Re: Stumped

[luis@odf.ucsd.edu (Luis Patino)]

Hi Jon, I have an idea that might start your car when it's warm. I am assuming here
that both your pump and your relay work. My '85 4ksq behaves in the same way as you
described in your "Stumped..." posts, and both mine and yours have the same KE-Jet
(CIS-E) injection for 5 cylinders, so my idea might get you going until you (and I
can find the permanent cure).

When your car refuses to start, unscrew the top bolt on the fuel distributor, but
be VERY CAREFUL since gasoline will gush/spray forcefully the minute you loosen
the bolt (this again is the top bolt that screws in horizontally and has no fuel
line attached). Therefore before you unscrew the bolt make absolutely sure that
you don't have left the fuel pump bridged and running, and also surround the bolt
with a sizeable enough rag to catch the escaping fuel - cover the bolt with the rag
so you can not even see it, to prevent the upward spray of fuel that will result
and likely hit your face and eyes! It goes without saying that this must be done
as far as possible from any ignition sources. With the bolt covered, loosen it
with a 12 mm open-end wrench about 10 turns and wait 10 s. You can now look at
the bolt (but still keep the rag under it to catch the seeping gas) and wait until
no more gas comes out. Loosen it some more and if more gas comes out wait until it
stops. Now tighten the bolt and start the car (having put that gas-soaked rag in
a safe place). If it doesn't start again repeat the procedure time and time again
until it does. If the first time doesn't do it maybe the 5th or 6th will, but the
car _will_ start if you try enough times.

It may take a few repeats of this procedure to start it because your car, like
mine is in a hot climate (FL and CA resp.). What causes this frustration is a 
phenomenom called vapor lock: the heat from the warm engine rises to the fuel lines
and gets tranferred to the fuel. If the engine is runing, the fuel circulates, so it
doesn't have time to absorb much of that heat. However, if the engine has warmed up
and then been turned off the fuel sitting still in the lines will stay there to 
absorb all the heat that the engine gives them. Certainly, the temperature under
the hood exceeds generously the boiling point of gas at atmosferic pressure, so the
gas can potentialy boil inside the lines creating bubbles of fuel va[por that 
intefere with the meticulously engineered fuel circulation which invariably prevents
further flow of fuel. Since the boiling point of a liquid rises under higher 
pressures, the K-jetronic was designed to conserve some of the pressure accumulated
from the effort of the fuel pump, after the engine is shut down to keep the boiling
point of fuel beyond the reach of underhood temperatures. Further, this residual
pressure must remain high enough for long enough time while the engine bay cools 
down. In order to maintain this pressure at its specified value, the system has
components like the pressure accumulator and seals at points where pressure may be
lost. With time and bad fuels (new "emissions controling" gasolines?) they eventualy
fall off adjustment or wear out. Sometimes the failure is partial, so they maintain
enough pressure to prevent vapor lock in cold days, so you'd only notice the problem
mainly in hot days or after particularly hard runs of the engine or if using a 
particular kind of gasoline with lower boiling point. Sometimes even with the 
injection in perfect order a combination of these other circumstances might cause
the freak no-start problem - especially in turbo engines which for this reason have
extra cooling lines, fans and shrouds (and still get vapor locked!).

Actually, before you try the last-recourse solution above, try this other one that
works for me except the in hottest days. Bridge the fuel pump and run it for 5 
minutes by itself, then quickly replace the relay and start the motor. Again, don't
give it up too soon: try this a few times, and then try it at cooler times of the
day, since this sure beats spilling fuel. Let me know if any of this advice works,
since it helps me make sure that I'm on the right track.

This weekend I'll try to fix what I think is leaking. In your case perhaps the
permanent cure is the sensor-plate play that I already fixed in my car and for
a while solved the no start. Read my previous post and request for help.

Good luck!

Luis
85 4ksq
----- Begin Included Message -----

>From owner-quattro@coimbra.ans.net Mon Oct  6 19:36 PDT 1997
Date: Mon, 6 Oct 1997 18:07:48 -0700
From: luis@odf.UCSD.EDU (Luis Patino)
To: quattro@coimbra.ans.net
Subject: 4ksq doesn't start when warm
X-Sun-Charset: US-ASCII
Sender: owner-quattro@coimbra.ans.net
Content-Type: text
Content-Length: 4688

Now that the list helped me so much with the overheating "problem" - thanks! - I
would like to ask you fellow audiphiles about my car's other significant annoyance.
The symptom is that if the 4ksq runs long enough to get warm and then it's turned
off and sits for more than a few minutes, it will absolutely refuse to start - no
matter how long its engine cranks. And crank it will, with plenty of juice, but it
will absolutely refuse to start.  However, if I let the car sit until it cools down 
completely, it will start perfectly, as it always does when completely cool. Other
than this problem, the engine performs fantastically, actually so well that I
wouldn't care about this annoyance except for the _hours_ it takes for the car to
cool down.

Did I sound stupid in my first paragraph? I hope so, because I did not want to
prejudice anyone with what I have figured and found about this problem, so that
in the - unlikely, I also hope - case that I am wrong, someone will be able to
tell me.

Of course I don't wait for hours until the car cools down. Since it has done this
or something very similar since I bought it 1.7 years ago, I quickly researched
the reason for these faults. Having eliminated lack of spark and pump-relay 
malfunction, I realized that fuel wasn't getting to the injectors when the car
was warm. I didn't know why until I couldn't start it at a gas station after
fueling. There someone who knew VW's showed me that if I opened the bolt on top
of the KE-Jetronic fuel distributor and let fuel come out, the car would start.
He didn't know why, but it sure worked. However, seeing the bubbles and pressure
that got relieved along with the fuel (I guess doing this was a fire hazzard) I 
realized that gasoline was boiling somewhere in the lines and that the resulting
vapor was blocking the flow of the fuel. Somewhere I had heard of a condition called
vapor lock, so, I thought, that was it. Later, when I got my Bentley manual, I saw
the series of checks that have to be done when a warm engine doesn't start. I was
supposed to check for how well the KE held its residual pressure, since it's 
precisely this pressure that prevents the fuel from boiling. Thus, what it boils
down to is that the pressure is leaking somewhere in the injection. At that time
(and also now!) I didn't have fuel pressure gauges, but searching the web for 
KE-Jetronic, I found an auto encyclopedia site that explained how KE-J works, and
explained that if the air-volume-sensor plate did not have the right play,
1.0 - 2.0 mm then the stop-screw o-ring would not seal and cause residual pressure
leakdown. It also mentioned all the other possible leaks, but since to check for
those I needed the pressure gauge, I went for the easiest cause of the problem
the one that cost nothing to fix, and sure enough the sensor plate had too much
play. Using the stop screw underneath the fuel distributor (I made the tool to
turn the screw by shaving the prongs of a minifuse), I adjusted the play, and bingo!
the car started starting - I had eliminated the vapor lock.

Unfortunately, last winter, the car started misbehaving again to the point that now
it doesn't start whenever warm. Luckily, since I learned to bridge the fuel pump
circuit at the relay connection, I bridge it now to let the fuel pump compress the 
bubbles instead of having to purge the bubbles through the top screw. Bridging the 
connection works almost every time except when the weather is hot. In those 
instances I have to both compress and purge many times before the car would start.
Thus, I can always get it to start, but have to put up with embarassment from
passengers and passers-by.

This time the cause of the pressure leak has proven more elusive, and I don't have
the testing gauge yet ($$$). The sensor-plate play is 1.1 mm still as I adjusted it,
so I guess that's out. However, when I checked the play last weekend, I discovered
that after running the fuel pump to pressurize the system (to measure the play
correctly) a mild fuel odor came from underneath the plate, that is from the air 
intake box. So I lifted the lid with the KE assembly on it and felt with my fingers
around the plunger lever from where I picked up brownish goo with strong fuel odor.
Thus, the leak is probably from the o-ring that seals the plunger in the fuel
distributor.

Should I replace that ring? There are no other obvious fuel leaks on the outside
of the fuel lines, pump, filters, etc. Can someone in the Southern California
(I live in San Diego) area lend me the fuel pressure gauge? Anything that I might
be missing? Has anyone with the same problems corrected them successfully?

Thanks!

Luis

4ksq 132k mi


----- End Included Message -----