Audi 100 Won't Start

[Marc Boucher]

It was suggested to me that I shunt the fuel pump relay in order to test it,
but I'm not certain of which contacts to shunt.

Attached is a scan of the side of the fuel pump relay containing a
schematic.  It actually contains 7 plugs.  I'm rusty on this stuff but I'd
guess that a current flowing through T to 86 closes the switches and lets
current flow through 87 and 87a, with 30 being ground.  Thus I inferred that
the correct shunt would be between 30 and 87 or 30 and 87a.  This would also
seem correct since the two largest plugs (thus largest power handling) is 30
and 87.

Thus I decided that I would create the shunt between 30 and 87.  To be
cautious I did it through an Ampmeter and just touched it quickly in case
something unexpected occurred.

The results:

Shunting 30 and 87 created a noise in the rear of the car that I recognize
as the fuel pump operating.  Note however that the power through this shunt
was about SIX AMPS!!!   Since it operated as long as the shunt was in place
I thought that I would report this result before shunting it for a long
period of time.   Note also:

This same sound is heard for abut 2 seconds when I turn the ignition
key...i.e. I can hear that the Fuel pump is energized momentarily.  This is
the way its always been.
If I shunt 30 and 87 then there is a flow of 6 amps, from even before the
key is in the ignition.  I'm not sure how long I can keep this shunted
without either burning something up or killing the battery.

Shunting 30 and 87a did nothing.  There was no voltage measured across these
two.

I figured that assuming the fuel pump relay had been bad, then shunting it
repeatedly for 5 seconds at a time would build up enough pressure to test
whether this made a difference (not sure of this).  So after having done
this, I tried to start the car a few times (no shunt in place).  There was
no change from before...i.e. a no-start.

Could you give me your opinion on whether its safe for either the fuel pump
or the battery to shunt for a longer period of time...say 30-60 seconds, in
order to more definitively test the hypothesis that the relay is or isn't
the cause of my starting problems?

Also, if the fuel pump were really bad, would it draw current and make noise
?

What is the fuel pump 'check valve' and what purpose does it serve?

Thanks again, Marc.




----- Original Message ----- 
From: "Marc Boucher" <mboucher70 at hotmail.com>
To: <quattro at audifans.com>
Sent: Sunday, January 25, 2004 6:54 PM
Subject: Audi 100 Won't Start


My otherwise reliable Audi 100, 1990 has finally let me down.  (CIS-E-III,
NF engine, non-turbo, non-quattro, automatic)

I've owned the car in California for over 4 years and its been fairly
reliable with only routine maintenance required.  Almost a year ago I moved
up to eastern Canada and its also been trouble free, including through the
tail end of last winter.  Never gave any trouble starting or running.  Its
kept indoors in a garage heated to about 10C (52F).

A few days ago we set out to the store.  It started ok but about 1/2 a mile
down the road it died and would not restart.  Battery is fine, turns over
fine, runs for 1/2 second then dies, in other words it tries but never quite
reaches a stable idle.

We pushed it home and almost 3/4 of the way home we tried again and managed
to get it started again and I parked it underground.  I've left the battery
on a trickle charge to make up for all the attempted starts.

Right now, it will only start after several attempts and then die.

I performed the input and output tests as listed at this site:
http://www.sjmautotechnik.com/100ecuf.html

and pulled the codes.   Here are the results (in order as they appeared) and
their meaning:

INPUT TESTS:
2232 Air Flow Meter plate rotary Potentiometer defective, check sensor
output and system wiring
2232 Air Flow Meter plate rotary Potentiometer  defective, check sensor
output and system wiring
0000 End of Diagnostic Output
1231 Transmission speed sender (G68), Signal used by Ignition Control Unit
(ICU), check sensor output
2411 Exhaust Gas Recirculation Valve, (California Vehicles only)
2132 Data Wiring Defective, check wiring (California Vehicles only)
2132 Data Wiring Defective, check wiring (California Vehicles only)
2132 Data Wiring Defective, check wiring (California Vehicles only)
2132 Data Wiring Defective, check wiring (California Vehicles only)


POTENTIOMETER CHECKS OUT OK:
I checked the potentiometer.  It moves freely and is centered within the
venturi when at rest.  The resistance between the bottom and the middle
contact is about 2 kOhms at rest and rises to a max of about 13kOhms near
full open falling slightly to 11kOhms at full open.  The resistance between
the bottom and top contact is a constant 4.4kOhms.  The resistance between
the top and bottom contact is 5.5kOhms rising to about 11kOhms near full
open then falling slightly to 8kOhms at full open.  From what I've read this
appear to be functioning correctly.

NO APPARENT VACUUM LEAKS:
I've searched the hoses and they seem in good condition.

OUTPUT TESTS ALL PASS:
Tests of Differential Pressure regulator, Carbon Canister Solenoid, Idle
Stabilizer, and Cold start valve all perform as expected.

I changed the fuel filter and added two bottles of gas-line anti-freeze to
the tank (tank has been at least 3/4 full anyway).

I checked all fuses and all are good and none are missing.

I erased the codes from the computer (California version has a memory that
can be erased).

I performed a cursory check for any loose wires.

I started the car again (after many tries).  It ran smoothly (I had to keep
it at 2000 rpm) for a few minutes.  Exhaust was thick white smoke which is
probably a combo of rich mixture and the -25C air its exhausting into.
After a few minutes it died (perhaps I let the idle drop a bit).  No
restart.  At this point I pulled the codes from the computer again.  This
time the following codes, in this order were pulled:

4444 No Faults have been recorded.
0000 End of Diagnostic Output
2132 Data Wiring Defective, check wiring (California Vehicles only)
2122 Engine speed sensor signal missing
2123 Full load/full Throttle switch Switch open ckt, stuck closed
2121 Idle switch Switch open ckt, stuck closed

A couple of other notes...the one problem this car has had since I bought it
has been inconsistent electrical problems.  Interior lights will stop
working for months, bulbs and fuses check out, and then they start working
again.  This has included the trunk light, the under the hood light, and
some interior lights.  Speedometer once cut out then came back next day.

I've checked that the ECU is firmly grounded, as is the intake manifold.
I've read that I should check the grounding of the potentiometer, but
neither that nor the fuel distributor appear to have been designed to be
grounded as they're installed on top of the plastic air cleaner.  I also
think that both are made of a non-conducting metal.  Strangely the fuel
filter doesn't appear to be grounded, in fact there is a constant 10kOhms
between it and ground.

Obviously I'm way out of my depth right now but I'm not quite prepared to
take it to a mechanic yet, at least not without pointing him in a certain
direction.  The problem is despite all my investigations I don't have any
more of a sense of the likely problem.  Right now I'm thinking it could be
any of the following: Bad Fuel-Injection Control Unit; Bad Ignition Control
Unit; Bad Fuel Pump; Bad Fuel pump check valve; Bad Fuel pump relay;  A lose
wire or connection somewhere leading to either ECU or between them; or any
of a number of things.

I'd really appreciate any advice or any guesses at what the most likely
causes might be.


Thank you,

Marc