96 Passat GLS. Starter cranks but engine won't start

Hi all,

Would really appreciate a word of advise. I bought this '96 Passat GLS
(2.0 L, manual transmission) with 120K miles last summer and this is
the first time it gives me trouble.

It started to rain on Thursday but the car was behaving fine: no
problems with starting and smooth idle at about 900 rpm. It rained all
day Friday. Friday night the car started at once in the rain, but the
idle was not stable for about 20-30 seconds. RPM was going up and down
but in 20-30 seconds it stabilized. Drove home (~100 miles) with no
problem. I was alarmed of course. My first thought was that this is an
ignition problem. Coil or wires or old distributor cap/rotor when in
marginal condition are known to result in weather dependent problems.
So, I thought to look at it over the weekend.

It kept raining all night and Saturday all day. Since I don't have a
garage, I figured I will get to diagnosing the car at some later day.
Saturday evening, however, the car won't start at all. The starter was
cranking and nothing was happening. On Sunday there was less rain, but
still humid and the car still could not start. I checked the spark at
the ignition coil output. The spark was present and looked normal. I
took apart the distributor cap. Cleaned the electrodes inside the cap
and the rotors. They were not immaculate but didn't look too bad
either. Reconnected everything. The car still won't start. Disconnected
the ignition wire from the first cylinder and connected it to a spare
spark-plug to check for the spark. The spark was present. Reconnected
the wire back. The engine still would not start

I must confess, I am puzzled now. I owned an '86 Golf (1.8L, m/t) for
10 years. It doesn't have too much electronics, like this newer Passat,
and troubleshooting is rather straightforward. What I did with the
Passat now is what I would normally have done with my Golf because
experience tells me that most weather dependent problems are in
ignition system.

Now, I would really appreciate an advise. I don't own a VAG-COM tool to
check for codes at this '96 Passat but is it absolutely necessary in my
situation? My thinking goes that if this is not because of the
spark-plug, it is most likely a problem with fuel delivery. Marginal
fuel pump may also be "weather dependent" (when the pump was failing in
my Golf, it was running worse on the rainy days). However, I have no
idea even where it is located in the Passat. Is it in the tank or under
the car? My Golf has two pumps in each of these locations. Can I kill
any sensitive electronics by pulling out a Fel Pump relay and putting a
jumper in place to hear the buzz? Any advise would be really
appreciated.

Thank you all in advance!
Andrew

P.S. Just in case anyone's wondering, there is gas in the fuel tank

Just a followup to my note. I did some homework that I should have done
before this posting by checking online parts catalogs and Bentley.
Please don't curse me for not doing that before writing. Anyway, if I
understood correctly, this car has a single Fuel Pump located in the
trunk accessible through the luggage compartment. I also pulled out the
FP relay and fuse and tested them. They are both fine.

Now, the Bentley says that I have to "connect a remote control VAG
1348/3A adaptor cable 1348/3-2 between terminal 4 a battery positive
(+)" and try to hear the buzz. I have no idea what these VAG tools are
but I suppose the jumper in place of a relay would do the trick. Am I
right? Or am I on a totally wrong track? Any advise would be greatly
appreciated.

Thanks to all who will respond!

Andrew

Just an update on the situation. Today the weather was nice and dry and
the car started at once. I figured I still have to try diagnosing the
problem and since the fuel pump was the main suspect, I got to it
through the luggage compartment. The top of the pump assembly has a for
pin connector on top (two conductors to power up the pump and the other
two for the gauge). The + conductor of the pump (a thick red wire) had
a visible damage near the connector. The insulation was worn out and
about 0.5 cm of the wire was exposed. Also, the wire was corroded with
about 3-4 of about 7 strands visibly broken.

I will fix it with a butt connector but it was getting late and dark,
so I just put some electrical tape on the damaged wire. It seems to me
that this may explain the symptoms. In high humidity the pump was
getting to low voltage because of the leak from the exposed wire. Hence
the symptoms.

I also found out that the spark-plug wire boot was cracked. I was able
to see the spark in the dark. So, I am looking at replacing an ignition
wire set or at least one damaged wire.

Thanks for the suggesting to check for the buzz, when ignition is on.
The pump is indeed supposed to turn on briefly to pressurize the
injection system and I could not hear the buzz. However, I can hear the
pump when the engine is running. I find this strange and would
appreciate any advise as well as any comments on this repair.

Thanks again!
Andrew

In rec.autos.makers.vw.watercooled, elms88 wrote:

>
>I will fix it with a butt connector but it was getting late and dark,
>so I just put some electrical tape on the damaged wire. It seems to me
>that this may explain the symptoms. In high humidity the pump was
>getting to low voltage because of the leak from the exposed wire. Hence
>the symptoms.

I would look for a different explanation for this part of your
problems. Humidity shorting the wires should not have much effect at
low voltage.

Tom, thanks for your reply. I agree with you low voltage cables are
less prone to humidity effects. However, let's remember that this wire
goes simply along the car body and about 0.5 cm of its length was
exposed due to damaged insulation. Of course paint provides some
insulation from the car metal but on the wet day there may well be
moisture in that area, so I would assume it could drain down the
voltage that the pump actually gets to the value not sufficient for the
pump to operate. Is this an unlikely scenario?

Also, the wire in the vicinity of the damaged spot is corroded. This
looks like a 7 stranded wire and some strands are broken. So, this may
result in heat losses (the pump may well draw several amps I think) and
again the voltage drop. I plan to cut off the damaged part, install a
butt connector and seal it with shrink tubing.

Right now the engine starts OK. I still don't understand why I can't
hear the pump turning on briefly when I turn on the ignition. I can
hear the pump buzzing when the engine is running when I get close to
the pump.

Thanks again!
Andrew

In rec.autos.makers.vw.watercooled, elms88 wrote:

>Tom, thanks for your reply. I agree with you low voltage cables are
>less prone to humidity effects. However, let's remember that this wire
>goes simply along the car body and about 0.5 cm of its length was
>exposed due to damaged insulation. Of course paint provides some
>insulation from the car metal but on the wet day there may well be
>moisture in that area, so I would assume it could drain down the
>voltage that the pump actually gets to the value not sufficient for the
>pump to operate. Is this an unlikely scenario?

Yes, unlikely to impossible.

The reason is that even salty water is not going to short out enough
current to keep the wiring from supplying enough current. The
resistance of a short due to moisture is unlikely to be under 100
ohms. If there were a conductive-enough short, I would expect the
fuse supplying the pump to blow rather than having the pump be
starved for voltage.

Now if you have moisture what I would not be surprised to find with
it would be a corroded connector. That would be the opposite of a
short: an open. An open can be intermittent and might start working
again after some vibration. I guess it is possible that the air
humidity to affect when the open circuit keeps current from flowing.
It is hard to see how that would occur at just those times, but I
could see that.

So the action item in addition to adding tape to prevent a real
short would be to clean any contacts. I don't know if there are any
connectors there. If there were, I would clean them up. I would
apply deoxit or other contact cleaner. I am not saying there is a
connector. I am saying that if it is, that I would clean it. I am
not at all experienced with dealing with the rear fuel pump, and I
am glad it has not come up for me.

Now for trouble shooting, here is what I would do from your case. I
would get a multimeter and set it to the 15 volt or next higher
range. When turn on the ignition. When the pump does not start, I
would go back to those wires and probe to see if there is voltage.
If you do have an intermittent open, what may well happen is the
pump starts up as you are wiggling the wires. That is a legitimate
troubleshooting indication. But if you can see voltage to the pump
and the pump is not running, expect the pump is the problem. If you
cannot see voltage, then you cannot be expecting the pump to run.
Somebody may be able to tell us how long that voltage should be
there with the ignition switch on and the engine uncranked.

I have gotten way past what I know (a moist/wet pair of wires will
not short enough current to keep the pump from running.) A problem
like yours is interesting. When you do find what it is, please let
us know.

Tom, thanks again.

Yes, there is a connector on top of the fuel pump assembly. It is a
four point connector (2 wires for a pump and 2 for a fuel gauge). The
wire bundle ends with a mating connector that plugs into the connector
on the assembly. The damaged wire is in the bundle and the location of
an insulation wear is about an inch from the connector. I looked at
the connector yesterday and it looked OK.I sprayed the contact points
with a contact cleaner from the can (don't remember the name) that the
local parts store sells.

There is a little chance for the hypothetical open at the connector
area to be closed by vibration. The car didn't start in the rain and on
the sunny day it just started. It was sitting motionless. Well, of
course opening and shutting the door may be enough vibration too, so it
is not impossible but unlikely. Your point that the fuse didn't blow is
a good one. Too much current due to the current leak should have blown
a fuse. As for your question for how long the voltage should be at the
pump with the ignition switch on and the engine uncranked, Bentley only
says "briefly". In my Golf I estimate It to be about a second.

I will tell what I find out and I would appreciate more opinions.
Andrew

In rec.autos.makers.vw.watercooled, elms88 wrote:

>As for your question for how long the voltage should be at the
>pump with the ignition switch on and the engine uncranked, Bentley only
>says "briefly". In my Golf I estimate It to be about a second.

I could see two ways for that voltage to be there briefly. One would
be that it is timed. The other is that a controller would shut down
the power once sufficient pressure is a achieved. I don't know which
way it is done.

Tom's VR6 wrote:

>
> I could see two ways for that voltage to be there briefly. One would
> be that it is timed. The other is that a controller would shut down
> the power once sufficient pressure is a achieved. I don't know which
> way it is done.

Don't know. My Golf's Bentley, which is written in by far more plain
language than Passat Bentley, says that the pump turns on briefly to
pressurize the system. I thought maybe Passat, being ten years younger,
simply holds pressure better than the Golf, and, that's why I could not
hear it. To check I pulled out the fuel pump fuse with the engine
running, so that it would stall by itself after a some seconds. Still
didn't hear this pressurizing buzz after that when turning on the
ignition. The engine started instantly though, as if it was
pressurized. Maybe it pressurizes in no time and I cannot hear it.

To see the voltage appearing at the connector when turning on the
ignition would be a good check. Will try it when I get to continue with
the car.

Thanks!

Did you do a wet test of the spark plug wires and ign coil (AKA ign.
transformer) to see if that is your problem now that the engine is running?
Take a spray water bottle and start misting your spark plugs, then wires to
the distributor working your water to the ign. coil.
I would say that your ign. coil is the problem. Engine will probably die
when the coil gets misted.
You can either replace the coil (BUY ONE FROM THE DEALER ONLY) or take it
off, clean it well and coat it with a thick application of epoxy. I have
done this to some ign. coils and coil packs with success. Some coil packs
go for over $400. yuck!
If the coil is bad you may want to change plugs, cap, rotor and wires too.
;-)
--
later,
dave
One out of many daves.



"elms88" > wrote in message
oups.com...
> Just an update on the situation. Today the weather was nice and dry and
> the car started at once. I figured I still have to try diagnosing the
> problem and since the fuel pump was the main suspect, I got to it
> through the luggage compartment. The top of the pump assembly has a for
> pin connector on top (two conductors to power up the pump and the other
> two for the gauge). The + conductor of the pump (a thick red wire) had
> a visible damage near the connector. The insulation was worn out and
> about 0.5 cm of the wire was exposed. Also, the wire was corroded with
> about 3-4 of about 7 strands visibly broken.
>
> I will fix it with a butt connector but it was getting late and dark,
> so I just put some electrical tape on the damaged wire. It seems to me
> that this may explain the symptoms. In high humidity the pump was
> getting to low voltage because of the leak from the exposed wire. Hence
> the symptoms.
>
> I also found out that the spark-plug wire boot was cracked. I was able
> to see the spark in the dark. So, I am looking at replacing an ignition
> wire set or at least one damaged wire.
>
> Thanks for the suggesting to check for the buzz, when ignition is on.
> The pump is indeed supposed to turn on briefly to pressurize the
> injection system and I could not hear the buzz. However, I can hear the
> pump when the engine is running. I find this strange and would
> appreciate any advise as well as any comments on this repair.
>
> Thanks again!
> Andrew
>
>