Burnt out ECU - help needed please!

Hi everyone, I'm looking for a bit of advice on some trouble I'm having with my ECU. Just for some background: The car (1998 C2 manual) failed when motoring at high speed and would not restart. The car was recovered and the garage found the ECU was badly burnt out. They then replaced it with another unit and that also over-heated while driving under load above around 3k RPM (fine at idle) and the same component on the ECU was damaged. Looking at the images it shows the problem to be with the first transistor nearest to the connector. I desperately need to know what that first transistor controls in the car as a fault there must be what is causing the unit to overheat and burn. We cannot put in another ECU and risk that being damaged. Does anyone know or have any more specific information which may help identify the transistor in question? Or some thoughts on the issue more generally? Thanks in advance for anyone who might be able to help. Will

Trace the track through to the connector if you can, then check the wiring diagram to see what it connects to. MC

The black chips attached to the side casing are i believe either injector or coil transistors .. This is not an overheat as such but a wiring overload / short circuit and im guessing it will be a coil pack . I can't say which one but from posts here if you have surefire coil packs fitted then they have at times caused issues on 997,s . MC has already said it .. you need to see which pin in the burnt area leads to which item so its studying wiring diagrams im afraid . For me though and a guess .. it's a coil pack issue . Check fault codes if the unit still powers up .. all you need is a code for a misfire to point you in a direction .

Thanks MC - I think I know how I'll be spending my weekend!

Thanks Ian - appreciate you taking the time to give it some thought. Do you think it could be related to the fact I've got one cam solenoid on each bank which are out of action? I'm thinking that might be causing the wiring overload / short you've mentioned? Having read another forum post which suggested that transistor was linked to the solenoids, it seems a bit too much of a coincidence to me...

Will Really sorry to hear about you car - I had been in to visit the garage where it is earlier in the week and they had shown me the first ECU that got smoked. I'm not sure if the 996.2 ECU (my car) is the same or not but I offered to let them have my car to compare yours to if it helped trace the fault. If the circuitry on yours is too burnt to trace to a pin and the ECU's are the same then the offer still stands. I will drop by in my car and they can compare what they need to help trace the issue.... Obviously they may already have this all under control or the ECU may be different but if it helps the offer is there.... James

OK, OK, so I know nothing of modern Porsches.... But, is it possible to check a wiring diagram to identify the pins/sockets at the ECU for each of the individual (?) coil packs and those of the injectors and check for a low resistance, to identify the coil pack or injector circuit that may be causing the issue..? I might expect signs of overheat on the loom plug/pin...? Had there been an injector or coil pack issue would this not have been picked up by the data logger..? Also I guess if nothing shows up it is possible that there could be an intermittent short in the wiring, given the time taken to blow the unit as opposed to it blowing on start up..? But then again that could perhaps be down to the low resistance in the circuit just drawing enough excess current to cause overheating in time.... But looking at the damage caused it does seem like an intermittent short, as opposed to a general overheat... Dunno for sure.. Sorry to read of your predicament hope you get it sorted soon.. :?:

I think you might just have found the fault .. the cam solenoids on early cars can short internally .. in theory it could damage an ecu .. it kinda rings a bell with me anyways . I would certainly get your garage to have a good look at them either way . Image of the dme pins ...

The main culprit for wrecking ECU's to that extent is a bad battery or alternator or fuse links that have been fitted with larger fuses than specified. Change the battery, alternator and check all fuses when fitting a replacement ECU. Jasmine porchalink have used ecus or can get them

Was the car running ok with its variocam solenoids not working? What symptoms did you experience with this?

My understanding of things .. which is in no way factually correct .. it's just how i interpret these systems .. The dme will usually switch an earth to a unit to activate it with the unit having a seperate 12 volt supply . If its an internal short or even a wiring short joining the power to the earth the dme supplies then the dme suddenly gets 12 volts shorting to earth internally . This will generate heat and pretty damn fast .. things can melt .. its down to the speed of the components to switch off the earth feed which is now compromised , flag a code and not try and switch it back on until the code has been cleared via a tester . Can i just say , this is in milliseconds . At which point it will repeat the procedure . i do know that more modern tech can switch faster than older tech .. obviously .. it's not uncommon to get a fault code and a system inop that won't work until the code is cleared .. in other words we tell the control unit its all ok and you can go back to work it then will do so up until it sees the fault again . Older tech and you may well burn out the control unit . Sorry for saying a 996 is old tech .. but i'm afraid it is in electrical terms The vario cam will default to basic setting , it only adjusts when told to .. if you have had the car for years and this failed then you would notice a difference .. if not you may well expect its normal and as such wouldn't know there was a fault . The dme can also supply a voltage to sensors .. normally 5 volts .. a fault on a single sensor .. a short wil drag down the voltage to all the other sensors on that rail .. a rail being the 5v supply to a set of sensors .. in this case you will get fault codes from unrelated sensors to what the actual fault is , a complicated and time consuming fault to find .. Been there and done that .. it takes ages and i know full well has caught out an OPC that had to send the car to us to fix ( us being a different OPC at the time ) .. me and my colleague took a couple of hours on that one to track it down .. Trust me i'm not brilliant and working with another tech makes life a lot easier .. especially in this case .. rare as hens teeth the fault was , ( a cam sensor short that flagged as a throttle position fault ) one of my / our best fixes ever ... credit to my mate for the help .. even though he will never read here ... Matt :thumb: If you use Mid-Sussex for work .. ask for Matt to fix your car .. top notch mechanic .

Check the fuse which supplies the 12v feed to the solenoid in question. Make sure a larger one hasn't been fitted. If there is a fault the fuse should blow. As mentioned, the ECU switches the 0v side, so the ECU fuse is not affected. MC

https://www.quora.com/How-can-a-battery-damage-an-Electronic-Control-Unit-ECU Your ECU has been totally fried and the alternator and battery would always be near the top of my list as the root cause. The above saves me having to detail why and as well as the ECU, other circuits can get damaged/fried on the way to the ECU melting down.

My ECU had to be replaced after a variocam solenoid failure (car is early 1999 M96.01 engine). Replaced both (new solenoid, reprogrammed 2nd hand ECU). No issues in the over 2 years since. Drivers side cylinder bank. However the car was drivable by disconnecting the solenoid cable, So ECU wasn't dead, but can't recall exactly why ECU needed replacing. Were your fuses all ok?

Thanks for the messages all - it's great to know there are so many people out there are willing to help however they can. So, we're going to go down the route of replacing both variocam solenoids, and seeing if that does the trick (which I'm pretty hopeful it will do), given the burnt out pin in the ECU connector (no. 25) is responsible for camshaft adjustment, and the pin correlates to the burnt out transistor in question. which too is related to camshaft adjustment. It's all pointing towards the world's most expensive solenoids! Luckily, the mechanic stopped the test drive short (having thought they had solved the issue by changing the coil packs), of wrecking the replacement ECU unit (although there was some burn damage to the same transistor), so it should act as a useful guide for a test run once the solenoids have been replaced. I've managed to pick up a second replacement ECU for £180, but I'm hopeful with some careful micro-soldering and a replacement transistor that the slightly burnt replacement unit can be fixed. I'll keep the thread updated on how I get on over the next couple of weeks. Cheers, Will