1996 Montero 3.5L DOHC misfire (6G74 engine)

Hello All!

My car suddenly lost 2 out of 6 cylinders. I hooked my oscilloscope to camshaft and crankshaft sensors, and the ECU output controlling the 3 coils (waste spark system). You can see that one of the ECU outputs is a flat line, while the other seems to be firing more frequent then needed. As if the ECU thinks it must fire the same coil two times in a row, instead of the correct one. I was wondering of you can advise me if the crank/cam signal looks good. Maybe there is something wrong with the sensors that makes the ECU think that it must fire the wrong cylinder?

Here are the readings:

Channel 1 should be crank sensor, channel 2 should be cam sensor. Channels 3,4 and 5 are ECU outputs controlling the 3 coils.

Or maybe the ECU needs replacing? Any advise would be helpful!

Thanks in advance!

Hook up scope on BLK/GRN, RED/BLU and WHT wires (#2, #1 and #3 PINS on the IGNITION POWER TRANSISTOR)

Look for square waves coming from PCM. If you're missing square waves, then possible bad inputs to PCM (cam crank).

If square waves present, then suspect faulty IGNITION POWER TRANSISTOR?

Funny it looks like the ramps are all there just on the wrong wire lol.

veliko, can you get us a more detailed view of the cam and crank signals? It's tough to be certain, but it sure looks like it's out of time. The big cam pulse appears to be intersecting two crank pulses, when it shouldn't intersect at all.

Been there, done that. Timing issues will make the ECM do goofy stuff on this system. Like firing one coil twice as often as it should. :silly:

Hi everyone!

Thank you very much for your replies!

MaroRussinovich, that is exactly what I did. Only I hooked the probles on the PCM unit connectors. The 3 top channels are just that.

Tyler, I'm attaching a more detailed view. Actually, this morning when I started the car, all 6 cylinders worked. I managed to quickly capture the waveforms before the engine acted again.
I'm sorry the setup is a bit different now (I was in a hurry) - Channel 1 is cam sensor now, Channel 2 is crank sensor, and the rest of the channels are the coils. These below should be good waveforms of Montero/Pajero 3.5L 6g74 engine.



After I stopped the engine, and started it again, the misfires appeared again.



To my eye, the crank and cam sensors look fine in all cases, so sadly, the problem must be the ECU.
I've replaced the electroilytic capacitors in the ECU, cause I read on internet they go bad with time, but nothing changed. I'll try to resolder all ECU joints, but probably must replace the E310a IC as shown here:


Weird enough, only 2 of the ignition control pins go to this IC. The third goes to another component.

If you have any ideas what outside of the ECU could be causing such problem, please feel free to share.

Thanks again!

Thanks for the new captures! DEFINITELY out of time. I would recommend at least checking the timing marks on the cam and crank before doing more work to the ECU.

If it's in time, there may still be an issue with the crank reluctor. It's held to the crank sprocket with roll pins, which have a habit of breaking and allowing the reluctor to spin on the crank. Thr lower timing cover must come off to confirm this.

Hello Tyler,

Thanks for your input! I'll have my mechanic check the timing and reluctor wheel.

It's weird though that the good and bad waveforms look exactly the same. I've made an overlap of the bad and good waveforms, and the cam+crank sensor waveforms are exactly the same. I've had to stretch the good waveform to match the engine rpm. What makes you think that the timing would cause the misfire?


Thanks again!

veliko wrote: It's weird though that the good and bad waveforms look exactly the same. I've made an overlap of the bad and good waveforms, and the cam+crank sensor waveforms are exactly the same. I've had to stretch the good waveform to match the engine rpm. What makes you think that the timing would cause the misfire?

I think it's all down to the programming of the ECU and where it looks for the CKP and CMP transitions to occur. With the signals out of time, the ECU is likely misidentifying which cylinders are coming up to TDC, and getting the coil firings wrong as a result.

Cases like this are what makes it hard for guys like Paul to generalize what happens when cam and crank signals get screwed up or go missing. :lol: Some cars won't start. Others set timing/sensor codes. These Monteros? Doubles up on one coil and forgets the other. :silly:

I hate to be picky, but did you accidentally invert channel one (CMP) in that last set of captures? Because it went from three small pulses and one big pulse, to three big pulses and one small pulse. Mostly want to make sure that's what happened, and not something goofy going on with the sensor between captures.

For reference, this is a known good:



You can see how the transitions of the CKP and CMP sensors never line up. You can also see how the big pulse of the CMP neatly brackets a pulse of the CKP. Yours never line up, inverted CMP or not.

As for why the misfire came and went, it may be that the signal relationship is moving just enough to make the ECU temporarily happy. :huh: The fact that the ECU controlling the power transistor correctly tells me the hardware itself is probably fine. If you want more reassurance, you can always grab some jumpers and rapid-fire tap the CKP and CMP signals to ground, KOEO. You should be able to get all coils to fire.

Hello Tyler,

Thanks again for your reply!

I understand, sometimes a microsecond can tip the ECU trigger, so maybe the waveforms are on the edge of good/bad now.

About the inverted channel, I noticed that one too, but come to think about it, I always put the probe's black tip on ground, and the red tip on the ECU pin. Maybe one of my probes have inverted wires/colors. I'll check on that to be 100% sure.

Thanks for the reference waveform! Is that from a 6g74 engine?

Thanks!

Hello All,

I'd like to update you on my problem. After my mechanic checked the crank sensor, it was evident that the reluctor wheel is off. The reluctor roll pins broke, and in the process damaged the crancshaft gear. This must have happened last summer when my crankshaft damper pulley was loosened due to unscrewed bolt (how could that happen is beyond me). After he changed the gear, the reluctor wheel, and several other related components, my engine is running fine again

Attached are the waveforms of my engine now.

Tyler, thank you very much for your advises, you saved me a lot of bucks for a new ECU! I am a php programmer by profession, so if I can give you any advise regarding websites, I'll be more than happy to do it!

Fantastic to hear! :woohoo: The new cam/crank waveform is lookin' good.

Those tiny little roll pins for the reluctor are pretty useless, huh? :silly: Wish they'd have made the reluctor part of the crank sprocket itself, like most other makes.