Re: Ur-q duty cycle wrong.

[Robert Houk - WorkGroup Server Firmware <Robert.Houk@East.Sun.COM>]

   Finally the Frequancy valve is working.
   I was given a properly working computer to try and see if the frequancy
   valve would work.
   It does.
   However there is a problem with the duty cycle.
   When the car is first started the duty is ~50% and moves up or down.
   Sometimes it actually goes down to about 37%.

Is this a MAC-02 ECU? If so...

"Cold", the duty cycle should be rock steady -- not bouncing around!
It should start anywhere from 65% (very cold) more-or-less slowly
but steadily dropping to around 50% (at, say, 80-85F block temp), 
taking a quick plunge to around 40% (at, say, 90F block temp), then
rising again quickly to 50% (call it 95F block temp) before going
closed-loop and bouncing around +-5% (nominally, more or less) from
what should be a 50% center point (my experience is more like 60%
slowing dropping as the block temp rises and finally stabilizes).

   The closing of full throttle valve causes the duty to go to 77%.

Fine.

   And yes the idle switch is working fine and the wire to the ECU is fine
   and the ground for the switch is also fine.

If you close the WOT switch at idle, the duty cycle should lock at 50%.
This tests both switchs, wiring, etc. does the engine stumble when you
do this (yes if it really needs the enrichment...)

   Once the engine gets a bit of a temp into it, the duty cycle shoots up
   to 96% 

This is NotGood(tm)!

           and stays there unless I close the full throttle switch at which
   time it goes back to 77%.

   I tried to adjust the CO from all the way rich to all the way lean
   (untill the engine stalls).
   No change in duty reading.
   Also unplugging the Oxygen sensor makes no difference, the wire is fine.
..........................................................^^^^^^^^^^^^^^^^

This doesn't necessarily follow, but I believe the ECU will go open-loop
if it thinks the O2 sensor is not working, and hold at 50%, so I suspect
the wire is indeed fine...

Running at 96% says the ECU is maxing out trying to enrichen the mixture.
(by the way, there is some anecdoctal evidence that the ECU can fry its
FQ drive transistor this way...)

The first thing I would check is the voltage coming off the O2 sensor.
0.5V is nominal, lower is lean, higher is rich. From the above, you
should see very low voltage, probably very steady.

My guess would be either faulty O2 sensor, or low fuel system pressure.

Look at the plugs -- do they look fouled (sooty-black)? That would in-
dicate actually running rich (bad O2 sensor); do they look white or
very light color -- that would be lean (too "hot"), or low fuel pressure.

That's where I think I would start... (O2 sensors are, relatively
speaking, cheap -- I'd just replace it on GP's...)

   I know that the air temp sensor gives reading of 2400 Ohms rather than
   13 Ohms at room temp, but it should only affect ignition timing and
   boost cut-off.

@4000RPM. Should be nominally around 80mv, as I recall, rising up to
140mv for highest temp... Definite problem, but shouldn't be causing
the main problem above!

   I checked all the wires for continuity and they are fine.
   The fuel pump relay I am using is has the right part number except for
   the letter which is G and it's supposed to be F.
   I can't see from the wiring diagrams how this would make any
   difference.  The fuel pump works fine.
   I am stuck.  Any hints would be greatly appriciated.

Good Luck!

   1982 Audi Coupe Turbo (475,000 km)
   1982 Audi Coupe Quattro (155,000 km) Coupe with 4kq floor
   1983 Audi Ur-quattro (220,000 km)
   ...

Sheesh, trying to corner the market, or something?

					-RDH