Is this problem more than an injector replacement?

[JedCoop]

My '94 just turned 100K last week! and it gave me new problem to deal with.

I've gone through the basic troubleshooting (with help from Q45.org & NICO. Thanks again!) but I think there is more to it. The engine runs very well (though maybe not a perfect 100%) all the time EXCEPT...

I'll drive the car so it is throughly warmed, then stop for 1 to 4 hours. When I start it again, it runs horribly. But it does run and can still accelerate. After a few moments the check engine light comes on. I drive on, with idle bumbling up and down, etc. Then, about two minutes down the road later a magic moment occurs, and in an instant the car is running like its good old self again. A moment later the check engine light goes out too.

This behavior has been completely repeatable for a week.

Trouble code on the ECM is 51: Injectors

I did the resistance check on the injectors, 7 of them are fine at 11.7 +/- 0.3 ohms, and one of them is at 20 ohms.

The fact that it is so consistent with engine warm up conditions makes me think that something else is going on. I don't want to replace one injector only to find that the real problem is something else. What else should I check when the car is running OK? Or is there something I must try when I restart the car and the check engine light comes back on?


Modified by JedCoop at 11:24 PM 3/12/2005

[DAEDALUS]

I didn't think 20 ohms will set off code 51 (sure it wasn't a code 45--leaking injectors?). I would guess there was an intermittent bad connection to one of the injectors, but the fact that it only happens soon after a drive would indicate a flooded condition. If you can hook up a pressure gauge, T in and shut off the car/pinch off the lines as soon as it starts to act up. Puzzling.

[JedCoop]

I didn't think 20 ohms will set off code 51

It doesn't... I read 20 ohms when the engine is cold and it will start up just fine. I doubt that varies with engine temp, especially since the problem goes away as if someone flipped a switch. Once flooded doesn't the engine run incrementally better over a short period?

Doesn't seem like an intermittent connection either, since it happens repeatable, predictably. My guess is that the 20 ohm reading just confuses the real issue, it was that way before the almost warm starting problem.

Once I start the car when this happens, the temp gauge still shows warmish, or reads above "C" in a couple of seconds. I am trying to figure out what is different about an almost warm start versus a cold start or fully warm start... or what changes after about two minutes of driving after an almost warm start.

[Q451990]

Back in the late 90s I replaced my injectors with remanufactured units. That caused a start-up problem after a hot soak, like you described. Drive the car, let it sit for a while, and start it and problems... wait all night or turn it off for only a few minutes and it was fine. My problem was only a hard start and a few puffs of black smoke. Yours may be more drastic.

Everyone's best guess is that an injector(s) were bleeding down slowly after the car was turned off. Let it sit for just a few minutes, and it wasn't enough to flood it. Let it sit all night and the fuel would seep past the rings...

Maybe you have a more severe problem with an injector or two? Since yours will act up for several minutes after start-up, I would try to get a power-balance test done to see if you can pinpoint a cylinder.

Heath

[DAEDALUS]

You think maybe an injector is "hanging open" for a bit when warm? A flooded engine will run better quickly, unless the cause is still present. If it occurs in park, is there black soot in the exhaust like Heath mentioned? Or a rich smell? The PBT is not a bad idea. Or check the plugs, especially if they're due for replacement soon anyway.

[JedCoop]

If it occurs in park, is there black soot in the exhaust like Heath mentioned? Or a rich smell?

Black smoke.... more like what Heath describes. Starting isn't hard though. Since I've owned this car for 6 months I don't know the history of all the injectors, though I do know that one or more were replaced.... and it is originally from Florida.

A power balance test - I read that requires the Consult tool, which I don't have. Isn't that the same as the old test where we pulled off spark plug wires one at a time an listenned to what happens?

[3rd Q]

It could be just the injector- I've noticed a couple times that if I just start the Q and pull it around back to park it in the garage and shut it down while it's still in its high idle mode (and running rich) it will have a similar symptom when I start it the next day- goes away after a couple minutes and runs fine. Might be like others have said and the injector is hanging open and is leaking down and flooding a cylinder.

Paul

[JedCoop]

Everyone's best guess is that an injector(s) were bleeding down slowly after the car was turned off. Let it sit for just a few minutes, and it wasn't enough to flood it. Let it sit all night and the fuel would seep past the rings...

Maybe you have a more severe problem with an injector or two?

Heath

I had another thought. Is the ECM smart enough to correlate a flooded cylinder with trouble code 51? If the ECM is smart enough, then this idea sounds good. If it can't, then I am not so sure.

I would expect a powerr balance test to determine clogged injectors, but can a it really determine whether an injector leaks?

If I need to replace injectors, I'll do the plugs at the same time.

[JedCoop]

It could be just the injector- I've noticed a couple times that if I just start the Q and pull it around back to park it in the garage and shut it down while it's still in its high idle mode (and running rich) it will have a similar symptom when I start it the next day- goes away after a couple minutes and runs fine. Might be like others have said and the injector is hanging open and is leaking down and flooding a cylinder.

Paul

My first two Q's both did that. I am now in the habit of never starting the cars without letting them warm up enough. Its a generaly good habit in any case. I've never given this newest Q a chance to do this yet, so I don't know for sure what happens.

The way the car '94 runs (horrible) when almost-warm is similar to what you describe, but the big difference is that the check engine light comes on. My first two Q's were pre-'93s, so there might be some differences in the ECM, dunno.

[DAEDALUS]

Leaking injector should be code 45. A pbt can isolate which cylinder, if any, isn't firing. Presumably you have a miss or at least a bad stumble on a single cylinder, no reason it can't be caused by too much fuel. Symptoms would have to be present when you do the test of course.

[JedCoop]

A pbt can isolate which cylinder, if any, isn't firing. Presumably you have a miss or at least a bad stumble on a single cylinder, no reason it can't be caused by too much fuel. Symptoms would have to be present when you do the test of course.

Any way to do the PBT without a Consult? I can't pull spark plug wires like the good ol' days. To make matters worse once I need to wait over an hour before starting, and then I have only about two minutes to execute the test.

Went back to the FSM. The trouble code 51 is actually "injector circuit". I have an old automotive stethascope - I'll try listening to the injectors today to at least know if they are working.

[911/Q45]

Lots of electrical things fail more hot than cold. Your injector problem is probably an example of this. Try ohming them hot.

[Q451990]

The ohm test while hot is a good idea...

The power balance test does need Consult. The interesting thing is that it shuts down each cylinder one at a time, but it can log how many RPMs you drop per cylinder. The tech should be able to test it at start-up after the hot soak (while stumbling) and get a reading... then test it again and see which cylinder (or cylinders) change.

It may be as simple to fix as a good rail flush if your injectors ohm out properly.

Heath

[JedCoop]

Lots of electrical things fail more hot than cold. Your injector problem is probably an example of this. Try ohming them hot.

Spot-on idea - thanks! The first time I tried the resistance "almost warm" I couldn't get a resistance reading at all from the #7 injector. The Q started with some stumbling for a few seconds, but evened out - the check engine light did not come on. #7 was back at 20 ohms wen I shut the engine off. The infinite resistance could have been my own error.

So I tried again. After a ride on the highway I stopped the car. I measured the resistance on injector #7 immediately. It was about 22 ohms, just higher than when cold. But as I sat there I watched the resistance climb in real time. In a couple of minutes it was over forty ohms. It level out at 57 ohms after 10 minutes, and then drop back down to 20 ohms after an hour with the hood open. (The engine cools much faster with the hood open).

One other injector raised from 12 ohms up to 13.7 ohms and went back down to 12. All the other injectors raised only 1/4 ohm.

So injector #7 is flaky, and flakiness varies with temperature. The proof is in the fixing, but I feel more sure now that this injector is THE problem. I'll probably change out injector #3 (the 13.7 ohm injector) while I am at it, it is inconsistent from all the other six. Under plenum hoses were done about a year or so ago, so this gaskets and hoses shouldn't be brittle. I'll order my parts from Scotsdale tomorrow.

Thanks much for the help! I'll post results.

[DAEDALUS]

You can't rule out the harness till you check the injector directly. The harness is a fraction of what an injector costs, but the plenum must come off.

[Q45tech]

The injector fuse is the +12 volt feed to all injectors, the ecu grounds them sequentially. So you can see with an oscilloscope/active amplified current probe, the current draw from each injector and the shape of the current ramp up as each injector is opened.........at idle or higher rpms........2 millisecs going to 11 milliseconds.

You can actually see the pintle open 1/2 way up the graph ramp [at idle] and measure latency [~~1 millisec] before fuel starts flowing and the shut down [ringing/oscillation] as the voltage/current is shut off and the minute bounce as the pintle slams shut against the fuel pressure [function of internal spring stiffness].

They should all look identical........bad ones show signs in the graph in comparison with the rest.

Same with coils and spark plugs analysing the current/voltage of the primary coils/transistor switches shows the abnormal ones.

Same with CAS dual pulse trains ........360 EQUAL height pulses [maybe not as the 360 slotted wheel gets dirty.......can confuse ecu [why there is the 90 degree magnetic backup]...........high rpm miss from some pulse drop outs........spark scatter [not firing at exact 1 degree point] from ecu losing count.

All engines work the same way.

[Jesda]

One other injector raised from 12 ohms up to 13.7 ohms and went back down to 12. All the other injectors raised only 1/4 ohm.

10-14 is considered normal, but I suppose the variation indicates a slow, eventual failure. Its not enough to cause any noticeable difference in performance. Even with one injector reading 25 ohms (the rest at 12-13) my idle seemed perfect. A few months later that 25ohm injector died completely and was replaced.

[JedCoop]

The injector fuse is the +12 volt feed to all injectors, the ecu grounds them sequentially. So you can see with an oscilloscope/active amplified current probe, the current draw from each injector and the shape of the current ramp up as each injector is opened.........

I love neat tools... Now I want an o'scope, too!

[JedCoop]

I think I've figured this out. High resistance happens when the injector is hotter. So why did my car only run like crap when it sat idle for an hour, and then run fine after two minutes?

Fuel injectors are cooled by flowing fuel. Keeps them from getting too hot. Once the car is running fuel flows past the injectors and cools them. Injectors that are working get cooled more just by spraying liquid into a lower pressure. (evaporation). But all injectors get cooled from the fuel rail itself.

I think this leads to a generally useful idea.... When checking resitance/ohmage of injectors, do it after you have run the car to warm and let it sit for a while with the hood closed to allow all the injectors to warm up to engine temp. As soon as you open the hood, check the resistance of the injectors - I'll bet that'll show up the not-quite-failed injector sooner, even though is works fine when continuously cooled while running.

(for reference, its is generally 65 degrees outside here all the time.)
Modified by JedCoop at 2:53 PM 3/14/2005

[Jesda]

I now have the same problem. It happens only on hot days. After parking for a few minutes, going into the store, then coming back, it runs like crap. There's a strong odor of fuel and it acts like the MAF is bad. I suspect a bad MAF until I read this thread, and think it could be hot injectors.

So, how the heck do you resolve the problem?

Even with a dead injector and some bad ones I never got this kind of behavior. My idle was extremely low when this happened, then was back to normal within a minute as I went down the road.

Could the MAF be too hot? The symptoms are just like that of a bad MAF. Perhaps I need to make a MAF heat shield?

Three injectors were replaced last year.

Thanks

[maxnix]

1.) Since I've owned this car for 6 months I don't know the history of all the injectors, though I do know that one or more were replaced.... and it is originally from Florida.

2.) A power balance test - I read that requires the Consult tool, which I don't have. Isn't that the same as the old test where we pulled off spark plug wires one at a time an listenned to what happens?

1.) Then get one from a dealer or Infiniti Customer Service after you register the car.

2.) It's all done electronically. Hard for humans to count electrons.

[DAEDALUS]

I now have the same problem. It happens only on hot days. After parking for a few minutes, going into the store, then coming back, it runs like crap. There's a strong odor of fuel and it acts like the MAF is bad. I suspect a bad MAF until I read this thread, and think it could be hot injectors.

So, how the heck do you resolve the problem?

Even with a dead injector and some bad ones I never got this kind of behavior. My idle was extremely low when this happened, then was back to normal within a minute as I went down the road.

Could the MAF be too hot? The symptoms are just like that of a bad MAF. Perhaps I need to make a MAF heat shield?

Three injectors were replaced last year.

Thanks

So, it really sounds like the car is running too rich at startup. An increase in ohms, unlikely as that may be, won't make the injectors flow more gasoline. A seal leak, perhaps related to temperatures, is more plausible. If you pull the injector fuse, and crank the car, does it fire up and run for a bit (rough)? Do you still smell the same gasoline odor? Measure the sensed MAF voltage when it's running poorly and compare it to the voltage when it's running well before you decide to make a heat shield. Readings need to be taken at the same RPMs. Usually (not always) it's good to have more than a hunch about the root problem before trying to fix it. Due diligence in diagnosis.

[Qgrappler]

So spec on injectors are 10-14 ohms, and all of yours but one measure measure about 12 ohms, which indicates, possibly, that another way to state spec is 12 ohms, +or- twenty percent.

If you measure 20, that means your resistance at that injector is 66% higher than spec, or 43% higher than upper limit spec.

The injectors work, as I understand it, with a very heavy spring inside which keeps the injectors closed. Its opened by a solenoid, driven by the injector firing current, which depends on the internal resistance of the injector coil and the voltage applied. If the internal resistance rises 66%, the voltage applied, and the pulling power to open the injector is going to be diminished by a proportional amount. Given the milliseconds framework of the timing in an engine, it doesn't take much thinking to see that the high resistance injector isn't going to do what it is supposed to when its supposed to.

The point of this is that the absolute amount of ohms that you are missing spec by is a bit misleading. At 18 ohms you'd be 50% above the spec, which has a dramatic impact on the current to the injector solenoid and the force applied to open the injector valve.

Actually, an increase in resistance ought to make the flow more lean, but if the resistance has increased BECAUSE of mechanical wear in the injector, its hard to guess what lean rich situation could be in the cylinder.