200 10V stumbles, poor mileage

[urq]

I spent some time reacquainting myself with this one by reading some of the
messages to date and pulling out the Bentley.

The patient is a 1991 200, 10V with MC-2 engine, correct?  I did not find a
direct quote of this, but inferred it from tidbits here and there, so I
wanted to start here ... 

The problem is that the engine runs poorly below 3000 RPM, but seems to
recover above that speed.  I don't recall seeing anything about ECU codes,
do you get any codes while the engine is running?  Have you read them while
the engine was malfunctioning?

The current suspicion is that the cold start injector is running when it
should not be; this being borne out by measuring the voltage across the
terminals to the CSV.  The power (12V) is supplied to pin 2 of the CSV via
the Blu/Blk wire which is ultimately connected to the fuel pump relay.
Seeing +12V at this terminal while the engine is running is expected
behavior (that said, you should not see +12V when the engine is stopped,
even with the ignition on).  Terminal 1 of the CSV goes to pin 1 of the ECU
via a Blu/Red wire.  This pin should be at ground only when the ECU wants to
add extra fuel via the CSV.  I don't have the detailed functional
description of the ECU, but I'd expect this to be when cranking the engine
(cold and hot), and perhaps in certain cases of high boost (I don't know
this for a fact).  I would *NOT* expect to see the CSV activated on an
engine that was running at or near idle or below 3000RPM with light load.
It sounds that perhaps the ECU has been replaced for a similar condition in
the past, is this true?  I'd recommend that you verify that there's not a
short to ground inside the CSV itself, or in the wiring harness between the
CSV and ECU before trying another ECU.  I'd be very skeptical that it would
be some sort of hack from a PO ... unless you have seen some other
"performance improvement" hacks in the car.  

All this said ... do you know for a fact that the CSV is the source of your
poor performance?  A simple test of this hypothesis would be to unplug the
CSV while the engine is running and see if the performance improves.  If it
does not, the CSV is not the source of your problem and we should start
looking elsewhere.  

Good luck on getting this nailed quickly!
Steve Buchholz

Happy New Year and all the best for 2010 to all the audifans!

-----Original Message-----

That what I thought how it should work. But there is 12V across the
connector when car was running for awhile (and was hot). Can someone test it
on a running car? :) I recall from long time ago that other people reported
that CSV was engaged well after cranking. 
-- Konstantine 

----- Original Message ----- 
From: "Ben Swann" <benswann at verizon.net> 

That does not sound right. 

The injector should pulse during startup when the engine is cold. Duration
of pulses is dependent upon engine temp. The pulse is caused by a switching
to ground AFAIK - either through the ECU or some other means - depending
upon ECU. 

There should not be a constant current path on both side of injector
terminals. There would be a positive voltage present on one terminal all
times and shared with other devices (blue/black wire), but other terminal
should not go to ground except through the ECU(pin 1). 

That is the way I recall without looking in more detail at the manual or
other schematics. 

See http://www.gtquattro.com/MCWiring.html for wiring details - MAC-14 is
similar. 

Ben