Login Register

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
'91 ecu troubles
#1
Ok, my daughter's 1991 8v m/t sidekick wouldn't start, and then when it finally did, it ran horribly. The ecu would not post a code, not even a 12.

Installed the ecu from my 1992 asuna 8v m/t, and her car ran great, and posted a 12 code. That means all her sensors are good, but the ecu is bad/dead. Put her ecu into my asuna, and it wouldn't even start.

My diagnosis is her ecu went into failsafe/limp home mode. The caps have already been replaced in her ecu. The output transistors seem to be ok, because the engine will run, but not very well.

Does anybody know where to send her ecu for repair that has a test rig, rather than some guy who replaces the caps and calls it rebuilt.

Jerry, do you repair these units?
Reply
#2
its stuck in backup mode
is the CEL dead key on too, ? flashing 12s missing is clearly bad ECU but only if the lamp is good, and" key ON" CEL glows. ??
Does the key on CEL glow.>????
as my page says, ECU. not all ECU are repairable.

1: some acid damage can not be corrected. (for 2 reasons, damage to copper too extensive or it destroys parts not sold now.)
2: there are many chips inside that are not sold now. they are fully Unobtainum status. (pro shops cannibalize the dead to make yours work..... $$$$$ method)
3: the only full tester service I know, is Cardone at $400 R&R. but why worry all this when good ecu can be had used for $50 , put in new caps and run 20 more years.

http://en.wikipedia.org/wiki/Unobtainium

cases in point,....
if the regulator goes bad, there are no parts to buy, they are not made or sold....... if the caps short, this can and does happen.
line 2 on my page.
"This page (1st published in 2007) covers the most rudimentary repairs & failures ,tests , the most common! Advanced ECU testing, is both complex, and expensive."

line one in the main menu covers who does it right. #1
9 out of 10 ecu are easy to fix.... the others, are impossible or takes the magic of useless donor ECU to fix yours... rare parts. or becomes a DONOR itself.
keep in mind most failures are the cause of the caps. shorting. most. but hot wiring eCU's outputs gets #2 billing,,,,or shorted injectors....


see these white and black S.I.P's , they are not sold on earth..... (new) SIPS = single inline package.. or the processor..(programmed)
http://www.fixkick.com/ECU/hell/html/image_16.html

http://www.fixkick.com/ECU/5vdc-Refr-sup...IC7_8v.JPG


does your ECU have a working +5VDC refr pin? 4.75 to 5.25vdc spec ?
is IC7 bad?

http://www.fixkick.com/ECU/secrets/56B30-5VDC.jpg


i do not do repairs now, I'm fully retired now. sorry.


the good used ECU is $50 (if patient and IGNORE scalpers on fleabay , real private sales are fair.....)
a dead one is worth zero. (a true pig in poke, it is) (but buy one and chances are good it can be fixed....)
the question arises. how much labor to put in to zero value ECU, about 1 hr. max. IMO.

as the car and ECU ages. the failures can get worse. what was true in 2007 may not be true today. (
and for sure if the acid is allowed to eat it up for years, un checked.
I hope i've helped you in some small way....
http://www.fixkick.com
Reply
#3
As I understand it, besides the caps, there is a diode or two that can go bad also. Probably is mentioned in FixKick ECU page. (ECU main relay diode? Reverse voltage protection diodes? Zeners?)

You can buy pens that let you re-draw acid barfed traces, but it's tedious.
http://www.amazon.com/CircuitWriter-Prec...B00B88B9KI
Reply
#4
there are lots of parts in all ECU, any can go bad. Any !
what matters is what is common..... (in most cases)
caps are the all time #1 failure, 1989 to 1995,, never seen them fail after that.,(they changed brand makers.... is why)
the diodes only fail when folks hot wire the ECU, this is number 3 fail. IMO
number 2 fail is injector driver transistors. (open, shorted or weak or leaking....)
one other failure is that 1 of the 2 injector drivers is shorted, causing out of control injection.

the top and bottom traces are very easy to fix. the center (multi-layer PCB) traces are not.
and those pen trace makers, do not work for power bus. it's useless for this and the #1 failure is the corner cap and is power, (jumpers do work, done vast numbers)

i worked in factory for years, and repaired like 1000s of boards... seen every fail there is, (we took in field returns, too) 'I worked the super dog shelf, no one there would touch"
i've done many boards damaged by lightning (hopeless). and many other causes.. flooding (water damage) (crash damage?)

here are the hot wire fails. (nincompoops that love to hot wire do this and is common,)
1: injector drivers 1 or more. (the cure is use a scope and make sure all signals are perfect !)
2: +5vdc refr diode blown (blew the protector) and then blew up the trace, to this line.... (have photos of that in my slide show)
3: battery reversal, and load dump events blowing up those diodes.
4: the guy that loves to hot wire the fuel pump. their fingers slip 1/8" of an inch and bam, he just blew , this diode to H3LL. D102 here
http://www.fixkick.com/ECU/8v-tbi-schematic91to95.jpg

if the ECU is in Europe (guess loc) and has an Immobilizer. option, fail 4 above blows the chip for that. feature or a transistor (never touched one but its there , im told.)
5: any output the badguys hot wired, bam, blown up. (lots there are , an most chips that drive these or not made or sold now.... lots of outputs there are....

inputs almost never blow. (resistor protected (series) and cap filtered they all are)

one other failure im keen on is the corner car
it has hidden center of board traces on that cap.
the cap blows. and the acid, eats up the pad ,the repair guy attempts to fix this and ends of with the center traces dead. (it's not his fault,its acids fault....)
remove the cap , with bad pads, invites, damage to this critical feed through (VIA's there are called) and that ends the JOY.

keep and open mind to ALL parts on the PCB can fail... ALL.
it's electronics, and if you do enough boards, you learn that. fast. or never fix most of them.
assumptions are the killers to success, fixing PCB or cars..... never assume. check it first.
all great techs are skeptics (max)!

keep in mind my page is just for the easy , gedunk fails... (easy peasy fails.... only....) or easter egg fails.... look and guess...
gee why is this diode blown in half, or something that looks like BBQ hot dogs. or carbon....
or smoke soot on the lid...

no lie, careful visual inspection are #1


keep in mind, if you see smoked parts, that can mean , something else is shorted. caused that part to smoke (a victim) to blow.
we check that carefully before doing the next smoke test.
letting out the smoke again...

the lack of 12's means, the ECU is in the non documented "BACKUP mode" (book has like one line of text for this mode)
backup means, the main processor is dead , and is running under a backup chip or backup processor or some simple, hardware sequencer (super dumb logic) that gives. static spark and static fuel. or fuel only based off TPS outputs. crude and can be. but gee it runs kinda........ floods and all that..

Backup mode is very serious fail. the worst.
in the old days, the ECU would end spark and fuel. (OBD1 changed all this) circa 1987+

in a factory ,we had the bed of nails tester , interface and Major ATE test system.. (and software for board, test vectors)
we run the machine and it stops on a vector (failed step) and is live or the results of State machine failure. (complex logic pre steps)
the tech can loop on this failure.. and find the failing NODE. easy.
and then find the failing part.
lots of fun this,,,, all this...
testing on a bench is basically pure luck,,,, using a meter and scope. and some skill.... but mostly just luck or "common fail points checks"
http://www.fixkick.com
Reply
#5
(06-05-2015, 11:05 PM)fixkick Wrote: its stuck in backup mode
is the CEL dead key on too, ? flashing 12s missing is clearly bad ECU but only if the lamp is good, and" key ON" CEL glows. ??
Does the key on CEL glow.>????
as my page says, ECU. not all ECU are repairable.

Yes, the cel light glows at key on.

1: some acid damage can not be corrected. (for 2 reasons, damage to copper too extensive or it destroys parts not sold now.)

Yes, we had to repair a circuit trace that had been eaten away by the leaking cap

2: there are many chips inside that are not sold now. they are fully Unobtainum status. (pro shops cannibalize the dead to make yours work..... $$$$$ method)
3: the only full tester service I know, is Cardone at $400 R&R. but why worry all this when good ecu can be had used for $50 , put in new caps and run 20 more years.

As a test, I used the ecu from my 1992 Asuna as a substitute for her 1991 ecu, and the car ran great, and pulled a 12 code.

http://en.wikipedia.org/wiki/Unobtainium

cases in point,....
if the regulator goes bad, there are no parts to buy, they are not made or sold....... if the caps short, this can and does happen.
line 2 on my page.
"This page (1st published in 2007) covers the most rudimentary repairs & failures ,tests , the most common! Advanced ECU testing, is both complex, and expensive."

line one in the main menu covers who does it right. #1
9 out of 10 ecu are easy to fix.... the others, are impossible or takes the magic of useless donor ECU to fix yours... rare parts. or becomes a DONOR itself.
keep in mind most failures are the cause of the caps. shorting. most. but hot wiring eCU's outputs gets #2 billing,,,,or shorted injectors....


see these white and black S.I.P's , they are not sold on earth..... (new) SIPS = single inline package.. or the processor..(programmed)
http://www.fixkick.com/ECU/hell/html/image_16.html

http://www.fixkick.com/ECU/5vdc-Refr-sup...IC7_8v.JPG


does your ECU have a working +5VDC refr pin? 4.75 to 5.25vdc spec ?
is IC7 bad?

Ok, I'll check that out when I get her ecu.

http://www.fixkick.com/ECU/secrets/56B30-5VDC.jpg


i do not do repairs now, I'm fully retired now. sorry.

Ah, congrats on retiring. "Live long and prosper"


the good used ECU is $50 (if patient and IGNORE scalpers on fleabay , real private sales are fair.....)
a dead one is worth zero. (a true pig in poke, it is) (but buy one and chances are good it can be fixed....)
the question arises. how much labor to put in to zero value ECU, about 1 hr. max. IMO.

Ok, the hunt is on for an ebay/amazon ecu.

as the car and ECU ages. the failures can get worse. what was true in 2007 may not be true today. (
and for sure if the acid is allowed to eat it up for years, un checked.
I hope i've helped you in some small way....

Yes, you have. I appreciate your honest posts; no guessing, just facts. Lots of other guys just love to bs people.
Reply
#6
thanks, i only try to offer things possible .
for example i even trace out the Ecu CIRCUITS TO FIND which transistor, vsv driver for EGR, id not expect others to try that it is tedious to max.

yes, ok the cel glows. it its not in backup. but will not flash, that means the ECU code is not running right, its lost or is crashing.
if the regulator , is dropping out, or going too low and , the processor can just keep rebooting and never get the chance to do the FLASHING LED, code... it's lost...
the acid can do harm not seen, easy, .. it can soak deep, or under chips or other places...

always have a spare ecu, even a wrong one is better than a dead one. (if it gets 12mpg its dead in my book)
for example , any 56bxx ecu will run well. a/t , m/t or even 91 to 95 years here,.
ive run the a/.t on my m/t car no issues.
if car is 3sp, you lose, TC lockup with m/t ecu. but car runs in 3 gears ok. and good mileage....
http://www.fixkick.com
Reply
#7
(06-06-2015, 11:17 PM)fixkick Wrote: there are lots of parts in all ECU, any can go bad. Any !
what matters is what is common..... (in most cases)
caps are the all time #1 failure, 1989 to 1995,, never seen them fail after that.,(they changed brand makers.... is why)
the diodes only fail when folks hot wire the ECU, this is number 3 fail. IMO
number 2 fail is injector driver transistors. (open, shorted or weak or leaking....)
one other failure is that 1 of the 2 injector drivers is shorted, causing out of control injection.

Great info! Smile Didn't you know you were a tech.

Got any old organ parts laying around (TDA1008's)? Smile

My ECU saga way back:

My ECU blew up back in 2004, total dead car. After the incompetent dealer advised me I needed a fuel pump, I picked up one for $90, peeving them of the $300 they wanted to charge. Then they said oh it's the ECU, $1200-2000 new, or $650 rebuilt. I bought one over the internet, $200 overnight I think, peeved them again, but there went my car to running again. Found their wrench on my hood in the gap once home, should have kept it, but returned that when I went to complain about my belt slipping. I had had them change the belts while there
Reply
#8
yes, many shops do not not know how to fix 1990s cars, lost books, lost expiences, missing scan toll tech1,,, or blown up long ago.
if id didnt check fuel pressure, (all 7 steps) then how can you say a pump is bad, Sure, the rail is dry with 12vdc at the pump 2 pins........ only....

yes with quality caps in the ECU, it can go for 40 years. i have brands i trust. (most are rated for 20 years,) but top brands beat that. all top brands do.... its their thing !
thanks for sharing real horror,... love to here how you got this worked out.... all those details are interesting.

the top brand and higher temperature rated caps are like a buck. no need to use China clones or the one that explode, due to stolen, cap acid, electrolyte formula's
http://www.fixkick.com/ECU/ECU.html#Nichicon

the ECU can do 3 things (in the horror department)
1: DEAD CEL KEY ON, OMG. (lamp good. and fuses all good)
2: REfuse to flash 12s, or any code at all.
3: starts to flash a code and halts mid stride , huh? really? (this is crashing the processor) all 3 are horrors. and x10 that if caps are fresh.... pure bad news.

most are a bad regulator, but there are lots of other parts and ways to fail.

get car fixed by any ASE shop, avoid stealerships... (unless rich)
http://www.fixkick.com
Reply
#9
Hi! I'm Noel from the Philippines. I currently have a problem with my Hyundai Elantra 1996 A/T. It suddenly ran rough as if only 2 cylinders are firing. It continued to start but roughly and then died completely. I checked for voltages at the harness side and ecu plug and all are normal. I used the screw driver method on all the plugs and confirmed no spark. We changed the coil and all sensors but no go. Went as far as checking continuity of the wiring harness. Engine timing is good. The ecu shows no sign of burned or overheated components. Very clean. It's a dilemma not to be able to speak to a true technician like you. Most only use the substitution method. I read schematics and do point to point checking. Hoping you can give me some pointers. I know this is a Suzuki forum but I found the system similar as most efi systems are.
Reply
#10
redux.
I used the screw driver method on all the plugs and confirmed no spark. (never do that, use real test spark plugs , or overload the coil and ECU drivers)

seems this is wrong?>? are both coils dead, or just one???
this post , fooled me,,, i cant see what is wrong here, both coil packs are dead. no spark x4, some weak,or?
the coils are independent on the primary side.


COIL PACK:
ground (this pin makes the internal capacitor work)
12v
fire 1&2
fire 2&4

BUT, one coil is shorted, this always burns up the ECU. Ignitors back in 1996

REDUX>


[Image: 3_03_10_15_1_49_19.jpeg]
http://www.fixkick.com
Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)