Mike, run that pressure test before you dig.
My split hose:

My fix:


If this is causing your trouble, look for SAE30R10 submersible fuel hose. 5/16" or 8mm.i opted to use a U-shaped tube than letting the hose bend.

 

Well just a little update on my issue.....it seems to be getting worse. This morning I turned the key on and off two times to give the fuel pump two chances to prime the system......that's be working for about a week. Well, this morning that wasn't enough. It still fired up and then died.....I had to prime it a third time, and then it fired up and ran great.

The part that confuses me is that other than the starting issue....the thing runs strong. I can run it up through the gear under full throttle and it doesn't skip a beat. If the fuel pump was so weak that it can't even prime the fuel system with three tries......why would it be strong enough to supply enough fuel for wide open throttle running? And it still seems to only be an issue if the car sits for more than 4-5 hours or so....if it sits for anything less, it still fires up on the first try. Very strange.

I picked up a fuel pressure test kit at Harbor Freight last night......I guess this weekend I will be doing some trouble shooting.

 

Well, mine went thru what you're having now then gradually worsened to needing multiple priming & yoyo. The worst it got before I found the culprit, the car died on me a coupla times. That got me really stumped. Don't wait for it.

 

Well I just pulled the IC off and hooked up the fuel pressure gauge. When I turn the key on and the pump primes, the gauge jumps right up to 51psi. When I turn the key off the pressure seems to be holding. After the car sat for about 15min the pressure was down to around 49 psi.

It sounds like the fuel system is good to go. Now what to check? Could it be a MAP sensor? I'm a little stumped ...I was expecting to find a bad fuel pump.

 

Tried the purge/breather valve yet (mentioned earlier)?

 

No, I guess that will be the next step. Could someone explain to me what the purge/breather valve actually does...and how it works?

 

That should be good news that your pump is good!
Aside from the vent valve, try a new Coolant Temp sensor. It's near the T-stat.
A failing or bad Coolant Temp sensor will feed the ECU erroneous readings. You might be starting lean when you're supposed to go rich or vice versa resulting in hard starts. This cheap part was among those I swapped out. Mine turned out to still be good so I got into the other suspects 1by1.

 

Other causes are the ignition system. Plugs, wires, coil pack. Go OE w/ all these if you decide to do refreshing.
You can do easiest or cheapest 1st. The breather valve is pricier than the Coolant Temp sensor, leads & plugs.
Hope you hunt it down soon.

 

Fuel vapour is created in the fuel tank when the engine is not running. The EVAP system stores this fuel vapour in a charcoal canister.
– when engine is cruising: purge control valve opened, small amount of fuel vapour is drawn into the intake manifold and burned.
– when the engine is idling (starting cold): purge valve prevents any vapours from entering the intake manifold and causing excessively-rich fuel mixture.

A faulty EVAP system will only affect engine driveability when the engine is warm.

 

The plugs and wires are brand new...only about 2000 miles on them. Whatever it is...it's related to the temp of the engine. After its totally cooled after sitting for 4+ hours it takes a few tries to start. If its already warmed up, or only been sitting for an hour or so it fires right up.

I'm thinking that my next step will be the coolant temp sensor. That makes sense... I can understand why that would be causing the problem. Is it pretty easy to get to the sensor?

The fuel breather valve doesn't make much sense to me...so I'm going to leave that alone for now.

One quick question on the coolant temp sensor....if it was bad and giving strange readings, wouldn't the temp gauge in the car be acting funny also?

 

Do you have a scangauge or anything alike? It'll tell coolant temp from the obd unlike our temp gauge that constantly sits at the middle. This will help you check the sensor's health. If it says cold or hot when it's supposed to be the opposite, you'd know its bad.
Again, be glad you won't have to dig into the pump.

 

I have a copy of the BMW diagnostic software on my laptop....and an OBD2 cable. That's a great idea! I will hook it up tomorrow morning after the car has sat all night and see what the coolant temp is saying when I hook up the lap top.

The only thing that I can't figure out is that when I try to start it for the first time and it dies, it doesn't run long enough to heat anything up at all....so why does it start on the second try. And why does it start when I turn the key on and off a few times....what would that have to do with the coolant sensor?

Very strange...but I guess I will know more in the morning when I do some more troubleshooting.

 

In this electronic age, nothing is simple anymore. In your case, I'm guessing 2nd starts run the engine instantly as fuel has already been fed into the injectors from the initial start.

 

But when I first turn the key the fuel pump will prime and run the pressure up to 50psi before I even try to start it. With that much pressure, those injectors should already have plenty of fuel at their disposal before they even need to open.

Getting ready to hook up the laptop this morning and see what other kinds of info I can collect.

 

Let's see what you'd find thru the obd.

 

Well I finally got a chance to hook up to my laptop and run INPA to check what the coolant temp sensor was reading. The outside air temp was 73 degrees, and the coolant temp was reading 36.75 Celsius (98 degrees Fahrenheit) before I tried to start the car......it was sitting all night. So it does seem that the coolant sensor is a little off.....but is that enough to make the engine not was the start?

When I finally did try to start the motor, I turned the key and the fuel pressure shot right up to 51psi instantly....and when I tried to start the car, it cranked over, fired right up, sputtered, and then died again. On the second try....it fired right up and idled nice and smooth.

I went ahead and reset all of the adaptations thinking that might help.

So far I'm still stumped. The fuel system is obviously good to go.......but I'm not sure about that temp sensor. It's reading about 25 degrees to hot.....but I don't know if that's a big deal or not.

I haven't tried cleaning the MAP sensor yet, or the TMAP sensor...but both are pretty easy to get to.

Any ideas...I'm at a loss. The problem seems to be getting worse...I'm surprised I'm not getting a CEL.

 

What's your operating temp reading at? Start up temp seems off. Might cause ECU to lean out the initial start up AFR a bit.

 

Just to have some benchmark comparison, my scanguage temp reading hovers around 195-205F (90-96C) w/ AC & hi-speed fan on even in city or heavy traffic. My 1st stage fan resistor has been bypassed & I've the Canton coolant reservoir. How do your readings compare?
How many miles on your car now?

 

I know you said you're skeptical, but the breather valve is easy and cheap to replace:
http://www.pelicanparts.com/techarti...eplacement.htm (figures 2 to 6)
https://www.northamericanmotoring.co...475-post9.html

Is it worse starting up right after you fill up with gas (even with a warm engine)? That was exactly my situation.

 

^That is among the suspects & a very good find @ 1/4 the price!
But, before you go buying parts, check if Vac line 13 is connected & doesn't have cracks. Clearer & better accessed w/ IC off. It's on the passenger side of the fuel rail.

 

I did check that hose (#13), and it's good with no cracks. I checked all the other hoses under the IC, they all looked good also.

Funny thing...this morning it fired right up on the first try. Strange how the problem is so hit and miss. I'm still not getting any CEL's.....and this morning I filled the car up at the gas station, and it fired right up.

It's never had a problem unless it's set for more than 4hrs. I'm just trying to work though my mind what it would be....something related to how long it sites, and the temp of the motor. Very strange......I guess I'll keep trying to trouble shoot. It would be much easier if the problem happened all the time.

 

Saw the responses you got over at MT. Most simply pointed you to a failing pump. I won't say they're wrong but our VDO units don't go as easy as those in the 56.
Your symptoms, being intermittent aren't easy to hunt down. Remote control diagnosis makes it a bit more of a challenge.

Here's another freebie check you might wanna do:
Go back to the diagram, part 7 is the rail pressure/pulse regulator. To check its health, hook up your pressure gauge again. Remount your IC & start your engine. At idle, you should see about 42psi instead of 52 at priming. If it doesn't do this, say it stays at 52, the regulator diaphram isn't working as it should or you have a vacuum leak. This trick is courtesy of the legendary Maestro, k-huevo, who used to be here a lot in the old days.

It would have been easier had you a scangauge to check how your temp swings are. The coolant temp sensor is suspect, but I'm not definite about it.

 

Thanks for the info on the rail pressure regulator. I actually tested that when I had the gauge hooked up before. When I prime the pump the pressure is at 52psi, when I start the motor and it's running the pressure drops down to around 42-43 psi.....when I rev up the motor a little the pressure climbs.

So I guess the pressure regulator is working properly.

 

Yeah, that rules out the regulator & vacuum leak. You've eliminated a number of possible causes now. Usually w/ a bad coilpack, you'd get misfires & they pop codes. Seems like your remaining suspects now are the $12 coolant temp sensor & the $22 identical purge valve from an 06 Saab93. They're both easy swaps if you decide to. And, why not try some techron?

 

Well, this has been the second morning in a row that the car has fired right up on the first try......strange. I didn't really do anything at all to it, so I'm not sure why the problem has gone away.

I think my next step (like you said) is going to be the coolant sensor, and then that valve. Thanks for the info on the Saab valve...much cheaper!!