Date:         Fri, 7 Apr 1995 00:49:27 -0500 (EDT)
Sender:       Vanagon Mailing List <vanagon@vanagon.com>
From:         ringo@spike.aerodyne.com (Chris Getschow)
Subject:      Re: 85 Vanagon starts hard

> 
> On Thu, 6 Apr 95 18:51:14 CDT I complained (abridged by Joel):
> >My '85 Vanagon has trouble getting started.  Here's the scenario:
> <snip>
> >depressing the accelerator when the engine catches and then driving away
> >immediately.  The engine runs rough for a couple minutes while I'm driving
> >(in a low gear) but then runs smoothly thereafter. I can smell gas when I'm
> >backing out of the driveway.
> >first or second try.  The second is that there's something screwed up in the
> >FI system that's causing the starting mixture to be richer the colder it gets

And Joel wisely replied:
> 
> i think you are onto the culprit, but have some "facts" wrong ... imho it's
> the coolant temp sensor gone bad or the wiring/ground associated with it.
> it sounds to me like the computer is NOT adding the necessary extra gas to
> "choke" the car when cold ... so the engine runs, but dies. then you have to
> literally flood the car to get it running. that's the gas you smell when you
> get it running. 

I *like* this interpretation!  My only problem with it is that I believe the
coolant temp sensor and its wiring to be ok so I'm looking for another root
cause. The reason I trust the sensor is that I was able to get the bus into
the garage tonight and performed some of the electrical checks in the Bentley
manual (page 24.20).  I measured the resistance of temp sensor II *at the 
computer connector* (pins 2 & 7) to be 4500 ohms.  According to the graph
on page 24.19 that corresponds roughly to a temp of 5C.  Guess what the temp
was in my garage:  right around there.

I *did* notice that resistance across the throttle valve switch terminals was
open-circuit at the idle stop.  According to Bentley (24.35) this should be
zero ohms.  I measured zero ohms at the full-throttle stop, but not at the
idle stop.  Is it possible that I neglected something in making the measurement
or is this a real live data point?

The only other abnormal thing I noticed was that the tube leading from the oil
breather to the air cleaner was both collapsed upon itself and ruptured.  I
can't figure how this enters into the problem but thought I'd mention it....

I'm a little perturbed that I see in Bentley's FI chapter a section on 
air-cooled engines, one on Digijet, and one on Digifant -- but then there are
a couple pages on 1985 Digijet!  So I got stuck with a transitional year; 
bummer. 

Joel also sez:
> check the connections first ... it might just be a loose wire.

This is truly sage advice in any case.  It applies equally well to making
test measurements.  And it's SO easy to forget.  I do however have a story
(non-VW) about a situation in which I insisted on checking every connection
in the tail-light circuit of a 1978 Olds Cutlass ("there's *got* to be an
open circuit here somewhere!") when my then-girlfriend had suggested: "maybe 
the light bulbs (6 of them) are just burned out!"  guess who was right 
(grrrrr....)  I married her, of course.

-- 
Chris Getschow (getschow@an.hp.com, ringo@aerodyne.com) 
Bedford, MA
'85 Vanagon GL