MB fuel injection issues/problems

[quk@sievers.com (Phil Payne)]

HELP!

I'm having problems getting my fuel injection system to work properly.

1988 TORSEN Ur-quattro MB001421, K-Jetronic, UK specification (leaded
fuel).  135k miles, no obvious signs of excessive engine wear except
that the rear crankshank oil seal is worn, giving the characteristic
"pulley howl" that can be silenced by pulling out the dipstick.  No
apparent hose splits - injectors replaced 500 miles ago.  Breather and
instrumentation hoses all tight.  Exhaust manifold cracked and blowing,
but not yet excessively.  A new Dialynx manifold has been ordered.

Working from VAG microfiche 000.5147.00.00 (German version) with
technical bulletins 000.5147.01.00 to 000.5147.04.00.

Questions:

a)   Why is the quiescent position of the air volume sensor plate
     that controls the fuel metering head given as:

                       "1.9 + 1.1 mm" ?

     The Germans are pretty good at basic arithmetic - but wouldn't
     it be simpler to say "3.0 mm"?
 
     (I noted the _fierce_ "ACHTUNG!" about the values on the stickers
      being obsolete.  Ouch - the ones in the microfiche are a bit
      different!)  

b)   000.5147.02.00 introduces a new service tool (VAG 1348/1A)
     for setting up the fuel injection (it holds the air
     volume sensor in pre-defined positions for measuring fuel
     delivery volumes, etc.)  Is this _really_ a VAG tool, or a
     standard item masquerading under a VAG number?

Problem:

If the injection is set up at idle as recommended, the engine runs very
lean - white plugs, zero exhaust deposit - and is underpowered.
Gradually enriching by experiment yields a better result, but the CO
values at idle go to hell in a hand-basket.  Throttle response is always
erratic - constant foot pressure at constant speed on level roads gives
random (small) surges, perhaps +/- 3mph at 70mph.  Successive pedal
depressions give entirely different results - sometimes nothing at all
sometimes "nominal" performance, usually something in between.  I
suspect the cold start valve of mis-behaving because the car sometimes
stalls after a warmish start with all the symptoms of a too-rich
mixture, but I think that's a different issue.

Any ideas?  Could the worn rear crankshaft oil seal _really_ leak in
enough air to distort the fuel/air mixture that much?  How could I test
for this - constrict the breather pipe with a brake pipe clamp?

The injection system has not been fully dismantled and properly checked
against nominal measurements - would this be worthwhile?  Can wear to
the fuel metering head cause this problem in some way?

And finally - do any of our UK readers have parts of an MB's fuel
injection system for sale, so I can practice on (and perhaps
recondition) a system out of the car, which is in daily use?


Phil Payne
Committee Member - UK Audi quattro Owner's Club