[T3] Art Sterrett's Brain

[Jim Adney <jadney@vwtype3.org>]

Some of you may remember that Art Sterrett has a '71 that would run 
as long as you got the fuel pump to run by some other means. We took 
him thru all the troubleshooting that we could imagine, always 
insisting that there was little chance that the brain was at fault. 
In the end, however, Art diligently did all the tests and finally 
bought a 311 D brain from me with the understanding that he could 
return it if it didn't fix his problem.

Well, as he reported here, the new brain DID fix the problem and Art 
was kind enough to send me the old brain (Abby Normal) so that I 
might be able to figure out what its problem had been.

This weekend I actually got around to opening it up, and one of the 
first things I noticed was that one of the transistors was sort of 
crooked. Out of habit, I reached down to straighten it a bit and was 
astonished when it just came loose in my hand. All three leads were 
rusted thru!

This transistor was one of several large metal can transistors, but 
this was the only one with a little heat sink pushed over the top of 
it. Now Bosch built these brains to really nice standards, so they 
used some little 3-pointed star spacers to position each transistor 
just the right distance above the PC board. Most of the transistors 
where just soldered in that way, but this one had been done 
differently.

This one had a little silicone rubber pad under the transistor body, 
presumably to give it a bit of a cushion since it had the extra mass 
of the heat sink attached to it. This was the only one with a heat 
sink and the only one with the little silicone pad. The transistor 
leads had just been pushed thru the soft pad, so the pad was tight 
around each lead. And where this pad was, that was where each lead 
had rusted thru.

I pulled off the heat sink so I could read the part number of the 
transistor, but found that it was unmarked. Nevertheless, the old 
transistor was still good, so I could verify which leads were which 
and tell that this was an NPN transistor. 

The next day I took the brain the rest of the way apart, so I could 
get to the under side of the PC board, picked a reasonable substitute 
from my stash, and soldered it in. While the board was up I checked 
and found that wire 19 went directly to the collector of this 
transistor and that its emitter was connected to ground. This all 
seemed reasonable and was exactly the way I would have done it.

Today I put it all back together, leaving the cover off the brain 
box, and installed it in my '71. Now it ran the fuel pump and worked 
just fine on a short test drive. I'm not sure why the heat sink is on 
there, because even after the test drive, this transistor felt no 
warmer than ambient (even if I removed the heat sink.)

I also took a few minutes and looked thru some of the other brains I 
have around here, and this is what I found.

    Late brains seem to have moved that transistor. It's not in the same 
    location. I'm not sure where it moved to.

    Early brains have that transistor in the same location, but they 
    don't have a heat sink or the silicone pad on that transistor.  

    I have one other brain that I could reach in where that transistor 
    seemed "loose." It's not like all the leads were rusted thru, but 
    maybe one was. Any one would still put that transistor out of 
    commision.

I'm guessing that this is likely to be a problem in mid generation FI 
brains where they are in damp climates. I believe Art lives in 
Alabama. Is that where this car spent most of its life, Art?

If there are other of you out there who had this same problem, I can 
now fix them, but I'm afraid that getting in and out of there takes 
several hours, so repair is not likely to be any cheaper than a good 
used brain.

And, BTW, I traced the wires that lead out to the injectors and found 
that they appear to lead directly to a 6 Ohm resistor in series with 
each injector. So Dave and Constantine were both right about the 
injectors not getting the full 12 V. I guess Bosch figured that full 
voltage would be okay for the short time that they were energized by 
the tester (or maybe there are resistors in the tester that I've 
forgotten about.)

-- 
*******************************
Jim Adney, jadney@vwtype3.org
Madison, Wisconsin, USA
*******************************

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
List info at http://www.vwtype3.org/list | mailto:gregm@vwtype3.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~