Q14 gone thermonuclear, hole in PCB, salvage???

[FridgeFreezer]

A V3 PCB has come to me that has a "significant" fault with it, namely Q14 has exploded and blown a hole straight through the PCB if you imagine taking a ~6mm drill and running it through the PCB directly where the transistor is supposed to be, that's what it looks like, but with charred edges. I'm assuming that R37 has gone open-circuit or something and all the injector current has gone through Q14. The damage is quite "neat" and localised, which may be a blessing.

As per usual, I've been asked if it can be saved, or failing that at least salvaged as a test-rig or something. Now, I'm fairly sure we can live without the transistor, but I can't find PCB layouts for the V3 PCB so I can't tell what's underneath it or close to it that could bite me.

Can anyone help? Are there PCB traces out there that I've missed? I can find V1 & V2.2 but no V3.

[trakkies]

I don't think there are details of the PCB track routing published - might make it too easy to counterfeit. And I don't have a bare PCB here at the moment to see where the 'surface' tracks close to that go. My feeling is that since it's close to the edge, wouldn't be too hard to bodge a repair.
I think I read somewhere you can buy a new PCB only at a favourable price if you return the old one. Try DIYAutoTune for details.

Incidentally, if it was the failure of the high current R37 that caused it, was it still fixed to the heat-sink? I dislike the 'glued' on ones supplied, and substitute bolted on types.

[FridgeFreezer]

I'm sure you're right with the reasons for not publishing, I was kinda hoping someone who knows might be able to cast an opinion or even a small glimpse of the square cm of PCB around that transistor. Maybe they're all too busy with the shiny new MS3 stuff to talk to us luddites running MS1

I have plenty of bare PCB's, unfportunately the inner layers are hard to see even without components in the way and I don't consider it worth the time/effort to unsolder everything from one PCB and put it onto another - at best I'd salvage the MAP, CPU, and maybe the power devices + heatsink. I was considering asking a dentist to x-ray one just out of curiosity but I have little enough time as it is without running round on errands like that.

To be honest I generally bridge R37 as all the ECU's I build are aimed at vehicles with Hi-Z injectors, not using PWM, etc. so it keeps it simple & robust. Likewise I bridge out R43, and lately R40 (and omit Q19) as the protection circuit seems to cause more problems (running just off saturation and therefore generating heat) compared with just allowing it to run.

[trakkies]

Now I'm not certain about this - but on the V3 I can see the 'surface' tracks quite easily. And I thought the inner one was just a ground plane? As regards the current limiting, I have done several early Rover V8 EFI engines with low impedance injectors - originally with resistors - and have driven the injectors directly with no problems. But knowing the dangers, followed the setting up instructions prudently.

[FridgeFreezer]

Close:

The V3 PCB consists of four layers (top, bottom, and two inner layers), compared to only two layers on the V1.01 and V2.2 boards. The inner layers used for power and ground planes,

Which says to me you need to be careful about the inner layers/tracks shorting together, very hard to be sure without knowing what passes nearby. No problem cutting the surface tracks back a bit for safety.