Jump to content


Welcome to Ultimate Subaru Message Board, my lurker friend!

Welcome to Ultimate Subaru Message Board, an unparalleled Subaru community full of the greatest Subaru gurus and modders on the planet! We offer technical information and discussion about all things Subaru, the best and most popular all wheel drive vehicles ever created.

We offer all this information for free to everyone, even lurkers like you! All we ask in return is that you sign up and give back some of what you get out - without our awesome registered users none of this would be possible! Plus, you get way more great stuff as a member! Lurk to lose, participate to WIN*!
  • Say hello and join the conversation
  • Subscribe to topics and forums to get automatic updates
  • Get your own profile and make new friends
  • Classifieds with all sorts of Subaru goodies
  • Photo hosting in our gallery
  • Meet other cool people with cool cars
Seriously, what are you waiting for? Make your life more fulfilling and join today! You and your Subaru won't regret it, we guarantee** it.

* The joy of participation and being generally awesome constitutes winning
** Not an actual guarantee, but seriously, you probably won't regret it!

Serving the Subaru Community since May 18th, 1998!

Guest Message by DevFuse
 

Photo
- - - - -

In-depth informations about EA82 ECUs


  • Please log in to reply
31 replies to this topic

#1 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 22 June 2007 - 07:02 PM

I have a damaged MPFI ECU here which was fried when the previous owner of the car confused the battery connectors...
So now, the ECU seems to work, the LED is blinking and the starter cranks, but the engine won't fire on, the cause seems to be no spark, could be the fuel pump, too.

I was told that the ECU is always damaged when the battery connectors are confused since it's directly and always connected to the plus pole through the fusible links for storing the fault codes (you can check this with the FSM)

So, something was damaged inside the ECU but not that terrible that the basic functions aren't working anymore. But what is damaged?

I have opened the whole unit and the electronic isn't that complex (as you can guess with a design of the 80ies) that you can't fix it by yourself, but without better informations or a cicuit diagram this would take some real amount of time. So, are there any in-depth informations about the ECUs? Or has anyone even find that faulty electronic component?

Main cause for doing the work: ECUs are rare and expensive here in Germany and I have some kind of power problem with my non-Turbo MPFI car/engine which is now driven by a Turbo MPFI ECU which is possibly related to using the "wrong" ECU. So I want to rule that out before searching for the error...

#2 Petersubaru

Petersubaru

    1000+ Super USER!

  • Members
  • 1,088 posts
  • Sarnia Canada

Posted 22 June 2007 - 10:22 PM

I would believe that it would be much easier to purchase an ECU from someone here on the board..maybe $15-25 plus shipping..at least this is what I paid for mine which I keep in the trunk as a spare...I think rockauto.com had them remanufactured for about $100

#3 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 22 June 2007 - 10:35 PM

You're not serious? :eek:

These parts cost about 250+ US$ here in Germany (more than a whole car btw) and are more rare than rear break discs! Ah, and this are used ones ;) I just live in the wrong country...

#4 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 23 June 2007 - 02:50 AM

yeah, ECU's for an EA82 are a dime a dozen around here. it's too bad i didn't know, i could have thrown one in that box i just sent you.

ECU's do not always fry when the battery connections are crossed. in my experience, the fusible link fails more often than the ECU and the ECU occassionally fails.

i've heard of people opening them up and finding the failed part...pretty sure it was a resistor or was that transistor? the information may be on this board, have you searched?

i think i fried one last year, when i opened it up i could not find any particular parts or traces that were damaged. unfortunately mine is a freak of EA82's and has been a bear to deal with ever since. it's a unicorn 1987.5 XT Turbo that i converted to non-turbo. swapping electronics on an XT of that vintage is not a fun task.

#5 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 23 June 2007 - 03:01 AM

oh right, your box is in the mail, i forgot to email you i believe.

#6 daeron

daeron

    Cunning Linguist

  • Members
  • 3,600 posts
  • West Palm Beach

Posted 23 June 2007 - 04:38 AM

You're not serious? :eek:

These parts cost about 250+ US$ here in Germany (more than a whole car btw) and are more rare than rear break discs! Ah, and this are used ones ;) I just live in the wrong country...

I want to clarify something here, a non turbo MPFI ECU may be less simple and more expensive, even for a statesider, to obtain.. BUT we DO at least have a chance to see one in a junkyard. If I were you, I would make a post, with a subject of "Are you in North America? do you go to junkyards?" begging for someone to find and ship you an ECU. Even though it is a "parts wanted" post, I think that your current situation would justify posting it in the "older gen" tech forum, where it will get more views.

Are you the German who just last summer used a turbo ECU in his non turbo car to get it home??????? if NOT, then I have you confused with another german member who swapped turbo/non turbo ECUS (i forget whether his car was turbo or the ECU he installed was a Turbo) with no problems a year or so ago.. SO, if you are NOT the person I am thinking of then there is a decent vote from one of your own countrymen on this forum that says that the turbo/non turbo ECU question is urrelevant.

#7 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 23 June 2007 - 04:44 AM

nope, same guy. this is the legend!
what vehicle is this for again? it's a loyale right?

#8 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 23 June 2007 - 08:00 AM

Yeah, it was the same guy ;) What an adventure: Travelling through the whole south of Germany with the train and one rucksack + one big bag full of some car parts (alternator, starter, ECU, battery... man, was it heavy ;)) and fixing that car up alone somewhere in the middle of nothing :) And the neighbours were shocked when the car moved away with its own power after standing and rusting there for at least 1.5 years... :grin:

And yes, I was able to get this car home with no problem, but I noticed some kind of lacking power between about 2.500 and 4.000 RPM, but since I was using highways most the time without many hills and avoiding cities, I really didn't need to accelerate much and I believed that this was some kind of imagination or some specific problem with the car. But it runs well with about 32 mpg and over 100 mph (we are allowed to drive that speed on some parts of the highways). So I thought it was perhaps related to some other part of the car or to the missing knock sensor (which the MPFI engine is lacking and the Turbo-ECU wants it)

Yeah, but now that engine is in my GL-10 (should be match the US types best) , together with that Turbo ECU because the original one isn't working, and I notice this lack of power, too. And everything is different besides the engine and the fuel pump. I installed the knock sensor some weeks ago, changed oil, spark plugs, ignition cables and ignition coil, but the problem is still there. And before getting more into this I want to rule out that it could be related to the ECU although I doubt that, but you never know... And fixing it up should be very easy if you know what part is damaged, so I really didn't search for a spare ECU...

The search function isn't really helping because searching for "ECU" gets no results, I guess words with three letters are too short.

Hmm, the "power problem": If you accelerate up from 1.000 rpm everything is fine until about 2.500 rpm: There you feel some kind of power loss, it feels like the car isn't getting enough fuel and it bucks a bit, but not much. The car still accelerates but slower. And starting with about 4.000 rpm, it's like a boost and runs fine again. It's not really bad if you drive normal, but it's worse when you try to climb a hill... And we have some decent ones here around ;) And I want to fix that up together with the other parts of the car during summer.

@grossgary
Thanks for shipping! You indeed forget to write about it ;)

#9 Cougar

Cougar

    Elite Master of the Subaru

  • Members
  • 6,340 posts
  • Anchorage

Posted 23 June 2007 - 08:59 AM

I have repaired a few ECUs from models like yours in the past. A couple of them needed new pre-driver transistors (TO-92 style) to get the fuel pump circuit running. The second unit had the same trouble so it took less time to repair and the first one didn't take very long to locate the bad transistor.

The fuel pump circuit won't turn on unless the ECU sees pluses from the distributor CAS circuit. You can check the fuel pump circuit by listening for the pump when you turn the key to the RUN position. It should run briefly. If you don't hear it at first try cranking the engine briefly and then try it again. If you still don't hear it then something is wrong.

Since the error light is working hopefully you are getting a code 13 which would indicate a problem with the CAS sensor in the disty.

#10 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 23 June 2007 - 09:50 AM

stickedy, try a place holder like this to search ECU* or *ECU and see if that helps, it likes at least 4 letters for the search. sounds to me like getting that ECU figured out will likely solve your problem and mine.

I have repaired a few ECUs from models like yours in the past. A couple of them needed new pre-driver transistors (TO-92 style) to get the fuel pump circuit running..


could you elaborate on where and how to locate that transistor. after crossing up the battery cables, if i wired the fuel pump directly my XT Turbo would run fine. if i didn't the computer didn't blink any codes and the fuel pump would not prime. not too mention this vehicle would run (not well, but it would run) on an XT6 (yes a 6 cylinder ECU in a 4 cylinder vehicle). it ran enough to tell me that the engine and other sensors were working fine.

anyway, i opened the ECU and couldn't find anything obvious...but i am very novice at electrical boards and such.

#11 Cougar

Cougar

    Elite Master of the Subaru

  • Members
  • 6,340 posts
  • Anchorage

Posted 23 June 2007 - 10:31 AM

The test method I used to find the problem was using the diode function of the digital meter and checking the forward and reverse directions of the transistor junctions. I can't remember now if the junction was open or shorted.

Your unit may have a different problem since the polarity was reversed. My unit just quit after the alternator was replaced. I think the higher output of the new alternator took out the weak'nd junctioned.

#12 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 23 June 2007 - 11:56 AM

YEAHH! I just made some testing and installed the damaged ECU:
1. The fuel pump isn't running. I verified it with the green test connector, with my working ECU the fuel pump kicks in every few seconds, with the damaged one nothing happens!
2. I tried to start the engine - and it fired up instantly!! It just lasted for a few seconds as the pressure in the fuel line disappeared, but that's very, very good!
3. The ECU throws absolute no error codes.

So, it seems just to be a burnt circuit inside the ECU which controls the fuel pump. If anyone could give me a hint on which parts I need to have a look, it would be great, otherwise I will try to find anything burnt...

#13 joostvdw

joostvdw

    Subaru Nut

  • Members
  • 609 posts
  • Holland

Posted 23 June 2007 - 11:57 AM

the stumble you encounter sounds a lot like what my XT did at first, drivers side timing belt was off about 3 tooths :rolleyes:

so please check that first, you sound like you know enough about ea82s to do this right, but everybody makes stupid mistakes and it took me 3 months to figure it out....

#14 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 23 June 2007 - 12:21 PM

So, it seems just to be a burnt circuit inside the ECU which controls the fuel pump. If anyone could give me a hint on which parts I need to have a look, it would be great, otherwise I will try to find anything burnt...

funny we've been communicating all this time off the board and have the exact same problem, cause and situation. i could have typed that thread you just posted, identical situation.

let's keep each other informed on anything we find out. all i can say now is that i didn't see anything obvious when i uncovered the ECU, but im' not a pro. hopefully Cougar will respond with more details and help us out. oh, i see he did respond, thanks a bunch. do you remember Cougar if you had the same symptoms, everything appeared fine but the fuel pump circuit was faulty? i'm betting this same transistor will be our problem. i'll try to grab that ECU and get some pictures of the internals. i have a multimeter so i can test with some guidance. i just happen to be in the state where that vehicle is right now, so perfect timing. thanks guys!

#15 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 23 June 2007 - 01:14 PM

I just mad some pictures of the eletronic inside the ECU. Just klick on the image to get a high-res version:
Posted Image Posted Image

Cougar, please have look and perhaps you remember which part was faulty. It's a pita to test all components without a circuit diagramm...

@grossgary
Yeah, there are coincidences all the time :) So let's try to solve this problem and I guess we could help many other people with a faulty ECU, too! ;)

#16 Cougar

Cougar

    Elite Master of the Subaru

  • Members
  • 6,340 posts
  • Anchorage

Posted 23 June 2007 - 03:03 PM

Well this may not be very easy to do by posting advice but let’s give it a try. First off, I will say if you would like me to work on them myself I would be willing to see if I could get them to work for you but I can't guarantee it. If you don't have the proper tools and skills you can damage the board if you aren't careful, though it wouldn't be any real loss right now as it stands.

The transistor I replaced is similar in style and location as Q600. That is a TO92 style. I'm not sure but, Q702 may be the transistor that is tied to the fuel pump ground lead. By tracing the lead from the fuel pump lead connection it should lead you to the proper driver transistor if that isn't the correct one. The transistor I replaced was the stage before the driver stage unit.

Using a meter in the diode check position should work in finding the problem if it is a transistor. You need to be aware of the circuit it is tied to though for any possible discrepancy in the reading, as it could mislead you in thinking the transistor is bad and it really is good. It just depends on how the circuit is designed and what it is tied to. This takes experience to figure it out.

Hopefully you have the same problem I did. The replacement transistor was less than 2 dollars.

I will be gone for the afternoon today but will be back this evening to check back with you.

#17 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 23 June 2007 - 05:01 PM

I checked it, Q702 is the one which switches the ground of the fuel pump. I hope you understand my thinking and explenation, my english is somehow limited in this area...

The system is somewhat hmm strange: There is an always powered positive wire connected to the fuel pump relay which is switched by the ECU (but this circuit can't be damaged since it is not directly connected to the positive pole of the battery. If this circuit would be damaged, there must be more damages in the electronic). From the fuel-pump relay the current is routed to the fuel pump and from it to the ECU where the ground is switched by this transisor.

Posted Image

It's a bit odd that this ignition control for running the pump is solved outside the ECU when the pump itself is also controlled inside the ECU one more time, but OK, could be possible that this security circuit was implemented after the first design or that it's some kind of extra safety.

Nonetheless: The transisor is designed to beat 7V difference between Base and Emitter and I guess with switching the polarity this difference is somehow higher and so the "switch" get destroyed. I have no other explanation. I don't know how this could happen after changing the alternator, but I guess this transistor is just one weak point of the design. The circuit should simply be protected by a diode and everything would be fine. Perhaps this was designed in another way first and inserting this fuel-pump relay changed the circuits so that there is no protection any more.

OK, I have friend who worked at an electronic company (Lear Corporation) and I'm pretty sure that he can get me one piece of this transistors or some equivalent modell on monday :)

In the meantime I could try to directly ground the circuit and bypass the transistor switch. Should be a working solution for testing...

Edit: I don't really understand why JECS used this power transistor which is designed for up to 20 A and does not simply use a relay for switching the pump on and off. Especially since there is already this fuel pump relay involved. Very strange... This transistor creates a lot of heat and therefore it's connected to the metall case of the ECU so that the heat could be lead away... Here is the datasheet of the transistor.

#18 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 23 June 2007 - 07:10 PM

i don't have time to read that, i gotta go but i'll check it out later. great thinking tracking down the wiring for the right part, i didn't even think about that!

#19 Cougar

Cougar

    Elite Master of the Subaru

  • Members
  • 6,340 posts
  • Anchorage

Posted 23 June 2007 - 11:28 PM

I wouldn't change Q702 unless you prove it is bad first. It isn't the transistor I changed out to fix my ECUs.

#20 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 24 June 2007 - 06:02 AM

Ah, OK, I misunderstood you on that point! I will make more checks this afternoon.

Perhaps I will make a circuit diagramm if I find some good tool for it. But now I have to repair my WLAN-Internet connection first :)

#21 Cougar

Cougar

    Elite Master of the Subaru

  • Members
  • 6,340 posts
  • Anchorage

Posted 24 June 2007 - 12:43 PM

It looks to me that it may be Q600 is the one I replaced. By checking the transistors with a meter you will hopefully find the trouble.

#22 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 27 June 2007 - 10:41 PM

To give an update: My buddy at Lear took the wohl unit to work and is checking every transistor for damages and replacing them when necessary :) He seems to have some extra time at working and more important, he has the equipment to do the job real quick. He will start with Q702 and Q600... So, I will get some results tomorrow in the evening.

#23 stickedy

stickedy

    USMB is life!

  • Members
  • 178 posts
  • Kronach/Bavaria/Germany

Posted 29 June 2007 - 07:59 PM

Some minor update: Q600 and Q702 are both in good condition, but it seems that Q701, which is some sort of pre-amplifier of Q702 is damaged. But we couldn't confirm this yet, so stay tuned :)

#24 grossgary

grossgary

    Elite Master of the Subaru

  • Members
  • 19,807 posts
  • WV

Posted 29 June 2007 - 08:47 PM

thanks, i'm following along. i'll want to have a look at mine as well.

#25 Cougar

Cougar

    Elite Master of the Subaru

  • Members
  • 6,340 posts
  • Anchorage

Posted 29 June 2007 - 09:39 PM

Some minor update: Q600 and Q702 are both in good condition, but it seems that Q701, which is some sort of pre-amplifier of Q702 is damaged. But we couldn't confirm this yet, so stay tuned :)


This is exactly the same problem I found in the two units I worked on. The stage before the final output was bad.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users