I just purchased a 96 850 Turbo wagon. I live in Canada and when it is cold the car will turn over and catch for an instant but not start. I can keep the starter running but it will not start. Some times after an extended period it will catch.

I have found that if I turn the engine over for a few seconds. Then remove the fuel pump fuse, it will immediately start - run for a few seconds and then stop - I assume running out of gas. If I replace the fuel pump fuse it will start right up.

I cannot figure out why. Any suggestions?

For background here are the things I have investigated / changed which have not solved the problem to-date.

- New spark plugs - no effect
- New cap and rotor - no effect
- Then I used an obd2 reader to collect any codes. - found no codes.
- Reviewed Volvo technical - procedures published on the web for 850's giving instructions on how to trouble shot conditions - when there are no codes present.
- Other than the obvious = this document points in the direction of - potential fuel injector leaking, or faulty pressure in the system. - However, the car runs fine and starts fine after it is started once ... so it seems to me it is likely an intermittent problem. (However, this my be the next place to investigate if all else fails)
- Repaired leaking - vent pipe attached to turbo - no effect
- Purchased a subscription to Alldata - found a technical service bulletin - (Engine Ignition Conditions and Starting Difficulties) - which required checking the coil resistance and grounding of the coil, connections and battery terminal. Looked promising but - after a few hours of work had no effect.
- Further investigated Alldata and found that - a defective engine temperature sensor could cause the computer to alter the fuel injector timing. I.e. if it were defective the computer assumes a 68F temperature - felt that might be the problem starting at 32F. However, after confirming the temperature gauge on the dash works (and is powered by the sensor), and then confirmed the resistance of the sensor when cold and when hot was to spec. gave up on this idea.
- Exhaustively reviewed the wiring diagrams and component repair and operation data for a hint of why - so far to no avail.
- One form respondent speculated that this problem may be due to flooding. I.e. the fuel leaks into the head and washes the cyl reducing compression. In my case it does start after the removal of the fuse. Hence I believe my compression is ok. Might be flooding, however, but seems to start quickly after the above mentioned sequence (removing fuel pump fuse).
- Removed the main relay for the fuel system (located on the radiator cowling) and inspected and cleaned the connections - seem fine - coil resistance seemed fine as well. Also the fuel injectors do work so the relay must work as well. Could be sticking however, removing the fuel pump fuse to un-stick it sees like a stretch ...
- checked cam shaft position sensor - connections as per Alldata - found ok
- Did inadvertently touch the rubber boot on the high tension cable on the coil and got shocked (garage floor was wet from melting snow)

Why removing the fuel pump fuse - 15 amp #2 and replacing it allows the engine to start - has baffled me -HELP

I know that it is nice to fix one's own car, but when you consider your time and effort, it might be time to take your car to the dealer.

After all, that's how a lot of people repair their Volvos.

Thank you for sharing your problem with us, I certainly hope to hear what cures your problem.
--
'96 855R,'64 PV544 driver, '67 P1800 basket case, '95 855, '95 854, the first three are mine, heh, heh, 435,000 miles put on 9 bricks

James,
Thanks for the advice.
In my case I am retired do not watch a lot of TV nor play video games but enjoy a good challenge. I can start the car when it is cold any time I want with my toggle switch and so far I have spent only $200 Cdn on parts(Cap, rotor & FP relay) I am learning a whole lot about Volvo's; to say nothing of interacting with a lot of nice people who like to offer advice and help on line. The bottom line is I am a "gear head" who enjoys solving problems (not to say I do not get frustrated at times)

Cheers

David

Hi,

how about this logic:
- playing with the fuse helps>it is electrical
- next to the fuse is the pump relay, replace it for $50

my 850/-94 had once starting/keep on going problems due to the 10 years old relay not able to keep the contact providing power to the pump.

cheap and easy fix

br Tapsa

Tapsa,

Thanks for the suggestion. I nearly did that but before committing the money I wanted to see if I could rationalize if it failed what happens.

My logic went like this. If it only fails to work in the morning when it is cold yet will immediately work if I cycle it on and off then the contacts must be sticking on when cold. I.e. the engine appears to be flooded upon start up so when the ECM gives the signal to the relay to stay on for 1 second it must stick on. So I took the relay out of the car and put it in the freezer for an hour. Then I reinstalled it in the car and tried to start the engine. It started right up. This implies that it is not (sticking) when cold.

In addition I have also determined (right or wrong)that even if the fuel system is pressurized by the fuel pump longer than for 1 second upon start up the ECM which controls the timing of the injectors should determine if the engine is flooded or not. Therefore, in my mind the fuel pump should not be the determining factor in flooding unless the injectors are leaking.

However, if I cannot resolve this soon I will definitely give that a shot.

Thanks

Have you measured fuel pressure? It should be 43.5 psi static pressure. If you put a vacuum pump on the regulator while the pump is running pressure will decrease as vacuum increases. That will also tell you if you have a leak in the diaphragm.
--
Tom 69-1800S, 72-1800ES, 96-850T,2000-S70 GLT-SE

TomL

Thanks for your suggestion. To answer you question no I have not measured the fuel pressure yet. I have a fuel pressure gauge for an American car but not for the Volvo. Do you know if a pressure gauge for a Volvo is expensive or is possible to purchase fittings and use my existing gauge? I.e. what type of fittings connects to the fuel rail. It looks like a shradder valve of some type.

In any case looks like this is the next step.

Thanks again. I will let you know what I find.

David

The schrader valve at the end of the fuel rail is the same size as Ford.
The fuel pressure will fall off normally, so that after 2 or 3 minutes the pressure will be about 20psi.

When the FPR failed on my 1995 854T, it cranked for a very long time and when it fired up, black smoke came out of the exhaust. Car ran fine all day, just the initial start in the morning.

On my 1995, the FPR is under the fuel rail with the vacuum line going to the tree. On my 1998, the FPR was moved down to the fuel filter area, and what looks like a fuel line goes under the fuel rail and then down to the front of the manifold.

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Klaus

After reading your first e-mail response I was encouraged. I put the car in the garage and pulled off the vacuum line attached to the pressure regulator. I looked for any evidence of fuel since I had driven the car earlier. I could not detect any fuel or fumes. Then I put a vacuum gauge on it to see if I could find a leak in the diaphragm. The regulator held fast holding a vacuum for 15 minutes or so. I suppose that means the diaphragm is ok / not leaking at least. Unless when it is really cold it leaks, but it seems to me it is more likely to leak when it is warm.

What do you think?

With respect to the info on the gauge. Great I will source a gauge or the parts to build one soon.

Thanks again for you help.

David

David,
When you turn the key to posII in the morning routine, does the fuel pump spin for a second or two? Could it be that the 15A fuse is not making good contact when it is very cold? Swap that fuse with a spare tonight. Also push down on the 103 relay, just to make sure is has good contact.

What a hell of a good puzzle :)

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Klaus,

I am glad you have not given up on my case yet.

The pump works and pressurizes the fuel rail (I can hear it working)and I am positive the car is flooding when I first try to start it in the morning. This I believe is supported by the fact when I remove the fuel pump fuse temporarily the engine ejects the fuel and start quite quickly. I have changed the fuse and have currently installed my fake fused toggle switch. W.r.t the relay I have taken it out and cleaned the contacts a few times.
However, as explained in my recent response to NBZ I have virtually exhausted my options.... So I have bought a relay from the dealer and will install it tomorrow morning if the car does not start after all the efforts put in today.

I will post the results tomorrow.

Thanks again for your interest

David

Somehow, fuel is getting into the intake manifold when the engine is ice cold and not running. A look into the 1996 specs, shows the FPR similar to my V70, and because of the extra long travel of the vacuum line and the need for gasoline to go up hill over the engine, I doubt that it is leaking there.

The only other source of leaks could be the injectors or the charcoal cannister. The cannister is also too low to deliver raw gasoline.

Is there a mouse nest in the air filter? :)

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Once again thank your for the suggestions they give me some hope.

For info as suggested this morning (29 degrees F outside) I tried holding the gas peddle to the floor instead of disconnecting the fuel pump fuse to no avail. As soon as I disconnected the fuel pump fuse it started right up and stalled as it ran out of gas. Then I reconnected it and the car started right up. For some reason I have to cycle the power to the fuel pump relay, however, at least I can start the car when I want.

I tried disconnecting the hose leading from the intake manifold to the EVAP valve just in case the valve failed open and negative pressure was drawing gas from the Canister when the engine cooled down. The hose to the EVAP held a vacuum and I could see nor smell any gas in the hose. Next …

I purchased an inexpensive fuel pressure gauge and fitting last night. Following the Alldata - testing and inspection procedure - for a 96 Volvo Turbo I found with the pump running the system generates a steady ~45 psi. I am not certain if gauge is accurate but assume this is ok for my purpose. Pressure increased when I clamped the return line on the regulator and pressure fell when I applied a vacuum to the pressure regulator. However, the pressure did not fall 1:1 as suggested by Alldata but I assume who knows how accurate my gauges are so I am assuming the readings are ok for evaluating my starting problem. I then checked for residual pressure and the system held at 40 psi for over 30 minutes. As a result it began to look like leaking injectors might not be the problem.

In any case I wanted to be sure so I dug up an old ACTRO fuel injector tester I had and installed it along with while the pressure gauge. This time with fuel pump relay installed and cycling the ignition on/off 3 times a pressure of ~ 42 psi was generated and held steady. I assume the system generates a little more pressure when the pump is continuously running. Testing each injector in turn using the single pulse mode (for plugged and leaking injectors) I found that for all 5 injectors the system pressure dropped from the base 42 psi to ~15 psi and then slowly increased to 20 psi. All injectors showed essentially the same characteristic. According to the manual a leaking injector should shown more pressure drop than the others. Of course I suppose all injectors might be leaking. However, this test seems to imply that the injectors are ok.

To complicate matters last evening I put injector cleaning solution in the gas and drove around a bit. Although it did not have any impact on starting this morning maybe it is beginning to work. Who knows. As a desperate measure I also removed the engine control module sprayed the connections with contact cleaner and re-installed it. Just in case a poor connection is playing havoc with the signal to the injectors.

I guess I will not know if any of what I have done will help until I try and start the car tomorrow morning.

Any other suggestions would be appreciated.

Just in case anyone is interested in the testing process and logic (or not) I followed is outlined below in more detail:

Following the Alldata - testing and inspection procedure - for a 96 Volvo Turbo:
Checked vacuum hose to pressure regulator first - ok
Connected pressure gauge and started fuel pump - first jumping relay base terminals 1 & 2
Pressure measured ~ 45 (not certain if gauge is accurate) but assume this is ok for my purpose.
Clamped the return line on the regulator and the pressure rose immediately.
Connected vacuum pump to regulator and pressure fell. Not quite 1:1 as per Alldata but I assume is ok again for my starting problem diagnosis.
Disconnecting vacuum and system returned to normal pressure.
Checked residual pressure and the system held at 40 psi for over 30 minutes.

Tested fuel injectors using an ACTRON fuel injector tester with the pressure gauge connected and found:

With fuel pump relay installed and cycling the ignition on / off 3 times a pressure of 42 psi was generated and held steady. This is less than the system pressure with the fuel pump running, however, it held fast so I assume it is ok for my evaluating my starting problem.
Using the single pulse mode on each injector in turn (pressurizing the system in-between each injector) found that for all 5 injectors the system pressure dropped from the base 42 psi to ~15 psi and then slowly increased settling at 20 psi. All injectors showed essentially the same characteristic.
Using the multiple pulse mode (used to determine if injectors are sticky or sluggish) as above found that for all 5 injectors the system pressure dropped from the base 42 psi to ~ 24 psi and then slowly increased settling at 30 psi. Once again all injectors showed essentially the same characteristic.
According to the manual the single pulse mode should have identified a leaking injector I.e. the leaking injector should have shown more pressure drop than the others.
My testing implies the injectors are ok, not clogged, sticky, sluggish or leaking. Of course this is when they are warm.
Of course since all the readings are the same all injectors might be leaking yet the car runs so well other than upon a very cold start up I cannot believe this to be true.

Follow up:

This morning I tried to start the car again normally but it simply fired once and then cranked over to no avail. I then inserted my new F/P relay and tried again. Unfortunately there was no difference. i.e. the car did not start.
So I switched of the current to the #2 FP fuse, cranked it over -- and it started. I switched back on the current to the FP and it once again started fine and ran great.
Oh well ---- the computer must be giving the injectors the wrong signal and flooding the engine. It appears I am going to have to take another look at all the sensors feeding the computer.

David, have you looked at the engine coolant temp sensor yet? The volt meter readings are as follows:
0C(32F) = 7300 0hms
20C(68F) = 2800 0hms
40C(104F) = 1200 0hms
80C = 300 0hms
100C(212F) = 150 0hms

It is this sensor that determines the length of injection. Inside my garage it needs to read 7300 in the morning :) There is no entry for -20C.

Have you tried to start the car with the fuel pump off, just to see if there is residual fuel in the manifold? Do not let the fuel pump prime, just hit the starter. There should be just enough fuel in the rail to let the engine cough once or twice. Then trip your switch and see if it starts right away.

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Klaus,

Yes early on in this problem I figured it had to be the temperature gauge. I checked the connections (determined some had replaced it recently) then performed a resistance check early in the morning and then after it warmed up and the readings were within spec. Also the dash temperature gauge seems to function as expected. That would have been a nice easy fix!

I am now focusing in on the MAF sensor. Even though I do not have any error messages I suspect the MAF may be sending a faulty signal to the ECM which in turn adopts a substitute value impacting the injector period - resulting in flooding when very cold outside.

However, I am a little confused on the results of my testing. One test for the sensor stipulates while the plug is connected and key on (not running) the voltage from ground to pin #4 should be battery voltage and the voltage between pins # 2 & #3 should be .1 - .2 volts. I did find battery voltage between ground and pin #4 and found .18 volts between #2 & #3 pins. Therefore it meets that criteria yet .... when I tested the resistance between the sensor pins #1 & #4 I found 753 ohms compared to the ALLDATA spec of ~ 110 ohms. Do you know if the ALLDATA spec is right and if this means I should replace the MAF?

With respect to starting the car without the fuel pump coming on. I have not tried that yet, but great minds think alike, I was planning to try that tomorrow morning. I will let you know what happens.

Thanks

David

It is 110 0hms according to Volvo's pocket data booklet. 750 is a bit high, but I don't know what that means. When the MAF goes bad, you should not be able to idle. If you unplug the MAF, it will idle but not increase in rpms.

I guess you will find more interesting things tomorrow? If there is gasoline in the fp regulator vacuum line?? If you get a chance to bring it inside over night, unplug the vacuum line next to the FPR and see if any gasoline comes out in the morning. It will only leak when cold, not hot. At least, that was my experience with my 1995.

This HAS to be something simple,

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

This morning before trying to start the car I measured the temperature sensor resistance. It is ~21 degrees F (-6C) and the resistance measured was 9400 ohms. Looks about right. (did a quick graph and this fits the curve)

I pulled the vacuum line from the fuel PR and did not find any fuel or fumes. Next I took off the line from the intake to the EVAP valve. Once again found no fuel or fumes and the valve was closed.

Then changed my starting routine. This time I switched off the power to the fuel pump before trying to start the car. The engine turned over but did not fire. I tried twice. Then I switched on the toggle switch (power to the fuel pump fuse #2) and tried to start the car. It fired up immediately - with no hesitation -- and this is with an outside temperature of ~ 21F!

Analysis -
It appears that there was little if any fuel or pressure in the rail otherwise it should have at least fired during the initial attempt try to start the car.
I assume this also shows that no other fuel path, flooding the car, was present prior to starting the car - i.e. leak down during the night.
This leads me to conclude that when it is cold the ECM must be giving the wrong signal / injection period when it is cold.

The ALLDATA info on fuel injection - shows that the ECM receives input from the Engine Speed Sensor, the Camshaft Position sensor and the engine temperature sensor. and During the start-up phase, the ECM computes the injection period on the basis of the engine temperature and speed signals.

Yet I know that the temperature sensor is giving the right signal from resistance measurement and from the fact the ECM is sending the right data to the dash temperature gauge.

As per the Engine Speed Sensor, ALLDATA states that "the engine will stop if the signals from the sensor to th engine control module (ECM) cease." Yet the ECM gives a proper signal to the dash tach and the car does start if I go through the F/P fuse routine. Hence the sensor must be working.

Therefore, this leads me back to the MAF. ALLDATA states that "The Mas Air Flow sensor measures the engine intake air mass and supplies the ECM with continuous signals, which the module uses to compute - Injection period - ignition timing - Turbo Charger bost and Post-running of the cooling fan." and "The ECM will adopt a substitute value if the MAF sensor signal is lost or becomes faulty."

So if the MAF is kid of working but giving the wrong signal upon start up (if this is even possible because the car runs great other than upon cold start up) then it might default to a substitute value - (probably for a warm climate) - Of course if it does send a signal for a warmer condition it should command less fuel not more ... but maybe the timing and injection combination does not work and by continuously trying to start the car it keeps injecting fuel until it is flooded. hmmmm.

Sure would be nice to know if the resistance between terminals #1 & #4 of the MAF must be approx. 110 ohms for the MAF. IMPORTANT NOTE: I just re:measured the resistance across terminals #4 and # 1 on the MAF sensor (with the plug disconnected) and found out it is actually 23900 ohms (23.9K) and not ~ 750 ohms. (My feeble excuse : Hoping to find a resistance nearer to 110 ohms I had used the 2000 ohm setting on my DVM which apparently also serves as the setting for a diode check. When the leads are reversed + to -ve there is no reading. Using the 2000K ohm setting and higher gives a consistent 23.9 K reading.)

This is likely significant given the spec calls for 110 ohms.

Maybe someone who has experience with MAF's or has a similar car can advise or if at all possible measure the resistance across these terminals to help confirm what is normal. I am confused since my car starts when it is warm idles even when it is cold and runs great other than this early morning cold start issue.

Cheers,

David

Update:

Armed with what I though might be good intelligence with respect to the MAF sensor I took a trip to a Volvo Scrap yard. There, I found a used MAF sensor and measured the resistance between which - showed 29K. So I guess unless the 2nd MAF sensor is also bad it is not the MAF sensor. I wonder why the spec for the MAF is rated at 110 ohms ...? Oh well.........

I wonder if the compute is holding old codes that impact starting?
I did successfully clear all the codes using my OBD2 reader - at least it said so. However, on my INNOVA 3110 tool the CAT, EVA, O2S and HTR icons were flashing indicating the vehicle supports the monitor but the monitor has not run the test yet. From from the ALLDATA info OBD II Readiness Codes: Revised Reset Instructions - it implies there is a whole procedure requiring an extended driving period in order to reset readiness codes from INCOMPLETE to COMPLETE. Yet I read somewhere this may not be possible for the OBD II reader. Any thoughts?

Cheers,

David

1996-7 ECUs have a terrible time going through the checklist and setting readiness codes. These codes go blank when the battery is disconnected and then show up as "errors" when people bring their cars in for emmissions testing facilities. There are no drivability problems during the ECU learning period. California test stations have learned to ignore these "errors".

I still think you have excess gasoline in your intake manifold before cold start. When you tried to crank the engine without the fuel pump running, it may have purged the extra vapors.

=== thought === have you taken a look at the throttle body lately? Could it be dirty and the idle air control hole partially blocked? Meaning no air getting to the manifold. Try a "cold" start with the gas peddle partially depressed to open the butterfly.

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Klaus,

I just took my car for a 50 minute - 7 stop road trip in the country following the OBDII revised rest instructions in ALLDATA (i.e. accelerating maintaining speed and idling - 7 times according the the instructions for the M4.3 In the end the OBD2 reader was satisfied and I found no codes and no flashing icons. Just in case I cleared them three times in a row. Superstitious behavior but --- you can never be sure. So it now appears as far as the OBD2 system everything is a-ok.

At this stage I will try anything. Sure would be nice if it were the idle control hole. I have to admit I have not even looked at the throttle body. I will try a cold start with the gas peddle partially depressed - I assume just enough to open the butterfly - and see what happens.

Thanks again for the suggestion.

David

Dec 07 update:

This morning (23 degrees F) I tried a cold start with the gas peddle partially depressed. The car sputtered a few times but did not start. It seemed different but not significant. In any case I tried a second time with the peddle slightly more depressed to no avail. I then switched off the toggle switch and the car fired up quickly and I immediately turn the toggle switch on and the car continued run / idle fine.

Oh well. When I get a moment I am going to inspect the throttle body anyway. Cannot hurt.

Cheers,

David

LOL. You are getting closer! Remember the old addage: gasoline, spark, and air. You know you have gasoline and spark, but air is questionable. If you have more gasoline than air, sputter sputter.
I remember with my 164, long may she live, when it was running a little strange. The intake manifold was filled with liquid gasoline. It is amazing how much gasoline is injected during cold start, especially when I had a broken intake valve spring :)
Your throttle body probably needs cleaning anyway. Get a new gasket, you will need it, the old ones are a little difficult to take off and tend to tear. Remember not to use carb cleaner - too harsh. The auto supply stores have very good cleaners specifically for TBs.
Good luck, it is getting cold here, so you don't have much warm weather left.

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Klaus,

Good news. This morning (30F a little warmer) I tried starting the car with my foot a little more depressed on the peddle. After a few cranks it fired right up. So I am now cleaning the throttle body checking hoses and adjustments etc.

Hopefully it will be a little colder tomorrow to test this but in any case the diagnosis is looking better!

When I disassembled the TB, however, a couple of questions came to mind. When the throttle body is closed it looks like no air is supplied to the combustion process through it. Therefore, I assume combustion air for a cold start must rely on what is in the intake manifold along with any provided through the "Idle Control Valve" ... yet the documentation for that ICV does not refer to start up as one of its functions. When I started the car with my food on the peddle this morning additional air was provided but where should this air come from on a cold start normally?


Thanks again for your help

David

The AIC motor is always on when the key is in posII or III. That little hole supplies enough air to do the job and a little dirt is enough to slow down the supply. Poor design, as the hole should be in the top of the manifold, not the bottom where all of the gunk gets deposited.
My 1995 is also in need of a cleaning, but that will have to wait until spring. My garage floor already has ice on it from the smow melting in the wheel wells.
I still like the way you switched the fuel relay :)

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Klaus,

While cleaning the TB on my 96 850 Turbo I did not find a hole int TB connected to the AIC. However, from your description I followed the AIC output hose and it dumps air via ~3/4" dia hose directly into the intake manifold via the top of a vacuum port/tree just beside the throttle pulley and cable. It looks like Volvo agreed with your assessment and moved the air inlet to the top of the manifold!

I inspected the AIC inlet and outlet hoses for blockage. I found there was a film of oil on the inside surfaces of the inlet hoses right back to where they enter the top of the rad ... I assume as part of the inter-cooler piping. The insides of the AIC itself looked pretty dirty as well. I am speculating here but --- it is possible when it is very cold outside the dirt build up & oil film does not let the AIC valve open or otherwise allow enough air in for a cold start. So carefully using the throttle body cleaner and air I cleaned up the valve - taking care not to let fluid seep into the motor - a lot of dirt did come out and the valve seemed to move much more freely. I also checked the resistance specs across the terminals and they were withing spec.

The car started right up after the fix -- about 5 hours in the cold ... but this is not a good test. Tomorrow I will try to start the car as if it were normal!

Cheers,

David

Sunday update:

After letting the car sit outside for 10 hours lat night I tried starting the car normally. Although it tried to start after a few cranks (best effort so far) it settled back into its old routine. I held the peddle down slightly but it was too late. Then I flipped the toggle switch - it fired right up - and I was able to switch it back on quick enough to keep it going. Of course it then ran perfectly normal.

Interestingly late last evening after letting the car sit for 7 1/2 hours outside car and it fired right. This was the second attempt following the successful 5 hour wait. For some reason it takes quite a long time sitting in the cold to demonstrate this problem.

It may just be superstitious behavior but am going to clean the AIC again just in case there is still some residual oil still in it hampering is cold performance.

Oh well back to the drawing board. Maybe I will replace toggle switch with a push button and mount it on the dash! He he

Cheers,

David

Some degree of success.

After researching on the internet problems with AIC valves and cold start issues on Sunday I made two more attempts cleaning the valve. After the fist attempt I let the car sit for 9 hours and although it fired up it would not continue to run. I then cleaned the valve once again only this time aggressively and sprayed WD40 into it to lubricate the parts (I read this on another forum)

This morning (28F) the car turned over a few times fired and continued running without the help of the toggle switch. Very positive indeed. Hope this continues the next few days. Only time will tell now. In any case best results yet.

Cheers,

David

For the 2nd day in a row the car has started without turning off power to the fuel pump relay. (Outside temperature was 26F - although now wind so felt warmer)

In any case it appears the mysterious "cold start problem" requiring disconnecting power to the fuel pump for a few cranks then reconnecting it was caused by a malfunctioning Idle Air Control valve. Cleaning and oiling it has made the difference.

Thanks all for your input and advice.

Cheers

David

Whew!! I am glad that worked. Beats throwing parts at it, like any profitable dealer would.
That just means that my 850 will get a cleaning as soon as it warms up again around here, probably May.

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

Fun problem...
When the fuel pressure regulator fails, it could allow fuel to get through its vacuum line and pour excess fuel into the intake manifold. That will result in an over rich condition at start up.
You shouldn't need to pull the fuse, pushing the gas pedal to the floor while starting also turns off the gas flow.

Klaus
--
The 164 has a new home, all I am left with are a 95 854T and a 98 V70R :)

KlausC

Thanks for the suggestion. You are the second to suggest it might be a failed fuel pressure regulator. I will test this asap and let you know what I find out.

David

Hi David,
So to date you have not replaced the fuel pump relay?
When the car wont start, can you still hear the fuel pump working
in the tank for a few seconds right after you turn off the ignition?
I still think that is the most likely culprit since all indicators
point to an electrical/fuel delivery problem.
My 94 850 would not start after a breif shut down period in warm weather.
Could be that yours wont start after a prolonged shut down in cold weather.
?
Armie
--
Luck favours the prepared.

NBZ

Thanks for your input.

I have not replaced the fuel pump relay yet.
I know the pump works and pressurizes the fuel rail (I can hear it working)and I have measured the fuel pressure after just turning the key on. My analysis (right or wrong) is that the car is flooding in the morning when I first try to start it normally and by removing the fuel pump fuse temporarily it allows the engine to eject the fuel and start. Which happens quite quickly after removing the fuse. I am certainly not an expert but it seems to me that even if the fuel rail is pressurized the injection sequence controlled by the computer should not allow the engine to flood. Which seems to imply the computer is seeing false signal and giving the wrong sequence. Hence my efforts to measure all the sensors that impact the injection timing. Yet by simply removing the fuse the whole timing issue gets resolved?
I have reviewed the wiring diagram and removing the fuse only deprives the fuel pump of current when the computer calls for it. It is hard to understand why this action resolves anything but starving the fuel rail and injectors of fuel, yet why does the computer push more fuel into the cylinder?

Maybe the lack of current to the pump inadvertently resets the computer and impacts the injection sequence. It seems to me this is quite a long shot however I have virtually exhausted my options.... So I have bought a relay from the dealer and will install it tomorrow morning if the car does not start after all the efforts put in today.

I will post the results tomorrow.

Thanks again

David

Good luck David,
My thinking is that if the relay isn't closing properly
then that could be what is causing the cylinders to flood.
Armie
--
Luck favours the prepared.