Board voltage low

[eman911]

Let me preface by saying that I know just enough about electronics to be dangerous.
The backstory:
I build an MS2 v3.0, loaded firmware and bench tested for approximately 30 mins on a JimStim with perfect results. The ECU then sat on the bench for about 6 weeks while I built the engine and wired the harness. I installed the ECU and powered it up, not realizing I had not attached the chassis ground. It is possible that it made intermittent contact with the ground stud while I was puzzling things out. I attached the ground and attempted to do a sensor check. I could not get the PC to find the MS so I spent 4 hours chasing a com problem with no results. I then removed the ECU and attempted to test it with the Stim and found it would power up and work normally for 15-30 seconds before powering down. I volt tested the board and found I did not have 12v anywhere I was supposed to, more like 2.5v. I resoldered all of the ground and power pins at the 37 pin connector with the same results.
How should I proceed?

[billr]

Remove the daughterboard (MS2) and check for voltages on the 40-pin socket of the mainboard, per the manual "build instructions". Post results.

[eman911]

Found the problem. Too much thermal paste on Q9. I cleaned off the excess, reassembled and bench tested for over an hour. Flawless!

[billr]

I'm glad you got it going, of course, but "too much thermal paste" can't possibly have caused the problem. You must have disturbed something that was shorting when cleaning Q9.

[racingmini_mtl]

I'm glad you got it going, of course, but "too much thermal paste" can't possibly have caused the problem. You must have disturbed something that was shorting when cleaning Q9.

Actually, some thermal pastes are conductive (for example those containing silver). They should be avoided in any application where you need isolation. Using little paste may help to keep isolation but that would still not be a recommended method as this may change with time and could lead to a less than optimal thermal performance.

Jean

[billr]

Ah, I wasn't considering electrically conductive thermal compounds!

[prof315]

Ah, I wasn't considering electrically conductive thermal compounds!

The heat sink paste that comes with DIYAutotune MS kits is in fact conductive. I have switched to using a non-conductive silicone based heat sink paste for any TO-220 cased component that needs a mica insulator.

[DaveEFI]

But surely, if only for looks, you'd clean off any excess afterwards? In the same way as you'd clean off flux residue?

[racingmini_mtl]

But surely, if only for looks, you'd clean off any excess afterwards? In the same way as you'd clean off flux residue?

Of course, you can clean off the excess but even that might not be enough if the paste is conductive. And it's not the same thing as flux residue because a bit more thermal paste is better than not enough.

Jean

[DaveEFI]

Surely if the problem was the supplied paste in the kits, others would have had the same result?

The reason I clean off any excess paste is it remains sticky. So might pick up something from the workbench that could cause a short.

[prof315]

Surely if the problem was the supplied paste in the kits, others would have had the same result?

The reason I clean off any excess paste is it remains sticky. So might pick up something from the workbench that could cause a short.

Others have.

[eman911]

My enthusiasm is short lived. Another related problem has reared its ugly head. The ECU runs forever on the Stim, all voltages check good. When I plug it into the car harness it will run for 2-minutes and shut down. The 12v side is good, but 5v side drops slowly to 2 volts.

Any ideas?

[eman911]

I just logged the episode and it shows the battery voltage (reportedly) starts to rise exponentially from start up peaking at about 25V and then shut down occurs.

[billr]

Disconnect the MAP sensor and TPS and anything else connected to the 5V ref; disconnect right at the MS DB37 connector. See if that changes things.

[eman911]

Ok, I've narrowed it down to the distributor pickup. I must have it wired incorrectly. Anyone familiar with Bosch Hall effect? 3 wires, red, green, black.

[prof315]

Ok, I've narrowed it down to the distributor pickup. I must have it wired incorrectly. Anyone familiar with Bosch Hall effect? 3 wires, red, green, black.

Yes the connector on the distributor side typically has a + a - and a 0 + is 12V - is ground and 0 is signal. The signal pin is pretty much always the center pin