Several years back I had intermittent  problems with my '82 a/c, auto
Transporter (Vanagon) that I traced to the ECU.  I didn't ever find the exact
reason the engine would stop, but often it had the symptoms of a rich mixture
either just before or after stalling.  Finally I tapped components in the ECU with
a toothbrush and the problem went away for ?4 years.

Recently it returned with a vengeance.  I held up several HUGE lines of peak-
hour traffic when my vehicle cut out on the way home from work.  Knowing it
was the ECU, I finally slung it from the roof of the vehicle in a cloth sling - open
to the world - and it didn't miss a beat.  I could NOT find the problem and finally
closed the ECU up and put it back in its place.  Next day, the vehicle stalled
twice within 2km of home and I was pretty exasperated.  I did my best to find a
replacement ECU - Australia, New Zealand, Holland, Germany, UK , Canada
and USA.  Finally I had one Dale Kreuger tell me he had access to one.  I had
also tried to find whether the manual-version ECU would suffice, but not one
person could tell me.

One Saturday I decided I would make one last effort to find the faulty
component or joint in the ECU before forking out big $$s.  I was very rough on
the thing, and it didn't hesitate ONCE!!  WAS it really the UCE after all?

I wiggled the head temperature sensor wire and the engine hesitated and
stopped!!!  I had replaced the sensor 4.5 yrs ago, to no avail.  I started the
engine and tried the same thing, but this time the engine ran perfectly no
matter WHAT I did with the wire.  I was puzzled!  What had I DONE the first
time, that caused the stalling?  I realised that had touched the PRESSURE
REGULATOR!!  I tied it again, and sure enough, the engine failed.  I could
repeat this at WILL, as long as I pushed on one side of the regulator OR
pushed in on the vacuum connection, the engine stalled.  THIS was IT!!!!
Eureka!  THEN I found that even by leaning on the engine tin anywhere across
the back of the engine I could stop the beast dead!  Further evidence it was
some sort of flexing in the pressure regulator (which is bolted through the tin).

After a long chase around, I paid close to $100 for the genuine Bosch part
(same part from VW was >$130). I put the regulator in and connected
everything up and the problem was STILL there!!  I was NOT impressed at this
stage.  The other possibility (I was grasping at straws) was that pushing on the
regulator reduced pressure in the fuel lines just enough to stall the engine.  I
measured fuel delivery rate and this was OK.  I spent half a day making a
pressure gauge to screw into the test point.  After two generous petrol showers
I convinced myself that pressure was OK.

Out of exasperation I decided to take out the temperature sensor.

The darn thing was<bold><bigger><bigger> LOOSE</bold><smaller><smaller>.  Flexing the tin had caused it to touch the
temp sensor and move it.  THIS is what the problem had been all along, and
this is why the rich mixture symptom was so obvious.  When the contact
between the temp sensor and the head was poor (high resistance) the ECU
enriched the mixture! I tightened the sensor, and all was 100%!

Moral to the story?  When you have a fault that you mutter about for months,
saying "It must be electrical", believe yourself.

For what it is worth.

MANY thanks to those who have commented and replied to various messages I
have posted over the past couple of months.

David Yates

'82 auto Transporter, downunder.