I have an '88 S4. I normally never have a problem starting the car until
yesterday. I left the car outside and we hard a very cold night and a hard
freeze. I went out in the morning to move the car and it would not start.
It cranked over but would not catch. I started giving it gas and I finally got
it running. It was running very very rough and finally died. I gave up and came
back out in the afternoon when it finally warmed up to around 50 degrees
outside. The car fired up as usual. I have only encountered this problem one
other time last year during a snow storm.
The other problem I have been having may be related. When it is around freezing
temperatures outside I start the car and it fires up as usual. The problem is
that it stays idling @ 1500 rpm instead of dropping down to @ 800 rpm. After it
starts to warm up, I get surging between these two rpm points very fast like a
switch. After a mile or so of driving, things go back to normal. It does not
happen every time it freezes, but I have noticed it 4 times this winter.
Several possibilities here...
The most common problem with hard cold-starting is a lack of fuel - that is, the
mixture isn't richened enough to crank. Only gasoline vapor will burn - liquid
gasoline will not burn. If the injectors do not spray enough fuel in that enough
evaporates in the cold intake and combustion chamber to make a burnable mixture,
hard starting and poor running will be the result. You might want to check the
Temp II sensor resistances on both sides.
Another possibility is a lack of air. The S4 idle is completely
computer-controlled by an electrically-operated air by-pass valve. If that valve
is sticking at low temps, it might be stuck at the last position, which would be
warm idle, which would not let enough air in to burn the increased amount of
fuel that should be injected for a cold start.
If you can tell whether the problem is too little fuel or too little air, you
would have a start on fixing it. The smell and color of the exhaust would be one
way to get a clue.
Another possibility is old spark plugs that are fouling from the moisture in the
cylinder. Each gallon of gasoline burned makes a little more than one gallon of
water vapor. If the plug insulators are dirty, the moisture from the first
firing in the cylinders can condense on the insulators, shorting them out. The
clue here is that the engine will fire and immediately quit firing, then start
firing randomly with continued cranking.
On carbureted engines, ice in a fuel line could cause this kind of problem.
Water would sometimes collect in a low spot in a fuel line, and the fuel flow
would be low enough that fuel could be pumped by the water - until it froze. On
the 928, the fuel pump moves the maximum amount of fuel at all times, and unused
fuel is bypassed back to the tank. The high fuel flows make it much less likely
that water could sit in the fuel lines to freeze.
My first suspicion, based upon both of your problems, is a sticking idle control
valve. This would "flood" the engine as the fuel injected is increased, but the
idle air isn't increased. The problem here is that if you open the throttle at
all, you kill the idle fuel enrichment and ignition retard, which would then
make starting harder. You can try holding the throttle down an eighth of an inch
after two seconds or so of cranking and see if that helps.
If the idle valve is sticking, you can try the solvent-in-the-vacuum-line trick
- you should be able to find that easily in the archives or on the Nichols web
site.
Wally Plumley
928 Specialists
==========When trouble shooting, always start with the “easy” and inexpensive
items first and work your way to eliminate problems until you end up to the
difficult and expensive items. The majority of electrical problems are
connection related.
If the ECU doesn't know that the engine is cold, it does not provide cold start
enrichment and it will not start. Eventually after cranking enough, there may be
enough “raw” fuel sprayed into the intake to start the engine. You would see
smoke after starting because the fuel wasn’t metered properly.
The Temperature Switch II is located in the front center of the engine just to
the right of the fuel pressure damper.
1. Check the plug to the sensor, is it on tight? Are the contacts clean? If not,
clean the connectors and clip it on tightly. Check engine starting now. If it
does, you’re done. If not got to step 2.
2. To check the temperature II sensor, you’ll need an ohm meter. Remove the
connector from the LH control unit. Connect the ohm meter to pins 2 and 5 of the
LH connector (not to the pins on the module itself). The correct value ranges
are:
0 C / 32 F: 4.4 - 6.0 k-ohm
15 – 30 C / 59 - 86 F: 1.4 to 3.6 k-ohm
40 C / 104 F: 0.9 – 1.3 k-ohm
60 C / 140 F: 480 – 720 ohm
80 C / 176 F: 250 – 390 ohm
Since your problem is cold start, you could just look for the proper resistance
range for the ambient temperature of the cold engine. If the sensor is shorted
(zero ohms resistance) the mixture will be too lean and the engine will not
start when cold. For problems at other engine temperatures, I’m including all of
the resistance ranges for higher engine temperatures. If the sensor has an open
circuit (infinite resistance), the fuel mixture will be too rich, the engine
will not run when warm, and will be difficult to start when warm.
If the values aren’t correct, measure them directly at the Temperature Sensor II.
You’ll see that there are two prongs on the sensor. There are two separate
temperature sensors housed in the one sensor body, one for the LH and the other
for the EZK module. Note the orientation of the protruding alignment notch on
the outside of the sensor. If you are viewing the alignment notch orientated to
the left side, the prong closest to you is the prong for the LH controller and
the prong further away is for the spark control module. Clip one of the ohm
meter leads to the prong closest to you (LH) and clip the other lead to a ground
point [Do NOT connect the ohm meter leads between the two prongs of the sensor].
Repeat the resistance measurements.
If the resistance readings are in the correct range, there is a problem with
wiring to the LH controller or the connector itself could be corroded. If the
resistance readings are not correct range, the temperature sensor must be
replaced.
If replacing the sensor doesn’t fix the cold start problem, check the fuel
filter, fuel line pressure, injectors etc. Fuel injectors can be removed and
tested for flow and flow pattern. There are several shops that provide this
service. The problem could also be attributed to bad ground connections. Also,
don’t overlook potential weak spark conditions. If you’ve eliminated all other
areas, a failing LH controller could be the problem.
Rich Andrade