Electrical issues following wash - need help solving mystery

I had to drop everything and go on a 2-week worktrip, in the meantime I had the starter freshly rebuilt. Both the starter itself, and the solenoid, were toast.
Installed it, and although the enthusiasm and note of the starter is crisp, new, and robust, the halting labored fast-slow-fast-slow pattern remains.
So, the failing starter, was not causing a bizarre electrical drain that affects the ignition.
I am now on to the crank position sensor, and I do have another that I can swap in if need be, but I will be testing the current one as thoroughly as possible.
If I plug in the uninstalled sensor, the cranking is smooth, fast, and consistent (although it is not contributing a crank position signal) and behaves the same way as if the currently installed sensor were unplugged.
I took the plastic cam gear cover off, and the timing belt looks good and tight.
 
Last edited:
Ok, tested the Crank Position Sensor as per this recommendation:

How To Test The Crank Sensor (1997-2004 3.0L Montero)

Mitsubishi 3.0L Index of Articles
easyautodiagnostics.com easyautodiagnostics.com
so basically, probing the signal wire coming from the sensor that goes to the ecu.
It showed that there were three distinct 5v signals being sent from the sensor, alternating with three periods of very low voltage, as the three lobes of the sensor signal plate passed through the sensor, per revolution of the crankshaft, as cranked slowly by hand with IGN on (but coils unplugged).
So it is not the crank position sensor.
It could still be that the pins holding the sensor signal plate have broken, and the plate is shifted somehow, sending improper crank position... but this is highly unlikely.

I am not hearing the fuel pump priming (with IGN on), does anyone know if it will prime for a certain length of time everytime IGN is turned on, or whether it only does it if pressure has dropped low?
Definitely smell fuel from the exhaust.
According to both the 1998 Montero manual, and the Magna manual, fuel pump runs only during cranking and engine running (does not mention IGN-On priming).
I also cannot understand how a no-fuel situation could make the engine more difficult to turn over. It should crank consistently and smoothly if the fuel delivery was the issue...

Again, when I say "intermittent difficulty turning over", I mean that with IGN disengaged (via either unplugging the IGN control connector to the coils, or the Crank Pos Sensor) the engine cranks quickly smoothly and perfectly. With IGN operational, it randomly quickly goes from cranking perfectly to slowing and dash lights dimming for a fraction of a second... the starter is fighting what can only be the engine firing at the wrong time, partial ignition in the cyls. It is why the original starter, albeit on its last legs anyway, was working so hard just to do normal turnover, and died. If anyone wants video of this difference, I can post it.

Having eliminated: engine codes, fuses, main relays, battery, charging system, starter system, ignition system, grounds, power, sensors, timing belt, engine cyls (broken rods/pistons),
I am now turning my attention to the ecu itself. Have found one to buy, and will be swapping it out to test.
 
Last edited:
Does that engine have variable valve timing? That makes more sense to me than partial combustion...
 
sk66 said:
Does that engine have variable valve timing? That makes more sense to me than partial combustion...
Click to expand...
What we have is definitely an ignition issue. I tested the sparking, and it is very weak, inconsistent, from all 3 coils. It is not strong enough to trigger my timing light. I also ran a brand new spark plug wire to a brand new spark plug, and observed spark directly, same problem duplicated.
- the coilpack has been swapped, no difference.
- the ignition control module (ignition power transistor) swapped out, no difference, and harness wiring verified
- ecu swapped out, no difference, and other ignition-killing sensors (crank position sensor) verified ok
- extra grounds to block added, no difference, extra grounds (duplicating the strapping) to coilpack mount added, no difference
- no trouble codes
- the fact that the spark is very weak, also rules out mechanical fails like, crank position sensor trigger plate broken/slipping
 
This is also difficult enough, without finding faults in the cobbled-together manuals, like this in the 1998 Montero ignition power transistor testing section :/
1998 Montero Manual Ignition Testing mistake.jpg
 
I waited until dark tonight to test startup again, to look for any shorts and arcs that shouldn't be there. After finding none, I hooked up a new spark plug wire and spark plug to each coil one at a time, replacing one of the spark plugs so that only 5 were hooked up to the engine at the same time, and the new one was out in the open with a ground wire to the battery. Coil C (to the right as looking into the engine bay) fired that plug what looked to be once per engine revolution... that's about a spark every 1.5 secs, and it did it consistently. Coil B (in the center) did the same, consistently firing the plug once per revolution.
Coil A (the coil to the left) fired that plug fast, 3x as fast as the other cyls... looks like it fired that coil about 3 times per engine revolution... hotter and brighter than the others.
Does anyone know if the Delica has a type of "batch fire" or "quick fire" for startup designed to help for quicker starts? I know some Hondas have a startup "batch fire", mid-90's Civics for sure. I understood that to mean briefly firing all cyls at once... so this would be different.
Just kinda 'firing' from the hip here, but, that could be robbing the other cyls of simultaneous spark power and possibly firing cyls 2/5 at the wrong times, fighting the correct cyls?
Again, none of this is picked up with my timing light. *Edit: later on, I bought a new timing light that verified the odd different sparking for each coil as described above.
 
Last edited:
Sorry you're not getting any help on this... looks like you're doing the exploration for those to follow...
 
As stated above, you are pretty deep and past some of my electrical knowledge. Only thing I can think of is back to the primary side of 12v, are you getting good voltage/current to all of the ignition components? IE ignition switch and relays supplying reliable power? Good grounds?
 
DustAngel said:
As stated above, you are pretty deep and past some of my electrical knowledge. Only thing I can think of is back to the primary side of 12v, are you getting good voltage/current to all of the ignition components? IE ignition switch and relays supplying reliable power? Good grounds?
Click to expand...
The voltage supply to the coils is indicated in red in the 2nd pic of post #16

Electrical issues following wash - need help solving mystery

Hello all, thanks for reading. L400 S2 ('98) V6 has been nearly flawless for 5 years. So here is the progression. Washed the van, using pressure washer (just externally, not in the engine bay lol). Van runs/starts fine. Two days later, it is dead in the morning. So something out of the blue is...
delicaforum.com delicaforum.com
It tests good at 12v with IGN-on, drops voltage a bit during cranking as expected to about 10v.

I do not have diagrams that can indicate all of the grounds. Or relays. I have checked contact for the main group of grounds that ground right below the ecu, but I do not know to what they belong and cannot therefore trace them. The main battery ground is cleaned, and was (temporarily) doubled up for good measure, at the battery and at the body and the same for the starter.
The ground that is part of the ignition harness (power transistor unit, or electronic control module pin #4 of the 6) is good, and I ran a new one completely just to make sure it was truly independent, no difference. The ground straps on either side of the coils are good continuity, and I doubled them up to the battery too to test. Since the starter has no trouble cranking without the ignition system I know we can draw plenty of battery power.
I need to figure out what is causing erratic ignition sparking that does not appear in any order. On the two days that the van suddenly ran perfect again, the timing looked good. I haven't met a timing belt that can jump out a tooth or two and then leap back to correct position and then leap out again :)
 
Last edited:
New update: I bought an Innova OBD2 scanning tool, model 3100, and it is unable to establish communication at the port, ignition in the ON position... going to try and clean the port and check for obvious wiring issues there.
??wow?? ... really not catching any breaks.
 
It's JOBD.
AFAIK, no one has found an available reader that works with it. There are some that are supposed to be compatible with new versions of JOBD, but it seems they are not backwards compatible.

JOBD diagnostics; error codes

Copied from delicaclub.ca: 1994+ Mitsubishi Diagnostic Trouble Code Retrieval Without Scan Tool Ground Data Link Connector terminal # 1. Pin4 and pin5 are ground so you can connect pin 1 and pin 4 with a paperclip or short wire (or use another ground point if preferred). Set the ignition...
delicaforum.com delicaforum.com
 
Ok, latest update.
Picked up a new timing light, my old one seemed to have inconsistencies, the pickup seemed to have trouble, found a crack in the pickup. New one is consistent.
But that only deepens the mystery.
Because the laboured cranking is very hard on the starter, cranking ease and consistency is restored if either fuel or timing is missing.
So, if I disable the spark by unplugging the 3-wire plug from the ignition control module (power transistor unit), cranking is easy, smooth & fast.
If I disable fuel by unplugging the fuel pump relay (this takes several seconds of cranking to clear all the fuel out of the cylinders) the cranking also becomes easy and smooth & fast. The laboured cranking is due to partial combustion and really bad timing or spark pattern or both.

It seems much easier to check timing with the fuel pump relay disconnected, I feel bad for the starter otherwise, and the power draw weakens the spark because the starter struggles. But regardless, I cannot see the timing mark at all when timing light is hooked up to cyl 2/5. The white mark is clearly visible on the main pulley, so I should be able to see it. In fact, I cannot see the timing mark at all with timing light hooked to ANY plug wire.

At the master cylinder/brake booster, I only have two check connectors: a black one (blue-black wire) for the fuel pump check which works perfectly when 12v applied, and a blue connector with a white/black wire, which I believe should be the timing check connector. Regardless, I cannot see the timing mark at all whether this check connector is grounded or not.

And, the strange sparking (unequal spark patterns on all 3 coils) continues.
- coil B (center) is a slow blink, perhaps once per second
- coil C (to the right as looking into the engine bay) fires twice as fast.
- coil A (the coil to the left) fires 3 times in the same amount of time, in a rhythm though, b-blink blink b-blink blink b-blink blink

And it does all the above no matter if I swap ecus. Or swap coils, or wiring to the coils.
Swapped out the cam sensor, just in case, no difference.

When the van magically ran perfectly for two days in the middle of all this, the timing looked normal, visible above the timing marker plate clockwise a ways (without the check connector grounded). So how can a mechanical timing issue go wrong, perfectly right itself, then go completely wrong again? Combined with the unicorn sparking pattern, it seems to point to an ecu-controlled spark-timing issue.
What would be the sensors or parameters that influence ignition timing? Does the ecu think the temperature is -400F?
- coolant temp sensor
- IAT sensor
- ?

So it almost seems if something is very significantly altering the timing. Perhaps something is shorting somewhere that has a drastic effect on the ecu's treatment of timing, making it freak out.
Just for due diligence I am going to verify tomorrow if my harmonic balancer has slipped on its rubber, compare the location of the timing mark to the location of the keyway. Timing mark should be just shy of 90-degrees counterclockwise from the keyway... not that that would actually change timing, but it would change the appearance of it. I'll also look to see if the cam timing mark lines up with the crank one.
 
Last edited:
Isn't the 6G72 (3L V6) an interference engine? If the timing was off that far wouldn't it mash valves?
 
sk66 said:
Isn't the 6G72 (3L V6) an interference engine? If the timing was off that far wouldn't it mash valves?
Click to expand...
Yes from what I have read, if it is off enough. But, if the ecu is altering it within its range of capability, perhaps not.
Other thing is, if the crank pulley has slipped on the damper ring (for the harmonic balancer) it would appear to be off, but that wouldn't actually change timing, just move the timing mark...
I'll go and eliminate that variable right now.
I'll also check to see if the check connector (blue with a white/black wire) that I am trying to ground actually has power to it...
 
The only thing that makes any sense to me is that the engine is firing early (at/slightly before TDC)... but if that was the case it would still run (until it destroys the engine).
As far as I can figure, that is the only condition that disabling either fuel or spark would correct/prevent.
 
sk66 said:
The only thing that makes any sense to me is that the engine is firing early (at/slightly before TDC)... but if that was the case it would still run (until it destroys the engine).
As far as I can figure, that is the only condition that disabling either fuel or spark would correct/prevent.
Click to expand...
Now, think about it like this: if the mechanical is all sound, because this condition occurred, then went away 100%, then returned exactly as before... the timing belt, timing components, crank, valves, cyls etc... are all fine, and no chance of internal damage exists.
The components of the ignition... the ecu, injectors, ignition control module, coils, wiring to coils, spark plug wires, plugs etc... are all sound, but firing the wrong signals at the wrong time.
Either, both ecu's have coincidentally independently failed in the same manner (one of which corrected itself for two days, then faulted again)
or:
- there is something else that is shorting or influencing the cam position sensor signal as it goes to the ecu, shifting it and resetting it every revolution, so that we get six cyls firing, but each coil fires something different. And it is not happening at the coils.
- there is a ground somewhere that is ecu-related, that is not contacting, and it affects the way that the ecu perceives the signals it receives.
- there is a relay that can mess with signals being sent to the ecu, so that every cam revolution, there is an unwanted grounding or power application which resets the TDC notion (interrupting normal ignition firing and starting over in a looped flawed pattern).
- and the fault is happening in a manner which cannot be detected via continuity testing of the harness to ecu (pins to/from the ignition control module 11,12,13, for the coils, the pin for the crank pos, pin for the crank pos) and they aren't grounding out to the other wires within their respective plugs. I do not yet have a definitive ecu pinout diagram for this vehicle.
 
Verified that the main crank pulley and harmonic balancer are fine, the mechanical aspects of timing appear perfect, the main pulley and right cam sprocket timing marks are perfect. Belt seems due for a change soon, but cannot see any reason to believe that these are the cause. I didn't snap a pic of it, but the large tooth of the signal plate behind the cam sprocket is about to/beginning to enter the cam position sensor.

IMG_E9570.JPG

IMG_9571.JPG
 
Last edited:
You must log in or register to reply here.
Top