Mechanical Problems & Technical Chat If you've got a problem you just can't figure out, a noise you can't diagnose, or a check engine light that won't go away, ask about it here!

ECU Burned Out?

Thread Tools
 
Search this Thread
 
  #11  
Old 07-07-2010, 04:42 PM
RonJ's Avatar
Recognized HCF Member
Join Date: Sep 2007
Location: Houston, Texas
Posts: 9,453
Default

Originally Posted by Glenn22
I have 12.7v on both sides of 10a fuse #9 via test tabs
The fuse is good and gets power. Scratch it off the list and now move to the troubleshooting in the diagram.
 
  #12  
Old 07-07-2010, 05:13 PM
Glenn22's Avatar
Junior Member
Thread Starter
Join Date: Jul 2010
Posts: 20
Default

I performed the DLC circuit test and passed every thing but i am hesitant to put in a new ecu since it might blow it. also when i was removing connector e from the ecu (step 11) i noticed that the ecu felt pretty hot.

Also is there any way to test the MCU without the "special tool"? and would the ELD be a contributor to this problem?

I appreciate all the feedback and great info Im receiving thank you guys
 

Last edited by Glenn22; 07-07-2010 at 05:15 PM.
  #13  
Old 07-08-2010, 12:05 AM
civicalized's Avatar
HCF Member
Join Date: Jun 2010
Location: Vancouver Island
Posts: 41
Default ecu

Hi, the hot ecu makes me a tad nervous... However I believe that in more recent manuals honda has eliminated or toned down ending the decision tree with "try a new ECU" because it is usually not the ecu. The basic trouble is an ecu is made to determine specific faults and it is not always right. I'd suspect the ecu only when all other possibilities are exhausted. I had an acura legend once, a honda mechanic and I agreed that ecu substitution was what the manual was suggesting for my car, actual trouble was the distributor, even though the ecu code was was pointing at the cam sensor. The fact that a local wrecker had a $50 return fee for used ECUs made me think twice as well, obviously he was aware that ECU failure was not too common. BTW, that distributor pickup coil had an intermittent open, argh.

And remember, the ECU was not programed to detect mis-wiring or intermittent connections. The same acura came to me with the temp and throttle position wires crossed, the ECU was no help with that....

Anyway, bottom line? Take our free advice with a grain of salt, but only as an adjunct to the troubleshooting guides in the honda manuals, which in your case is both a 2002 and a 2003. The honda books are pretty good, go through the steps, sleep on it and look for unanswered questions. "ie, what about that orange and black wire?" Next day check out that hunch and repeat ad nauseum. The amateur repairman has a big advantage, time and lots of it. Sooner or later the light will go on and you will at last say "eureka", sadly no one else will be as impressed! This moment always comes about 12 hours after the big fire starts to become an attractive option. Fight the urge!

Also, when testing for voltage, really one should test for power with a test lamp. A turn signal bulb in a socket with long wires and alligator clips. Clip one side to ground, insert a pin in the other clip and probe at will. Refrain from piercing the insulation.

In some situations both sides of an open fuse will show voltage, due to a backfeed. That backfeed however will offer too much resistance (we hope) to allow enough current through to light a lamp. That said be cautious about drawing power through the ecu with a test lamp unless instructed to do so.

So, all that said, if the trouble looks like it involves the ICU (MCU?), open it up and check the solder joints. I suppose the ECU could have bad solder joints too, could be worth a peek. Look at large transistors and harness connections. Be extra cautious and disconnect the battery unless you need it in circuit for test purposes.

If you have links to the manuals post 'em and I'll be happy to offer possibly worthless opinions on schematics.
 

Last edited by civicalized; 07-08-2010 at 12:07 AM.
  #14  
Old 07-08-2010, 04:33 AM
Glenn22's Avatar
Junior Member
Thread Starter
Join Date: Jul 2010
Posts: 20
Default

Well i should be due for figuring out this problem any day now since ive been wanting to either set it a blaze or push it off a cliff.

this is basically the only manual on wiring i have: http://www.clubcivicquebec.com/HowTo...iagramsEM2.pdf

The ecu being as hot as it is makes me think that there has to be a problem in the ICU (MCU), ELD. i have another ecu which i picked up for $100 with 100 day warranty but i dont want to put it in with the possibility of it failing and not only losing the ecu but also having to pay for my keys to be programed twice.

On a side note why doesnt honda just program their keys to the pcm/imobilizer rather than the ecu as well? i see no logic fin the thinking of their current setup.
 
  #15  
Old 07-08-2010, 05:01 AM
mk378's Avatar
Recognized HCF Member
Join Date: Jun 2007
Posts: 10,023
Default

The only thing that seems to be not working is the OBDII test plug, right? In that case check the continuity of all the test plug wires end to end from the plug to the ECU. Also check them for shorts to ground, each other, or other wiring.

Edit: This is on page 12 of the diagram you posted. Note that the brown and light blue data wires are also bussed to the ABS control unit. It is possible that a problem with that unit is interfering with the ability to read engine codes. So unplug the ABS unit and try the OBDII scanner again.
 

Last edited by mk378; 07-08-2010 at 05:08 AM.
  #16  
Old 07-08-2010, 06:33 AM
Glenn22's Avatar
Junior Member
Thread Starter
Join Date: Jul 2010
Posts: 20
Default

The circut tested out fine but ill go back and check for continuity on the wires from the connector to the ecu. It looks like e3 to dlc5 e23 to dlc7 and e28 to dlc9 all these showed correct voltage during the test but illl check for continuity and that they are in the correct pinouts. the rest of the DLC wires go to the multiplex is there acess where i would be able to check continuity on DLC ports 14,16. 14 goes thru the multiplex and connects to both the immobilizer as well as the srs unit and 16 connects to the immobilizer. Would a fault in the wires or in the path to the immobilizer cause the shifting problems and the cars inability to rev high and excell past 5mph in first? or does the immobilizer unit just impact cranking the engine?
 
  #17  
Old 07-08-2010, 06:51 AM
RonJ's Avatar
Recognized HCF Member
Join Date: Sep 2007
Location: Houston, Texas
Posts: 9,453
Default

Please give more details about the results of tests detailed in the diagram. Did you complete all tests in the diagram? Was the answer YES at every YES/NO juncture? Does your current ECU have the updated software?
 
  #18  
Old 07-08-2010, 07:09 AM
Glenn22's Avatar
Junior Member
Thread Starter
Join Date: Jul 2010
Posts: 20
Default

tests 1-5 had all yes answers so i preceded to to test 10 as directed from there i concluded the test with the final test being also a yes. i am not sure if the ecu has updated software or not since it was from a salvage yard. it is a PLR-L54 from what i remember 54 represents the software update. i would put in my other ecu which is a PLR-L58 which should have newer software but as i stated earlier im hesitant due to the current ecu getting very hot and i do not want to damage my other ecu.
 
  #19  
Old 07-08-2010, 07:11 AM
mk378's Avatar
Recognized HCF Member
Join Date: Jun 2007
Posts: 10,023
Default

I think you only need DLC pins 4 and 5 (ground), 7 (K line data) and 16 (battery power) for a basic OBDII scan. The wire that goes from 14 to the SRS and immobilizer is for some sort of Honda proprietary thing. The only other thing the pin 7 K line goes to appears to be the ABS.
 
  #20  
Old 07-08-2010, 07:17 AM
RonJ's Avatar
Recognized HCF Member
Join Date: Sep 2007
Location: Houston, Texas
Posts: 9,453
Default

You may want to try another junk yard ECU.
 


Quick Reply: ECU Burned Out?



All times are GMT -8. The time now is 01:35 AM.