ECU Failing at Random

[weirdchris555]

Hello everyone. My name is Chris. I live in upstate New York and am from Boston Mass just for some basic info about me. I have just been going back and forth with Matt via email, and decided to ask the question on the forums for more information.

Well a couple of months ago, I bought Nistune from a 3rd party and they installed it in my SR20DET redtop #62 ecu.Well, I got it back and it just didn't work right. It would work fine, then it would die by kicking into the processor fail mode. This occurred at random times both at idle and when driving. Ran pig rich and sometimes just fouled out. There did not seem to be any repeatable procedure that would cause it to happen. The only thing that I noticed was if the car was warmed up it would happen sooner than if the car was sitting overnight or with the battery unplugged for a while. I thought maybe some sensors were failing, even though everything was working fine prior, so I traced everything and it was all good.

At first, I found out the base image on Nistune was for a blacktop S13 ecu. But I knew that people would run the blacktop ecu even with the wrong O2 sensor. So I removed the board and tried to reseat the board in the pins. That did not help. So I personally re soldered the jumper back to the factory position and the same issue occurred no matter if it read off of Nistune or the stock maps. I do not have any images of the install in this ecu, but it looked good. I will post them in the near future when I am back near that ecu.

So I purchased a "virgin" ecu from an online forum. Well, it was far from factory. Someone had tried to put a daughterboard at one point or another in there. The pins were bent, some were missing, hot glue just kind of thrown about, a random jumper on the front side of the board (not where CJ1/CJ2 are), and what looked like burn marks or severe corrosion marks, as seen in the pictures attached. I feel its kind of one of those things that you are afraid to touch it because it is working fine the way it is but any slight alteration could send it to its doom. I do have pictures of this though, as seen below. Somehow though, it actually does work. I am surprised. Matt told me to check the througholes on this ecu to see if it is repairable. The images below are of the working 2nd ecu explained in this paragraph.





Anyways, I am trying to figure out why my original ecu is kicking into processor failure after a random number of minutes. Anywhere between 5 to 11ish if cold. Matt told me to check pins 1 and 40? To see if there is 5V going through it? Possibly the original install shorted the power regulator or something is shorting and overheating and once it cools down its fine? Below is a video of it kicking into the failsafe mode right near the end.Please I realize my mistakes and I am bummed so I don't need people busting on me about my purchases thank you.
Any help is greatly appreciated since I really need to get Nistune running. Worst case scenario I need a third ecu, this time from a trusted person.

http://www.youtube.com/watch?v=sRPjF6bhD1A

[skyline_stu]

Ouch. That board is going to need some surgery. Either send it to Australia and I can fix it, or go find a stock one on ebay and Nistune that..

Stu

[weirdchris555]

So here's the weird thing. Today I decided to clean up the jumper on my original board. I then resoldered it (which I have done in the past that yielded the same failure) and threw it in the car. Drove it around for about half an hour and so far hasn't tried to kick into limp mode. Leaving it in the car for the rest of the week to see if it fails on the way to work and then if it stays, I'll try switching the jumpers back to read from the Nistune board.

[Matt]

That isnt the ECU pictured above is it?

On S13 ECUs header going out to the board can be damaged and it wont affect the ECU in 'factory' mode

All it takes one line to the header to have the track cut, cracked or throughole damaged and the ECU wont be able to run with the board

[weirdchris555]

No its not the ECU pictured above. This is the original ECU that had Nistune installed the first time of which I do not have any pictures uploaded currently. A couple of months ago it was randomly failing both when CJ1 was soldered and then when I soldered CJ2. Both randomly dumping into fail mode at any time. Now for some reason it hasn't yet, so I am going to be driving with this ECU installed for the rest of the week and then if that stays running I'll switch the jumper to CJ2 and try Nistune.

I will not ever try to install Nistune in the above ECU until it has been tidied up, whether that means sending it to stu or not.

[skyline_stu]

All it takes one line to the header to have the track cut, cracked or throughole damaged and the ECU wont be able to run with the board

There's atleast 1 line busted on there.. prob more when that EVIL header is very, very, very carefully removed.


Certainly won't be the worst ecu i've fixed. Think Nippon Denso, Electrolytic capacitors, leakage, pcb combustion.. Happens frequently on older cars that have had recent battery or alternator problems..

Even had to repair my OWN ecu the other week, the RT board in my holden came loose and broke a pin off... So on the side of the road here's a guy heating up a 1lb plumbers iron with a propane torch in the car park of macca's.. I certainly got some strange looks from passers by.. and I realized afterwards I had a power inverter and old soldering iron in my toolbox... DOH .
But hell, it got me home to better equipment.


Is that third picture one of those wierd 3d images ?? Makes.my.brain.hurt.
Stu

[weirdchris555]

Sorry it was a terrible terrible picture. Was trying to get a picture of what I thought was a jumper wire but I "jumped" to conclusions and its just a component of the board. Well, at least on both of my ECU's they are there. I will keep this thread updated. Come this weekend I will try out the Nistune board if all is well. I still don't like having the board with a blacktop image on it installed in my redtop ECU.

[skyline_stu]

Naa, don't stress. the red top ecu runs a lot leaner mixture and different timing, (more advanced), the black top is softer in timing and richer in fuel. The red top just revs 100 rpm more.

[weirdchris555]

So a sooner update than I expected. Went to go get lunch quickly from work and I was driving through the town when all of a sudden the car bucked bad for a split second and then resumed its normal pace. Now this is what I had previously experienced the exact second that the ECU would kick to failure mode, except this time it didn’t stay in it. Just for the split second. I am wondering if from the beginning there was bad solder at the CJ1/2 joint and when I cleaned it all up it helped.

Now I think it just kicked for a split second today because I basically just blobbed down some solder and made a cold joint since I honestly didn’t think it was going to last this long. Maybe tonight I’ll clean it up and try out Nistune I guess.

On a second drive, I came to a stop and it kicked into the limp mode. I turned the car off then turned it on and it was fine again for about a minute. Then it kept going into and out of limp mode until I turned the car off since I was at my destination. Switched back to my 2nd ecu and got back home.

[weirdchris555]

So I've been driving it around for the rest of this week. Currently in the middle of replacing the oil pump. Besides the incident where it was going in and out of the failure mode, there was only one other time that it failed when I was sitting at my friends house idling after driving half an hour to his house. This is after I cleaned the ECU up and instead of using a blob of solder, I properly soldered in a single core copper wire connector. I am trying my best to diagnose why it is kicking, but it is definitely doing it a lot less frequently.

This is my original ECU that I am currently using with Nistune. Did not grab a pic of the other side at the time though. This is not the messed up one pictured above.

[skyline_stu]

What does your tacho do when faulting?? It almost sounds like another fault elsewhere in the electrical system. When faulting, does it only rev to 2000rpm and sound like a rev limiter ?

[weirdchris555]

When it faults, the tach does not do anything funny. Maybe it runs at a little lower speed since its trying its best to ignite the fuel. It runs in the 9.XX A/F. A couple of months ago when I first got all my Nistune stuff when it would kick into the mode I could rev it wherever it wanted to go. There was no sensor failure rev limiter. I could take it all the way to the actual rev limiter if I wanted to. Now though since it hasn't stayed in the failure mode I couldn't tell you with any certainty but I have taken it above 2,000. It will just enter or fall on its face while driving and then immediately go back to normal operation mode. There was only one time where it stayed in the mode so I turned the car off then back on and it would do as described above, enter and then exit failure mode over and over again.

I did happen to take a couple of logs a couple of months ago when this first happened that I will try to track down. I will post it up here if I can find them. I did notice that right before kicking into the failure mode, one or more sensors seemed to display faulty information, but it was random as to which sensor. So naturally I traced every wire back to the ecu and they all checked out to be fine (even though the car was running fine prior to all this). That's when I expected an ecu problem and got a new one.

[skyline_stu]

Is your ecu connector plugged in tight ??

[weirdchris555]

Yes it is. Just as tight as when I put my other ecu in. Sorry I haven't gotten on here sooner, I was replacing the oil pump.

[weirdchris555]

Sigh. I gave up. Purchased a 3rd ECU. Sending it off to Autovaughn Performance here in the states for the install. Do not really know what I am going to do with the other 2 ECU's.

Now I just wish there was a way for launch control. Drag strip is right next door and I miss running it there.