Those who scan my name on here will find that I've been chasing a
starting problem on an '85 Jetta for way too long. However, the car
runs after I get it started, and there are, after all, only 91 miles
of road in Juneau, Alaska, where I live, so it's very difficult to get
around to trading it off, particularly when I have this garage full of
spare parts, a really well-thumbed Bentley, and access to a really
well-equipped university auto shop for five hours every Friday night
for about $400 a year in course fees.

The car is an 85 Jetta GL with an auto tranny and CIS-E injection. The
problem is that I'm the only person who can consistently start it.
When I bought it eight years ago, it was a matter of getting in the
car, turning on the switch, letting the fuel pump stop running, and
hitting the starter once.

About three years ago, the trouble began. At first, it appeared
temperature related, but now it does it all the time. I get in, turn
on the key, let the pump do its thing, and then turn the key to start.
Sometimes, it starts right up just like the old days, but other times,
the starter just turns the engine with no fire. When it does this, I
can sometimes get it to start by "feathering" the gas pedal a little -
sort of like pumping it. Other times, I have to turn the key off, and,
usually, turn it on and off a few times so the pump cycles, and then,
with a little feathering, it will start, run for a few seconds like
it's flooded, and then it smooths out.

Sometimes, when I do this cycling the pump thing, the car acts like it
hydrolocking. The starter will spin, and then it sort of "grunts," and
(usually) starts after the "grunt."

The car gets 20-22 mpg in mixed driving, not bad for a 23-year-old 1.8
with an auto tranny. It just acts up when I try to get it started.

In earlier posts, I've listed the work so far. Over the last four
years (!?) I've replace the transfer pump, the fuel accumulator, the
oxygen sensor, the cold-start thermo-time switch, the coolant
temperature sensor, and the o-rings and shrouds on the injectors.

So, the latest development is that I managed to adapt the auto shop's
fuel pressure tester to the banjo fittings on the '85, and carefully
ran the procedure on pages 69-72 of the Bentley for CIS-E Fuel
Pressure Tests and Specifications.

I set up the tester gauge with a T-fitting with a valve on one line
connected to the cold-start valve feedline, and pulled the screw from
the fuel distributor test port (neatly hidden away under the air
conditioning line) and hooked up the other line from the gauge T. I
pulled the electrical connector off the differential pressure
regulator (DPR), and pulled out the fuel pump relay and rigged a
jumper that I could insert between the two big holes in its socket.
Then I ran the tests.

I got a system pressure of 79 psi, which is in the 76-81 psi range in
Bentley table k (p. 70: fuel pump relay jumpered, valve open, DPR
connector disconnected).

When I disconnected the fuel pump relay jumper, the pressure dropped
back to 42 psi and stayed there for the ten minutes suggested in the
residual pressure specifications table m ( p. 71: at least 38 psi and
steady).

I got a differential pressure of 74 psi, smack dab in the middle of
the suggested range of 72-76.1 psi (Table l, p. 70: 2.9 to 7.0 psi
less than my system pressure of 79 psi - fuel pump relay jumpered,
valve open, DPR connector disconnected).

The final test consisted of hooking a 15 k-ohm resistor across the
coolant temperature harness connector and reconnecting the
differential pressure regulator with an ammeter inserted into the
green-white wire going to pin 2 of its connector.

The Bentley says the pressure should have been between 61.5 and 69 psi
for my car (10 to 17.5 psi below my 79 psi system pressure), while the
differential pressure regulator should have drawn 50 to 80 ma of
current.

I got the same differential pressure as when the ignition was off: 74
psi. The meter indicated a current draw of .01 ma (ignition on, fuel
pump relay jumpered, valve closed , DPR connector reconnected).

Quoting The Bentley page 71:

   If both the pressure and the regulator current are out of
specifications, look for
   an electrical problem. Check all wires and connections. Check for
a good
   ground connection at the cold-start valve. If no other faults can
be found, the
   electronic control unit is probably faulty.

All the wiring looks OK, and the connectors all pass the wiggle-and-
pull test. The cold-start valve is functioning correctly, and appears
to be well grounded.

There's 7.4 volts on the green-white wire going to the differential
pressure regulator with the ignition on, and I measure 18 ohms across
its terminals with the ignition off, both within spec.

The troubleshooting charts indicate that the car should run rough and
get lousy mileage if the DPR is shot. This is not the case with my
car. As I said, after I get it started, it runs fine and gets decent
mileage.

So, the question for the VW gurus: what are the failure modes for a
differential pressure regulator? What should I do next?