1992 Galant ECU Problem or not?

Hi,

I drive a 1992 Mitsubishi Galant (2.0l GLSi DOHC) - at least I used to until it suddenly stopped on me (of course, in a narrow one-way street with lots of people behind me and only a handcapped parking space nearby. *sigh*). The engine just went out and I was unsuccessful restarting it. The AA guy that I called said that the sparks were infrequent and too weak. I checked that at home and the sparks are somewhat infrequent. I not sure if they are too weak, too. (Spark plugs are fine.) After reading about the ECU caps problem on the Internet, I checked the ECU. (I could sometimes hear the "clicking of death" from under the dashboard - a sign that the ECU is constantly resetting itself.) One capacitor had leaked some electrolyte. I orded the caps and replaced all of them. Out of curiosity I checked the capacitors that I removed and they were all performing within specification - even the one that had spilled some electrolyte.

*sighagain* Needless to say, the car didn't start even after replacing the caps, so I investigated the ECU error codes. When the car is off and I turn the key to "ignition" (2nd position of the key, i.e. I don't start the engine) I can hear sounds coming from the engine compartment - sounds like a valve frequently opening and closing to me. After about 30 seconds that sound stops, after another 30 seconds the "Check Engine" light comes on. ECU error code reads "25" - Barometric pressure sensor faulty. If I start the car shortly (i.e. crank just for half a second or so) the error code remains 25. If I crank for a little longer (a second or more) the error code reports a faulty ECU (constant signal) - that doesn't go away unless I switch off the car. The above behavior is reproducible even after a disconnect of the battery. I suspect the ECU processor to stall if I crank for more than a second or so. (At one time, the fuel pump stayed on after the ECU "failed" and didn't turn off at all.)

So, here are my questions:

- Any ideas, anyone? ;-)

- What about the code "25" - does that always happen if you leave the car on "Ignition" for too long without starting it? That error can't be produced by a faulty ECU. So either the sensor is faulty or it's a connection problem. Or a power supply problem.

- The ECU runs fine at first. Why does the ECU stall after cranking for a second or so? Faulty power supply? (I don't suspect a loose connection). That'd lead to back the ECU and a faulty component.

- I've heard that besides the capacitor problems other components in the ECU tend to fail, too (a diode was mentioned quite frequently...) Any hint on what components to check? Checking the components is no problem for me - it's just either complicated (ECU in the car under operation) or complicated (ECU not in the car - desoldering components neccessary...).

- Does anybody know a way to buy a rebuilt ECU within Europe? Any rebuilders?

Thanks a lot for any advice (and sorry for the long post... ;-) Norman

Reply to
Norman Weiss
Loading thread data ...

Assuming you've reset the error code in the ECU, what code is output after the reset, when you first turn the ignition key to the "run" position. That's the position just before the engine would crank over. Before installing the new capacitors, did you check the circuit board for damage, such as a corroded or broken trace. I've found broken traces on the circuit board before. It might be that under a certain load the break opens up. Try looking around the center of the ECU board, near the large electrolytic.

Yar

Reply to
Nirodac

Hi Norman! You may wanna read my topic: "90 Eclipse GSR4 repaired ECU diagnosis". Still didn't fixed my ECU, but I've read a lot about rebuilding one. I might be able to help. You've said your "ECU runs fine at first"! I don't believe that. Either functions, either not. You can see that when mine blow-up, nothing was working anymore: no spark, no fuel, no pulse to injectors. After replacing all the caps (not only the damaged one) and cleaning the entire board (which is very important), you definitely should look for circuit breaks (use a magnifier and bright light) as Nirodac mentioned. When finished that, you can test again the board. If you ask me, ppl should do this every 3-5 years, or at least when car is behaving strange. If the test fails in any way, means there are other components damaged by using the board when eated up by the electrolyte leaking from the caps. In my case must be at least one transistor. To test transistors you have to desolder them, check their resistance when not power-supplied and their continuity when power-supplied. I suggest using a multi-tester where you have a special place to stick transistors, both PNP and NPN. Transistors can be easily damaged by a short when high-intensity current goes to their base. If you can mail me some pics at boa-at-mapsuae-dot-com, I will do my best to help you. If your board is in fact not faulty (from this distance I believe it is), you may wanna check your sensors, starting with the Crank Angle Sensor. If you choose to buy a rebuilt unit, let me know to give you some links of companies doing that, but it's a bit risky to not check all the harness wiring and relays before spending money on a new unit. You might damage the new unit too if a short in the wiring is present, or you might find that your old unit is working too after replacing a damaged relay. Good luck and will talk more some other time. Hope this helps.

Reply to
Bornish

Well, I assume the ECU is stateless, so disconnecting the battery clears everything? I did that, the bevavior stays the same: code 25 in "Run" position, constant signal after cranking for a second or so.

Well, one of the capacitors spilled some electrolyte and hit a SMD capacitor and a diode post with it. It dissolved the PCB finish, but as far as I can tell didn't damage any traces. The diode is fine I think (I resoldered it), but I'm still worrying a little about the SMD capacitor. On the other hand, I doubt that it could produce these symptoms.

Norman

Reply to
Norman Weiss

I did and I've posted some comments.

Well, you can trust me on this one. The ECU processor runs if I put the ignition switch on "run" (not on "ignition"), because it outputs an error code after a minute or so. I only fails after I crank for longer than a second or so.

fuel, no pulse to injectors.

Right. Mainly because one of you capacitors exploded. Since the three or four caps on the ECU PCB are for the power supply, that means that your ECU suppy voltage was too low to properly run the processor. Now that you've fixed the caps I runs again, but now you see the symptoms of faulty components - like I do.

entire board (which is very important),

light) as Nirodac mentioned.

None found.

using the board when eated up by the electrolyte leaking from the caps.

I doubt that any component was damaged by the electrolyte. I think a component blew because of the improper operation of the ECU. Maybe the electrolyte leak and the fault are not related at all!

power-supplied and their continuity when power-supplied.

Right. But that's a PITA, so I wondered if there are any bets out on what ECU component is likely to fail besides the caps.

See you PM, please.

may wanna check your sensors, starting with the Crank Angle Sensor.

I suppose that's fine, since the ECU doesn't tell me it's broke... ;-)

Thanks, Norman

Reply to
Norman Weiss

I would definitely start checking the sensors as mentioned earlier, including the TDC and crank sensors in the dist. I believe Mitsu is using optical sensors to detect rotation. Also you may want to check all the power feeds to the ECU as well as the grounds. If you leave the ignition switch in the run position for a few minutes, without cranking the engine, does the ECU output an "failed ECU" code? If it doesn't, try heating the ECU with a hair dryer, and see if that causes it to fail. You may have a thermal intermittent, when a component heats up, it may fail.

no fuel, no pulse to injectors.

entire board (which is very important),

light) as Nirodac mentioned.

behaving strange.

using the board when eated up by the electrolyte leaking from the caps.

not power-supplied and their continuity when power-supplied.

transistors, both PNP and NPN.

goes to their base.

you may wanna check your sensors, starting with the Crank Angle Sensor.

companies doing that, but it's a bit risky to not check all the harness wiring and relays before spending money on a new unit.

you might find that your old unit is working too after replacing a damaged relay.

Reply to
Nirodac

I still don't see how checking the sensors will help me. A defective crank sensor is signalled by the ECU, if I'm not mistaken. And it wouldn't cause the ECU to fail, or? (Starting the car will cause the ECU to signal "Defective ECU")

That's going to be my next step - but the ECU is well grounded (bolted to the car body), leaves the 12V input. Does anybody know a little about the ECU power supply, i.e. the 5V voltage regulation circuit? I'd like to know where to test the 5V for stability.

That sounds like a good idea. I'll do that, too.

There's one thing left where I could use some help from a Mitsubishi owner, preferably a 89-94 model, even more perferably a Galant: If you leave the car in the "Run" position (without starting), does the "Check engine" light come on after a minute or so?

Thanks, Norman

Reply to
Norman Weiss

Sorry, may have misread your other posts I believe you had said something about the injectors not getting pulsed, but you had an output from the ecu.

Reply to
Nirodac

MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.