer even though there is no check engine light now? Or do I need a super expensive scanner to see stored code history?

2.) Lets say hypothetically, when I do the scan, if it showed only cylinder 1 and 6 were misfiring for example, then could the fuel pump be ruled out and the injector for the corresponding misfiring cylinder would be the issue? If I’m not mistaken a bad fuel pump would cause all the cylinders to misfire?

Locate fuel injection pressure sensor.

Remove the vacuum hose from the sensor.

If gas leaks out of this connection, replace the sensor.

Tester

Didn’t you already replace the injectors? But not the fuel pump.

Did you compression test it? Clean the EGR passages?

I did replace injectors 3 and 4 but the rest are stock. I managed to get my hands on a scan tool but i have no idea how to use it…it is very old. Below, the screenshots show what i believe to be the results showing which cylinders are misfiring. I have no idea how to read Or interpret the results though

from the numbers in the photos, does this mean cylinders 5 and 4 were misfiring? Because under the “value” column, it shows a number greater than 0 for those cylinders?

I am not familar with that scanner but yes, it seems that 4 and 5 are misfiring.

Do they share the same coil on the 6 pack coils?

Did you do a compression check and did you clean out the EGR?

Compression was normal but that was when i tested it 2 weeks ago. Also I wasn’t doing a live data stream while driving when i was scanning for misfires. I just plugged in the scanner with the car off. So does the scanner just count the misfires from your previous drive cycles then?

I actually replaced the EGR valve but it wasn’t an OEM one. But theres no EGR codes so i doubt that’s the problem

Which makes sense, because cylinders 4 and 5 are adjacent to eachother on bank 2 (drivers side) and the only code I have right now is lean bank 2 or P0174

Getting the whole story out of you is impossible. Like reading a book one sentence a week.

I am out.

1 Like

This misfire issue is not related to the misfire issue that was related to the intake runner control. Car ran fine for a couple weeks then a new issue came up. So that’s why I’m giving you information “one sentence a week.” Because the issues are not related and a new one came up I had no way of knowing was going to happen. You alright?

Yes. I see you’re using an older . . . but fairly high end . . . scanner and you’re using mode 6 data. You’ve already earned a lot of points in my book, for what it’s worth.

I’d forget about the fuel pump. That wouldn’t affect just one bank.

I’m not sure what your fuel trims are on bank 1, but bank 2 has obviously met the conditions to generate P0174. So I’d be concentrating on bank 2. In fact, misfires means excess . . . unburnt . . . oxygen, which means lean condition, last time I checked.

Apparently we can rule out compression for #4 and #5; if you posted numbers in your earlier postings, I must have forgotten . . .

I’m not sure if your engine has a fuel pressure test port. In any case, do you have a way of hooking up a fuel pressure gauge? Do you have a way of “buzzing” the injectors to determine if they’re either flowing too much . . . or not enough?

Something along these lines?

image

I’m also a pretty big fan of OTC tools, as you seem to be

For what it’s worth, I kind of doubt the #4 and #5 injectors are the problem, given the intermittent nature of the problem. I just mentioned the injector tester as a way of ruling out possible culprits

Thanks for the reply. I’m going to do some more diagnosing with mechanic on monday. I do agree with you that the fuel pump is likely ruled out as a bad fuel pump would likely cause both banks to misfire. But I won’t know for sure until I can do a fuel pressure test which I’ll make sure we do on Monday. As for the injectors, I wouldn’t say the problem is intermittent per say just because the misfire isn’t random. There is a clear, definitive pattern in terms of what causes or triggers the misfire, that is, the car only misfires when started on a warm engine but never does on a cold start…But maybe I’m just getting tied up on semantics. Regardless I’ve read and heard mixed things about bad fuel injectors causing warm start only misfires so like you said I’ll make sure we test those too. If the mechanic rules those out, what do you think would be the next step? I know the mechanic will probably know what to do but this is the 2nd person who’s worked on my car and is a bit stumped. It’s always good to have a 2nd opinion.

Hard to say at this point, since I can’t lay hands on the car?

Maybe the injectors are somehow leaking when warmed up?

But if you’ve got it consistently misfiring warm, your mechanic should be able to get to the bottom of it in short order, I would hope

Right now I’ve got a truck in the shop with leaking injectors, which is causing an interesting situation.

Mostly, it doesn’t start at all, because the leaking fuel causes hydrolock

but if it does happen to start, it keeps running and smoothes out just a little bit

But I’ve 100% verified the injectors are leaking

I’ve just got other vehicles to take care of first, before I replace the injectors on that truck

Getting back onto the subject at hand . . . I take it your plugs are once again fouled?

The coil(s) were mentioned earlier . . . I take it you’ve got a coil pack, rather than coil on plug, correct?

it’s not unheard of for coils to fail hot. In fact, I’ve seen that to be more common, versus failing cold

There doesn’t seem to be issues with the car starting - that is it will always start no matter what. However if it’s a warm start, it will crank and start normally and then a couple seconds later start misfiring. I have definitely considered the coil pack. Even though it’s only 2 months old it’s definitely not out of the question and I might have been overlooking it since it’s almost new, which could definitely be a defective part. You are right. The coil pack on this car is just one piece for all 6 wires rather than individual ones. Is there any way test it or not really since the coils aren’t individual?

As for the plugs I haven’t pulled them since the issue started a few days ago but it’s safe to say they’re toast again cause the cars been running lean. However the fact the car isn’t misfiring at all at warm speeds if the engine was cold started and not turned off and back on makes me think the plugs are still good

Would it be accurate to say the car is only running lean when it misfires during a warm start? Or could it possibly be running lean all the time even if the cars not misfiring

You’ve got spark plug wires . . . that means you can use some sort of spark tester

and you can also unbolt the coil pack from the engine and physically inspect it. In some cases, you can see a physical problem, such as a crack or a burnt coil tower

Please refresh my memory . . . the wires were replaced, yes?

Yes I did replace the wires recently. So if you were me, test the fuel injectors, do as fuel pressure test to test the pump and test the coil pack?

Also could I accurately test the injectors by using a scan tool that shows live data from them? Or could they actually be leaking but still show no fault on a computer? So in that case the would the most reliable way to test them be to actually remove them and buzz them?