[Diy_efi] Audi idle issue

urq urq
Tue Feb 28 05:21:18 UTC 2006


Yes indeed, Audi was using CIS on the I-5 turbos sold in the US in '89 ... 
'90 was the last year for the turbos, but ISTR the NA engines continued for 
a year or so after that (CIS-E).

It sounds like you have a Bentley ... use that as a reference.  I must say 
that a good many of the recommendations I skimmed from the earlier posts 
seemed a bit scary to me!

IME it really isn't very difficult to pull the air box and metering unit on 
your car.  I guess it can be a bit tough to get the big boot to seat, but 
even that's not *that* bad.  Rather than trying to figure something out, I'd 
recommend opening it up and doing the basic setting per Bentley.

There is some diagnostic capability into the ECU in your car ... follow the 
instructions per Bentley to read the codes out.  It might give you some 
clues as to whether the engine is running rich or lean.  I also heartily 
recommend use of www.sjmautotechnik.com ... Scott has collected a lot of 
useful information ... most of it is from the Bentley, but it is a bit more 
concise.

Think about the way the CIS mixture control works ... as the metering plate 
moves up it causes the metering rod to move up, exposing more of the 
metering slit, and as a result more fuel flows to the injectors.  Now use 
this to analyze what happens as you adjust the "CO Adjust" screw ... if 
screwing it in a particular direction causes the metering rod to be 
positioned higher for a given metering plate position, it is richening the 
mixture.  If you can get the engine to run ... perhaps by giving it some 
throttle as it warms up ... there is a signal you can hook a dwell meter to 
and measure the "OXS Duty Cycle" ... which is the feedback from the OXS to 
control the mixture.  Normall the duty cycle is around 50%, if you see less 
it means that ECU is trying to lean the mixture, and richen it if the duty 
cycle is greater than 50%.

Perhaps it does make a difference on a Ferrari, but on this Audi you don't 
need to worry about covering the mixture adjustment hole ... it's on the 
unmetered side of the metering plate; the worst thing that can happen is 
that *unfiltered* air will get into the engine via the hole ... but it will 
be *metered*.

It has been a while, but I'm pretty sure the metering plate is set up with 
an "anti-sneeze" mode which allows backfires to drive the plate below the 
bottom of the cone slightly before hitting the hard stop, minimizing the 
chances for damage.  If you think about it, turbocharged cars can be a 
little bit more robust in this regard as any "sneeze" has to go through the 
turbo's intercooler and the compressor before getting back to the metering 
plate.

HTH!
Steve B
San Jose, CA (USA)
... been messing with CIS on Audis in some form or another since 1978 ... 
and with two Audi turbos that have basically the same system you have ...


>   Good point about the plugging of the idle mix adjustment hole!
>   Another thing that can happen is a SEVERE backfire can bend the linkage
> between the plate and distributor. The reverse flow from the backfire will
> force the plate "the wrong way" and if it comes up farther than the damper
> mechanism allows for, the linkage can bend.  This will cause a severe
> rich-idle condition, with the black smoke and fouled plugs. If the bend
> isn't too bad, you can adjust it and restore it to proper operation.
>  I would expect an intercooler full of gas fumes, igniting could qualify 
> as
> a "severe backfire." Under normal conditions, fuel is not in the
> intercooler, but after sitting stopped, injectors could leak a little and
> the fumes can fill the intake blumbing.  Any little "cough" during 
> cranking
> would ignite this unintended fuel/ air charge.






More information about the Diy_efi mailing list