[Rhinoman]

I'm surprised that they couldn't repair that, maybe its more a case that they didn't want to spend the time on it or there was more damage than was evident.

 

[m000035]

While I do not know anything about this particular repair shop. I wonder if they have ever repaired anyone's ECU. $50 sounds like fairly good pay to open/close and send back an ECU. Sending back every ECU as "non-repairable" could make a good amount of money.... Scam?

Like I said, I have no information about THIS shop, but my mind works in odd ways.

 

[Rhinoman]

There is a company in the UK who are quite well known who charge the equivalent of $190 whether they fix it or not. Personally I don't charge for testing but sometimes spend hours on a 'no fault found', I'm tempted to bring in a charge but I don't haven t make a living out of it and the majority of ECUs are faulty.
I do wonder if they have ever repaired a 1.8 Sidekick ECU, when I get in a 'new' ECU I have to make up a new harness to connect it to my tester and programme the tester with the correct cam and crank signals, that often takes longer than the repair and there is a cost involved, I can see how some could be tempted just to return it as unfixable.

 

[magician35]

Just as an aside, it is good to 'study' before jumping into this - and with the internet, there is now no excuse not to do so. I brought my Tracker here from NY, when I moved here 18 years ago. When I had been here about 3 years or so, and before I found this forum and started learning the specifics about this car, my car wouldn't start - diagnostics indicated that it was the computer. In my desperation and haste, I found some guy in England who advised he could repair the computer - I was happy enough, as I was afraid that I would have had to trash my car. The FSM, of course, does not indicate anything about 'repair', only 'replace', which I was pretty ignorant about. Anyway, I sent the ECU over to England, and about a week later it was returned in working order - to the tune of about $300 (somewhat expensive for the replacement of $5 worth of capacitors). About a year later, I began to have the same problem - I pulled the ECU, and saw burn marks on the circuit board. This time, I found a replacement ECU (from California) for $100 core exchange. After waiting 2 weeks for the replacement, I plugged it in, the car started and then stalled again. As these replacement cores advise that the 'warranty' is voided if you open the unit, I never opened it, but sent the unit back for another replacement. And waited another 2 weeks. The same thing happened again. This time I opened the unit, and saw the burn marks in the same place - and that is when I found this forum. With due respect, Jerry diagnosed the problem for the forum - it wasn't the caps this time, but the injector driver. With his info, I found replacements, ran new injector wiring (rather than searching for the culprit that was continuing to cause the problem), had the new drivers soldered in at my local computer shop, and also installed a 5v fuse on the line, so that, if a wiring problem occurred again, the fuse, rather than the injector driver, would blow. So, a real learning experience - and my car is still on the road, thanks to the myriad of people on this forum who take the time to share their experience and information.
Hopefully your 'adventure' is over, and it has been a learning experience for you, as well.....

Even though I started repairing vehicles 65 years ago, this was my first encounter with an ECU. I was amazed to find (as you did) that there was no fuse protection on the board. Even though I thoroughly checked the run from the ECU to the IAC, I was very hesitant to hook it all up and connect the battery. It was quite a relief to discover that it didn't fry again (at least for now.) I understand that newer cars no longer have all of the functions on a central card and employ local modules for individual functions. (Could the designers be that smart?)

If I run into something like this in the future, the first thing I'll do is pretty much what I did here: Trace the damaged area of the board back to the connector pins so I know what control is involved. Then I'll know where to look for shorts or voltage that could have done the damage and correct that before I put a different board in.

Thanks again for the help.
Malcolm

 

[Rhinoman]

I understand that newer cars no longer have all of the functions on a central card and employ local modules for individual functions. (Could the designers be that smart?)
Malcolm

Not really, currently the trend is to incorporate as much as possible in one ECU to reduce costs, it makes them difficult to design because of the number of pins and wires required, a complex ECU can have 120 pins or more.
Modern ECUs are required to withstand short to battery and short to ground on all pins except power pins so they are more robust.
The design concept has little to do with us designers, ECUs are designed to meet the manufacturers requirements.

 

[colonelhogan44]

Nice! It's always a good day when problems get resolved. If I were you, I'd me breaking out the soldering iron and replacing components to try to make it work!

But I'm a tinkerer.

 

[magician35]

Not really, currently the trend is to incorporate as much as possible in one ECU to reduce costs, it makes them difficult to design because of the number of pins and wires required, a complex ECU can have 120 pins or more.
Modern ECUs are required to withstand short to battery and short to ground on all pins except power pins so they are more robust.
The design concept has little to do with us designers, ECUs are designed to meet the manufacturers requirements.

No offense meant. I was in Aerospace Engineering and it was the specs generated by Systems Engineers who didn't think at the component level that created frustrating problems.

 

[magician35]

Nice! It's always a good day when problems get resolved. If I were you, I'd me breaking out the soldering iron and replacing components to try to make it work!

But I'm a tinkerer.

I might take a closer look at it when I have the time. I think the biggest problem is that some traces were taken out so I'd have to do some patch wiring. I'd also not be too anxious to have it take out the IAC if it wasn't right.

 

[Bex]

I happened to watch Top Gear last night, and they were driving across Australia in a nice Bentley coupe. They said that the car has THIRTY computers in it. What a horror......

 

[m000035]

The twin turbo Audi had almost that many, and it was built in 1999. the computers "talk" to each other over the CAN Bus. One computer for each front seat (memory seats), one for the back up warning system, even one for the radio antenna.

I still have the interface to connect a laptop to the system.

 

[Rhinoman]

I happened to watch Top Gear last night, and they were driving across Australia in a nice Bentley coupe. They said that the car has THIRTY computers in it. What a horror......

I would think that its a lot more than that, maybe 100+, there aren't many functions that aren't computer controlled these days.

 

[magician35]

I would think that its a lot more than that, maybe 100+, there aren't many functions that aren't computer controlled these days.

I don't know if I should continue this thread or start a new one but the saga continues. The "new" ECU took care of the idle problem but a smog test today failed terribly on all fronts. Everything is way out of limits but it runs good and there are no codes and no Check Engine light. Apparently the ECU had been reset. The smog guy says it might help to just drive it for awhile to let the ECU "reset". My OBD2 says "not ready for smog testing" which means there are tests that need to be completed. Don't know if this is an indication that the ECU is malfunctioning or not and don't know where to go from here. I'm tempted to put the old ECU back in just for the smog test. It does not show any codes even with the IAC disconnected. The IAC is no real problem as long as the A/C isn't turned on.

 

[aquanaut20]

Ant time the ECU does not detect main power it resets to factory, all ready monitors (emissions) return to "NOT READY"

The vehicle must run through specific run cycles in order for the ECU to set the monitors (learn process).

Suggest you run it for a couple of days, stop and go, highway, coast from speed, hot/cold cycles..

Any OBD scanner can detect the ready monitors...

... Philip

 

[magician35]

After 3 days and about 60 miles of driving (streets, freeway, hot, cold, up, down) the OBD2 readings haven't changed. Reading Monitors gives - Misfire, Fuel System, Components: continuous, available, complete; Catalyst: not continuous, available, not complete; Heated Catalyst, Evap System, Secondary Air, AC Refrigerant: not continuous, not available, not complete; Oxy Sensor: not continuous, available, not complete; Oxy Sensor Heater, EGR SYstem: not continuous, available, complete.

In the list of Oxygen Sensors it shows Bank 1, Sensors 1 and 2 available. For both it shows $01 - $06 incomplete. And it still shows "not ready for smog testing".

I don't know how to interpret those results, whether they indicate sensor problems or problems in the ECU. (Still no fault codes).

Do you see any reason not to put the old ECU with the bad IAC circuit back in (with the IAC disconnected) to see if it will read differently?

 

[aquanaut20]

The ECU will not complete the ready monitors if you have any hard/pending fault codes...

Do you have a miss fire?

Do you have a Live Data Scanner, if so what are the fuel trims, and does the O2 switch rapidly with no bias?

What OBD scanner/tester are you using??

... Philip

 

[Rhinoman]

If you don't have an FSM for the 1.8 then I would suggest checking the 1996 1.6 FSM, available on Acksfaq. The FSM details the conditions required for the tests to be completed.

 

[magician35]

The ECU will not complete the ready monitors if you have any hard/pending fault codes...

Do you have a miss fire?

Do you have a Live Data Scanner, if so what are the fuel trims, and does the O2 switch rapidly with no bias?

What OBD scanner/tester are you using??

... Philip

I have the OBDWiz (on my laptop). I'm just trying to learn how to use it. There are no faults or pending faults showing. I don't believe it is misfiring although the idle does roughen up a bit with the A/C on even though the idle speed is OK. The Misfire Monitor test shows complete.

I know I can record live data so I'll go back and try to find the fuel trims and O2 data. Thanks. Malcolm

 

[magician35]

The ECU will not complete the ready monitors if you have any hard/pending fault codes...

Do you have a miss fire?

Do you have a Live Data Scanner, if so what are the fuel trims, and does the O2 switch rapidly with no bias?

What OBD scanner/tester are you using??

... Philip

Here's what I found:

Short Term Fuel Trim Bank 1: 0 at idle, spikes to 19 with RPM , then levels back to 0 with constant RPM, Idle 0

Long Term Fuel Trim Bank 1: Idle -3.1, RPM -2.3, idle -3.1

Short Term Fuel Trim Bank 1, Sensor 1: Idle 0, RPM 19 (constant), 0 at idle

Short Term Fuel Trim Bank 1, Sensor 2: Idle around -1, no response to RPM

O2 Sensor Bank 1, Sensor 1: Idle .2, RPM peaks at .5, levels at .2, idle .2

O2 Sensor Bank 1, Sensor2: Idle 0, No response to RPM.

Problem with Sensor 2? (No codes showing or pending.)

Malcolm

 

[magician35]

If you don't have an FSM for the 1.8 then I would suggest checking the 1996 1.6 FSM, available on Acksfaq. The FSM details the conditions required for the tests to be completed.

I have the '96 1.8 FMS with the '97 update but don't remember seeing that information. I'll take another look.

Malcolm

 

[aquanaut20]

Defiantly an issue with the O2 sensor circuit..

For the upstream(B1S1 and B2S1), the graphs should have a fairly rapid fluctuation (switch) from .2 to .8vdc, and fuel trim at idle and 2000rpm should be close to "0",+/-2, when hot and running properly...

If you have one with non action, it is either dead, or the wiring is faulty, This however should set a MIL code..

The alternative is that the ECU is sick....

... Philip