The Lotus Cars Community banner

1 - 20 of 23 Posts

·
Premium Member
Joined
·
64 Posts
Discussion Starter #1 (Edited)
My '05 Elise is blowing up ECU's, FIXED!

Guys,

This is a continuation of my car wont start woes. Here is the short version of the story:

Car ran fine after a clutch change. 4 or 5 trips to work etc.
One day I start the car, it runs for 1 sec. then died. No CEL, gas gauge shows no bars, fuel light on, fuel pump does not run, start button does nothing. All fuses are tested and good. ECU wont talk to the OCBII reader.
Disabled immobilizer, replaced Multi Function Relay. no change.
Sent it to the shop on a flatbed.
Shop found damaged circuit on inside the ECU. Plugged in another ECU. Fuel pump ran for a few seconds, fuel bars came back then...
POW!
Dead ECU #2
I have purchased a 3rd ECU but the shop is wanting to find the trouble before plugging it in (good idea).

Anyone have any idea what this might be?
BrianM

It was a short in the post cat O2 sensor. see last post.
 

·
Registered
Joined
·
733 Posts
can you post pictures of the circuit board damage?

I've always wondered if the t4 ecu was as delicate as the preceding k4 ecu- that one burns up a (cam position?) sensor microchip if the battery leads get reversed.

will you be probing the ecu-removed wiring harness via multimeter to look for battery shorts? or will the shop be doing the diagnosis?
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #3
can you post pictures of the circuit board damage?

will you be probing the ecu-removed wiring harness via multimeter to look for battery shorts? or will the shop be doing the diagnosis?
Shop is doing a harness check. I will ask for a photo.
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #4
Pics from shop. I already asked if they know which components were involved.



 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #5
OK, looks like the damage is around Pins 42-43 of the processor. Those look to be voltage reference pins V_RH, V_RL???
Anybody have a schematic of the ECU?
 

·
Registered
Joined
·
733 Posts
here is freescale's user manual for the mc68376 processor:

https://www.nxp.com/docs/en/data-sheet/MC683336_376UM_ZIP.zip

in it, look for document apbmech.pdf to see the pin labels. that just repeats the labels and pin numbers that you just wrote.

look for document c8qadc.pdf to read about the queued analog to digital converter module. the block of pins near 42/43 are analog inputs, so i guessed that they were related. sure enough, V_RH and V_RL are the high and low analog reference pins.

finally, look to table A-11 in document apaelec.pdf to read the electrical limits for those pins. V_RH - V_RL max is -6.5 to 6.5 V. same for the difference between those pins and the analog power and ground pins (V_SSA and V_DDA).

i don't know anything about these reference voltage pin explicitly, but i would guess that they connect to a dedicated voltage regulator microchip. so the external connection that leads to a fault may be removed from those pins.

if that is the case, i would also guess that improper voltage application to one of the chips lead to the higher current flow and pcb damage. this is all speculation on my part, but i would be looking for a short to battery voltage in the wiring harness as a first guess.

edit: in case it's not those pins directly, but a neighboring pin, would you like a map of analog pin to sensor type? i can add that if you think it's helpful.
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #7
edit: in case it's not those pins directly, but a neighboring pin, would you like a map of analog pin to sensor type? i can add that if you think it's helpful.
Yes Please. A second ECU failed in the same place so it must be a harness short of some kind. a map of sensor pins could help narrow it down.

BrianM
 

·
Registered
Joined
·
733 Posts
The attached image shows processor pin number, Sensor function, C121E0002H calibration ROM memory locations (16 bits, one each in the AtoD RAM buffer (0xFFF2xx) and one in static RAM (0x80xxx) where it is accessed by ECU code) and ECU wiring harness pin number. The ECU harness number is printed in the ECU connector next to the pin hole at the ends of rows of pins, I believe.

There could be errors here too.

For completion in this thread, I included an image of the processor pinout with a zoom to the analog input section.

AN51 and AN52 are the pins nearest to the V_RL and V_RH pins, so perhaps the oxygen sensor signal has a wiring short to its heater power?

Also, your images are not loading for me, if you would like me to look them over closely, could you please download them and save as attachements to Lotustalk?
 

Attachments

·
Premium Member
Joined
·
64 Posts
Discussion Starter #9
AN51 and AN52 are the pins nearest to the V_RL and V_RH pins, so perhaps the oxygen sensor signal has a wiring short to its heater power?
Oh crap, the O2 sensor. I rewired the after cat sensor with a Toyota one. I wonder if my solder joints gave up?
 

·
Registered
Joined
·
733 Posts
Okay, now I'm more confused: I don't think that the neighboring analog input pins are involved anymore. But I agree with your assessment that the V_RL lead was involved. Unfortunately, it looks like the other end of that connection goes under the processor.

Here's an image from my bench ECU (older revision PCB with aluminum canned V_RH regulator, rather than the solid state one on your PCB) that highlights the pin patterns from above and below to make the identification more clear. In some of my images I used a backlight to be able to see through the PCB.

In the image with the backlight, it looks like the pinhole that I labeled 'f' connects to a mid-plane copper trace. I expect that this plane is a ground plane (since it would make sense for V_RL to be at ground, and V_RL is connected to all of the other analog inputs through capacitors to stabilize them).

I am really curious to know where the other end of that V_RL line ('d') goes... It looks to be level with the third hole up ('a'), and about as far in as the VSSA (I don't think this should be at ground) and VDDA pins. I would guess it's connected to VDDA, the AtoD ground pin. If that's the case, then maybe there's an internal short in the AtoD, or one of the inputs is overloaded? aaand I am back to guessing again...
 

Attachments

·
Vendor
Joined
·
6,306 Posts
On 05s, it’s 99% of the time the 02 leads that fry the ECUs. A) the exhaust heat breaks down the wiring under the loom. Might look OK at the surface, but the wires are scorched earth when the onion is peeled back. B) Folks will buy aftermarket 02s with the wrong pin out or they will get the DIY plugs and then pin out incorrectly...

In either case, the 05 harnesses are the worst of the “Elige” lot and they’re the oldest and they will tend to fry the ECU if they were not well cared for... add to that the problems with aftermarket 02s and there ya go...

I’ve not followed this thread closely, but usually don’t need to know the **likely** answer on a case like this... There’s always a first for something new, so carry on just in case!

-Phil
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #13
I am really curious to know where the other end of that V_RL line ('d') goes... It looks to be level with the third hole up ('a'), and about as far in as the VSSA (I don't think this should be at ground) and VDDA pins. I would guess it's connected to VDDA, the AtoD ground pin. If that's the case, then maybe there's an internal short in the AtoD, or one of the inputs is overloaded? aaand I am back to guessing again...
I do appreciate your help. Don't think it's an on-board problem since another ECU plugged into my car blew in exactly the same place. Has to be external to the board. And since I just changed (Toyota re-wire) the post-Cat O2 sensor and those pins are right there, that's a lot of coincidence. My solder job was good enough to pass smog but a few weeks of heat may have shorted them out at the sensor.

I've passed all this info onto the shop and they are going to look at the O2 harness. If that's not it, I'm getting the ECU from them pulling the processor and we'll see where that trace goes.

BrianM
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #14
On 05s, it’s 99% of the time the 02 leads that fry the ECUs. A) the exhaust heat breaks down the wiring under the loom. Might look OK at the surface, but the wires are scorched earth when the onion is peeled back. B) Folks will buy aftermarket 02s with the wrong pin out or they will get the DIY plugs and then pin out incorrectly...

In either case, the 05 harnesses are the worst of the “Elige” lot and they’re the oldest and they will tend to fry the ECU if they were not well cared for... add to that the problems with aftermarket 02s and there ya go...

I’ve not followed this thread closely, but usually don’t need to know the **likely** answer on a case like this... There’s always a first for something new, so carry on just in case!

-Phil
I'm on my second (used) engine harness. Thanks for the tip on the O2. Hope that's it.
 

·
Registered
Joined
·
733 Posts
I do appreciate your help. Don't think it's an on-board problem since another ECU plugged into my car blew in exactly the same place. Has to be external to the board.
I agree; I was thinking along the lines that some sensor was overloading the AtoD function, and the associated electrical current has to pass out through the AtoD power pins. rather than a processor-internal failure, this would be caused by an external source. the tricky thing about this hypothesis is that such an event leads to damage on the AtoD power traces, rather than on the traces from the sensor that caused the fault.

I've passed all this info onto the shop and they are going to look at the O2 harness. If that's not it, I'm getting the ECU from them pulling the processor and we'll see where that trace goes.
rather than destroying the ecu further, you could always just use a multimeter. that way, you might be able to sell the ecu to someone who wants to try and tinker or attempt repair. in that regard, i have a conflict of interest here...
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #16
Late yesterday, the shop found the heat shrink I used on my O2 sensor swap has split and probably is shorting. Unfortunately they looked before testing for a short so I'm only 90% sure that is it. Obeisance thanks for the note on the pins that's what triggered them to go look.

BrianM
 

·
Registered
Joined
·
733 Posts
Late yesterday, the shop found the heat shrink I used on my O2 sensor swap has split and probably is shorting. Unfortunately they looked before testing for a short so I'm only 90% sure that is it.
there may be ecu actuated relays on the heater power lines, thus a multimeter probe of the harness could miss such a short. I'm just guessing here.
 

·
Registered
Joined
·
3,126 Posts
It is easier to lose confidence in electronics than to gain it back. Good luck. I hope it works out.
 

·
Premium Member
Joined
·
64 Posts
Discussion Starter #20
SUCCESS!
Car fired up today, is the good news, the bad news is it was my soldering that failed and caused me all the trouble.
Splicing in the Toyota O2 sensor was the worst money I ever saved.

Thanks All for your help and advice.
BrianM
 
1 - 20 of 23 Posts
Top