Date:         Tue, 6 Sep 1994 15:53:49 -0700 (PDT)
Sender:       Vanagon Mailing List <vanagon@vanagon.com>
From:         hugie@lindy.stanford.edu (John Huguenard)
Subject:      Re: Hesitation at low RPM

Jim Macklow scribbled:
> 
> Hello,
> 
> > 
> > On Thu, 25 Aug 94 14:45:03 CDT Jim Macklow said:
> > >My '87 Syncro still has the following problem:  ...
> > >For the FI experts out there:  Is this an indication that my computer
> > >is/has gone bad?  Or is there some sensor that the computer ignores
> > >in the first 30-60 seconds that is bad?
> > >This behavior is fairly consistent, whether the engine is cold or
> > >warm.  I seem to have a longer reprieve from the low-power demon on
> > >cool, cloudy mornings.
> > 
> > i'd bet on the coolant temp sensor. just replace it. don't check it. mine
> > all tested 'good'. finally, when i got disgusted and started just replacing
> > parts, when i replaced it, the problem was 'solved'. i still don't know why
> > it does that, but it has done it twice.
> > 
> > it could also be the oxygen sensor.
> > 
> > the way the behavior of the engine was explained to me was that if the
> > computer gets confused by null readings or out-of-whack readings from any
> > of its sensors, it (the computer) goes into Limp-Home mode, which doesn't
> > give you much power, but it keeps the car running.
> > 
> 
> I decided to check to see if I was going into Limp Mode, so I disconnected
> my coolant temp sensor.  There was no change, still had hesitation at
> low rpms.
> 
> Then I reconnected the coolant sensor, and disconnected the O2 sensor.
> Bingo!  Without the O2 sensor attached, I have lots of power at all rpms,
> and no degradation when the car is warmed up.  On the downside, it seems
> to my very subjective eye which is looking at the gas guage, that my
> mpg has gone to hell.  Like, maybe I'm getting 15, but probably less.
> Lots o power, though.
> 

The O2 sensor (on my admittedly air-cooled engine) puts out high
voltages (around .9 Volts) when the mixture is rich (too much gas).
So grounding the input (0.0 Volts) will make the computer think the
output is lean and will compensate by enriching as much as possible.
This will give terrible gas mileage, but in general will increase
power.

With the sensor disconnected the input should be "pulled" to the
middle of the range (around 0.5 Volts).  In other words the computer
thinks that the mixture is burning perfectly and will make no
compensation on the amount fo fuel being injected.  I think this is
"open loop" mode where injected gas is determined mainly by the output
from the air box and the engine temperature sensor.

I don't remember now how O2 sensors go bad; i.e. whether they short
out (0 V output), develop high internal resistance (which should be
the same as running open loop, or if they develop a high voltage
condition.  I think only the last possibility would explain your
symptoms.

You could try measuring the output of the o2 sensor with a DMM.


> Is this "evidence" that the O2 sensor is bad?  Should I ground the loose
> wire instead? (maybe the van goes into "drink till ya drop" mode when it
> gets 0 volts from the O2 sensor?)
> 
> Got any idears?
> 
> -Jim
> macklow@sybase.com
> 
> 

good luck

John Huguenard
hugie@lindy.stanford.edu