Odd intermittent fuel problem

[TiagE39]

Hi all. Got a bit of a confusing one (for me, anyways) on my '99 528i.

On occasion, I will go out to drive my car after it is sitting for a few days (2.5 or more) and it will run for about 30 seconds and then die. Subsequent restart attempts will catch but die shortly after, as one would observe running out of gas or a bad/dead fuel pump. However, I still have slightly less than a 1/4 tank, and the fuel pump is new (less than a year old). The passenger side sender unit and fuel filter are also new (within last year). Refilling the tank with a gallon of gas clears up the issue.

I am well aware that the above symptoms are describing a bad or marginal suction jet pump/transfer pump. However, there's some additional catches that make me wary of that being the case:

• I have observed the fuel circuit running with the passenger side sender removed so I can see the fuel pump and jet pump outlet. It's a strong flow even with the passenger tank lobe close to empty. No bubbles.
• Driving the car normally below 1/4 tank even to the light turning on doesn't kill the engine.
• I have journaled the fuel sender readings by INPA and by the cluster unlock and the readings align with a working jet pump (driver sender reports equal to or less than the passenger sender). I will be journaling this again as a precaution.
• Most importantly and the subject of my post: my parking space is laterally tilted about 15* so that the passenger side of the tank is higher than the driver's side.

I'm emphasizing my last point as I feel like I've stumbled onto a a fringe set of circumstances that don't agree with the E39's fuel system. My thinking is that low fuel level + parking space tilt + sitting a few days (so that the fuel loop loses a little pressure, although my car always starts strong) creates a situation where the fuel pump is a little too dry and then subsequently lets bubbles/cavitation into the fuel line. This prevents the jet pump from priming properly, so the return fuel is also cavitated, which creates a positive feedback loop until the engine dies from a slump in fuel pressure.

I will add that I can leave my car in my storage garage for weeks with less than a 1/4 tank and it will start fine on the first try. Two days in my tilted parking space and too little gas however and it fuel starves.

What does everyone think? I suppose I could "fix" this by not parking with below 1/4 tank, but I feel like I shouldn't be having this issue (the germans thought of everything, surely including laterally tilted parking spaces). Could the jet pump be failing marginally? What about my (still original) driver's side fuel sender?

[edjack]

When you say "jet pump," I assume you mean the transfer pump on the left side of the tank. This is a passive device that relys on ventury action; it has no moving parts.

I do not see how the "jet pump" is the problem.

When you first start the car, do you always pause in position 2 before starting? This should pressurize the fuel delivery system.

[StephenVA]

Easy first step
Install a fuel pressure gauge on the fuel rail
Read running pressure
Turn off engine READ pressure, any drop over time?
Before starting engine after "sitting for a few days (2.5 or more)" What are the readings?

You need two things to keep injectors happy, Volume and pressure. Any leak down in the system would give you your results.

Normal list of leak down issues
Injectors dirty
Regulator/ filter pressure leaking
Vapor return lines/fuel tank issues.

[StephenVA]

When you say "jet pump," I assume you mean the transfer pump on the left side of the tank. This is a passive device that relys on ventury action; it has no moving parts.

I do not see how the "jet pump" is the problem.

When you first start the car, do you always pause in position 2 before starting? This should pressurize the fuel delivery system.

Ding Ding! and count to 1,000, 1001...

Most drivers never do, they just crank and tell you their car is always hard to start.

[TiagE39]

I always wait for the fuel pump priming to end before starting the engine.

[sleuth255]

Try backing into your parking place so the low side of the fuel tank is reversed. Problem gone?

[Aussie528iT]

There is also a 2nd jet pump associated with the fuel pump surge pot. If the fuel pump and its holder are not fully clipped into the surge pot base this 2nd jet pump doesn't work properly to keep the surge pot full once the fuel level in the RHS tank half is below the top of the surge pot. Putting a gallon of petrol in the tank raises the fuel level in the RHS half of the fuel tank above the top of the surge pot and fuel can then fill the surge pot and allows the fuel pump to operate properly.
How do I know?
Been there and done that when fitting a new fuel pump and then found out later it was a faulty fuel pump relay, after getting back into the RHS tank, re-seating the fuel pump assembly properly and then finding the fuel pump relay was faulty - one of the green relays behind the boot/load area RH trim panel.
Hope this helps.

[TiagE39]

Try backing into your parking place so the low side of the fuel tank is reversed. Problem gone?

Haha, I thought that too, but I'd like to figure out the underlying issue.

There is also a 2nd jet pump associated with the fuel pump surge pot. If the fuel pump and its holder are not fully clipped into the surge pot base this 2nd jet pump doesn't work properly to keep the surge pot full once the fuel level in the RHS tank half is below the top of the surge pot. Putting a gallon of petrol in the tank raises the fuel level in the RHS half of the fuel tank above the top of the surge pot and fuel can then fill the surge pot and allows the fuel pump to operate properly.

Okay, this is interesting and sounds like it might be my problem. I did replace my fuel pump (i've had it out twice, actually) so it's possible it's not installed 100%. So what you're saying (to check my understanding here) is that the fuel pump may not be completely into the black "cage" that it's installed into?

[wagons ho]

I would like everyone to consider a different scenario.
Let's say the car is fine, meaning no pump install problem or bad parts.
528 still has a return fuel line correct?
The poor quality gas with alcohol we get now is expanding in the fuel line, while parked.
On restart, the regulator has to burp the air out, to have a completely filled fuel line.
Obviously, any air(fuel void) in the fuel line would cause it to die.
Problem doesn't happen except when cold? Correct?

- - - Updated - - -

Especially if turning to position 2 and filling the fuel line fixes the problem.

[StephenVA]

Pull a sample of the fuel and do a simple alcohol/gas mix test

if it is over 15% there is going to be an issue.

Pullled many only to find water in the fuel

[seagreen323i]

I'll give you a scenario. The previous owners of my father's 530i broke the mounting bracket off the fuel tank. We bought it as a non runner, literally threw gas in it and drove it away. All for 2k for a 100k mile 530i sport.

Basically the pump would randomly tip over and suck air instead of fuel.

Sent from my SM-S901U using Tapatalk

[Aussie528iT]

Haha, I thought that too, but I'd like to figure out the underlying issue.



Okay, this is interesting and sounds like it might be my problem. I did replace my fuel pump (i've had it out twice, actually) so it's possible it's not installed 100%. So what you're saying (to check my understanding here) is that the fuel pump may not be completely into the black "cage" that it's installed into?

That's what happened in my case. What I thought was a faulty pump turned out to be a faulty pump relay and I compounded the problem by not reinstalling the pump properly.

IIRC (it was several years ago) there are 2 black plastic clips that you have to press that release the pump assembly from the surge pot and when I put the pump assembly back in one of the clips did not engage properly (pilot error).

A bit more info on the pump relay. I tested the voltage at the pump and it was 12V+, but after the pump was operating for a while the relay contacts were overheating and the pump stopped operating.
The outside green plastic cover was distorted showing signs of over heating.
On my touring there are a couple of other green relays behiind the trim panel so a short term fix was to swap the green pump relay with one of the other relays one of which supplies the cancer stick lighter.

[sleuth255]

Haha, I thought that too, but I'd like to figure out the underlying issue.

Exactly. Its a simple diagnostic test to see if your driveway angle is related to the problem: If the issue goes away then this hypothesis is supported and the problem may be an engineering defect or (more likely) an issue with the transfer pump.

Also, responding to the "let it pressurize" comments, just saying that if one has to let the fuel system pressurize before trying to cold start then one has a fuel system problem. I have always been able to immediately start the car on ~the second engine turnover in any condition.

[R Shaffner]

I glanced at this thread a few days ago, and I'm real glad I did. It saved me a lot of hassle yesterday.

I was driving down the Interstate on a short trip when the engine suddenly stumbled as I gave it throttle. I moved over to the right lane as it stumbled more, and ended up driving slowly on the shoulder with the flashers on until I got to an exit ramp. I made it up the ramp to a shopping area right as it died and wouldn't restart. It was like it had run out of gas, but the gauge read just under 1/4 full.

I carry OBD Fusion in the car, and saw codes for too little fuel. I did the normal things. No sign of an intake leak. It was too noisy there to hear whether the pump was running with the key on. So I checked the fuses. Tested the fuel pump relay. I even jumped the relay. Didn't help. I put a screwdriver tip to the valve on the fuel rail...and just a little air came out! No gas. Very odd.

I was beginning to think I had a bad fuel pump and would need a tow, when I remembered this thread. My gauge read 1/4 full -- that rang a bell. So I walked to a gas station, bought a gas can, put 2 gallons in the tank, and it started up!

Now I know to keep the tank mostly full until I figure out the real cause. Thanks to all for sharing this information.