When i installed my Steve Wong chip, i re-mounted the DME on some cushy foam rubber to absorb some of the vibration- maybe cheap insurance...

I recall reading a thread recently relating to the ICV driver circuitry (output drivers)....Loren chimed in on this (spot on) with a study of heat specs on these particular components, as one reader speculated that these parts would bode well with better heat sinking capabilities. Ends up not really being the case. Member corrected a poor running condition (although) by replacing the drivers (IIRC). Haven't seen other failure modes relating to "running" conditions other than obvious, hard failures.

I'm looking forward to extracting mine (DME) and spending some "quality time" under bright light and magnification. As well, I'd like to scope several output points and have a look at the signals...will use my storage scope and post the photos of each waveform at the board's output connector. Should be interesting,......(wish I had a pinout box for this).

The cushioning of the box itself (mentioned above) is most definately a plus(why not?), as to reducing transmitted vibrations...smart.

Best,

I've long been intrigued with the idea of using a scope to diagnose engine troubles or motronic problems. I asked for info here:

http://forums.pelicanparts.com/porsche-911-technical-forum/387581-using-oscope-dme-engine-testing.html?highlight=sun+tester

but didn't get a single response. If you could post some shots of output, hookups, and info on obtaining an appropriate scope, I'd be greatly appreciative...

ianc

Steve-
with the magnified view of the failed joint, the solder joint looks pretty grainy, and given the component above it's heat output, the failure looks more like a "disturbed joint" which lead to the failure. A disturbed joint is where the solder is soft or not fully set and the then the joint is moved (like as the car is bouncing down the road)- the graininess is caused by the solder not setting properly (develops a crystalline structure I believe). Why that may be interesting is that if it is a disturbed joint, that means normal operating conditions are causing the solder to re-melt, which is a very different problem than just cracked joints, which would presumably be caused by fatigue failure or some other mechanically induced problem.

Part of the real fix may be a different solder alloy with a higher melting point...or perhaps a conformal coating on the board.

And any ideas on the yellowish precipitate? How did it get in the box and where did it come from? If it does conduct, even slightly, it will change the output of each component in the system.

Matt

http://forums.pelicanparts.com/porsche-911-technical-forum/387581-using-oscope-dme-engine-testing.html

best,

I've repaired several DME with solder failures under this exact transistor. I know mechanics that have also taken apart several DMEs over the years to resolder these exact joints for this transistor. See the end of this post from a year ago for the exact same failure:

http://forums.pelicanparts.com/porsche-911-technical-forum/330641-carrera-no-start-motivated-greed-not-throw-towel-yet-4.html


http://forums.pelicanparts.com/uploa...1172346599.jpg

I think most of the failures are due to heat and/or over current at these junctions. This transistor switches on and off and provides the ground signal to the ignition coil to allow it to charge. The other terminal of the coil is a fixed 12 volts. It's possible that defects in the ignition coil cause some type of short that overloads this transistor. I once worked on a car that had a full Nology ignition system installed. The DME ran so hot, you really could have fried a egg on it. The case temp had to be over 200 degrees and almost burned my hand. I believe the shop who was working on it had resoldered this joint once. I couldn't say if the ignition system was what overloaded it, as I don't know how it was wired it, but if it drew it's current from here, then I would not be surprised.

Bad injectors will also cause a similar failure to the fuel injector side of the circuit. The fuel injector transistors are on the opposite end of the board mounted on a standoff aluminum heatsink. They work in the same way in that the transistors switch on and off sinking the current, acting as the grounding, while the other end of the injector is hot with 12 volts. I have a customer with a race/rally 3.2 that could not figure out why his car kept shutting down in his ralllys. He burned up three DMEs, spent about $4k in parts, replaced every single component/sensor, removed and did a full wiring harness scan, before we traced it to bad injectors. The interesting thing is that he could drive it for hundreds of miles on the street and there was never a hiccup and the DME never was hot. However as soon as the car was under racing conditions, where full throttle was norm, the DME fried and shut down within 3 miles. The big grey resistor next to the burgundy capacitor was burned black, along with some damage to the auxiliary components.

So if you only subscribe to books and manuals, which one tells you to check the solder joints in their trouble shooting guide? Or is this just good ole common sense? Good Job!!!

Steve,

Great stuff! Now that I have my DME opened I will be checking. By the way a failed solder joint like this will induce it's very own heat because of the higher resistance in the joint and only make the matter worse. This type of failure/condition in electronic boards is a death spiral.

I also repaired such issues on my DME relay already.

Thanks for the great thread.

Now for the big question, when the transistor at T-504 fries, has anyone found a modern day equivalent?

The DME schematics show a NPN Darlington pair for that component. Steve W's pic shows a TO-3 package. A search turned up a transistor made by ST Microelectronics that is marketed as a ignition coil driver in the TO-3 package. ST's schematic for this part looks like the same one on the Bosch schematic.
http://www.st.com/stonline/stappl/productcatalog/app?path=/pages/stcom/PcStComGenerateTableView.onClickOfRPN&primaryheade r=Transistors&secondaryheader=Transistors%2C%20Pow er%20Bipolar&subclassheader=Darlington&subclassid= 88&rpncode=75077&producttype=product&open=&applile vel=

Is the ICV bad or has it been replaced recently? If so, the ICV driver circuitry (output drivers) will be the source of your problem. Ask me how I know.

Steve... I have a source! A good friend of mine restores old MG's, purely for hobby. He bought a case of these awhile back... :D



http://forums.pelicanparts.com/uploa...1202358554.jpg

sooooooooooo..... where do i buy the super zoomy full race/rally/street never have to think about the fuchin damn thing ever again 20 year leap in technology one for my 87 cab?????

or do i just carry a spare and hope to hell it hasnt come apart at the seams.


another item that has to have a 20 year quantum leap in techno to solve problemo once and for all.

NEW AND IMPROVED would apply here!

Damn, I knew this was a test. You.........

I am seriously thinking uping my DME

I don't see why that one wouldn't work. Digi-Key even has 212 in stock for $3.50 each.

Steve, here's food for thought:

I doubt that anyone on else on this board (well, maybe one person, but who cares) has as much knowledge about the DME and fuel maps as you.

What a fabulous thing it could be for those of us who are not electronically inclined to put some of that knowledge into a small pamphlet you could sell.

Personally, I'd pay $$$ for a little guided tour of the DME with pictures and info on what each component does, failure modes, and some basic troubleshooting info and perhaps an overview of each component for the layman. Shouldn't have to be much longer than the DME test plan I shouldn't think. The thing is such a black box that anyone without a detailed knowledge of electronics, like myself, just can't make head or tail of it.

With a few evenings work, you could probably knock out something pretty good. Post it on your website with a Paypal link and wait for the cash to pile up. If you did ever think about writing such a thing, sign me up as your first customer!

ianc

how about just a new and improved...........bombproof,vibration proof, beer proof, waterproof, dust proof, heat/cold proofed DME in its very own lil soft and cozy foam lil house with spare DME attached that nothing can bother and it outlives all of us for some reasonable cost amount with a lifetime guarantee and free shipping!

after reading about DMEs and then having one fail...........it seems this concept would never have made it on the NASA SHUTTLE.........well maybe once, since its built on lowest bid!

Had the same cracked joints in my DME 2 years ago. Drove me nuts trying to find the problem. Soldered them up on the advice of this great board, put the DME back in, never had the problem again. Prior to that I was all but about ready to sell my little car. Man, am I glad I didn't. :)

Shoot me his contact info!!

Whatever I know on troubleshooting and addressing all the problems on these cars, I'd post it online for everyone to access for free. I've just been too swamped with work and various projects these past couple of years to put it all up. Thanks for the idea though!

And hopefully he still has the ultra rare refill harness that is required for a proper fill:





http://forums.pelicanparts.com/uploa...1223670049.jpg