Fuel pressure q's...

[paanta]

'91 535i.

I'm trying to diagnose a problem where the car is hitting some sort of limit (feels like the rev limiter) at about 5600 to 5800 rpm (varies from day to day). It's also having some idle funkiness. Cold or warm starts, for the first minute or so it's fine, then idle will dip and any throttle input is bogged down and there's no power. It'll clear up after 30 seconds or so.

So far I've replaced the CPS and fuel pressure regulator.

I don't see any signs of vacuum leaks.

Hooked a voltmeter to the O2 sensor and it hovers between ~.3 and .85 volts, though it does go over 1.2v for a while after startup. Bad O2 sensor?

Checking the fuel pressure right now...

The pressure with the FP running and the car off is almost exactly 3 bar. When I cut the power to the pump, it drops down to about 2.8 bar and will stay there. Crimping the return hose with the pump on, pressure goes up to 4.5 bar.

Bentley doesn't have a test procedure for pressure in a running car, but with the car running at idle, it's about 2.5 bar. Blipping the throttle takes it up to 3. Normal? See: http://qik.com/video/1193891

30 second fuel flow, measured immediately after the (brand new) fuel pressure regulator (obviously with the car off) was .90 quarts. Bentley calls for .93 quarts. Is this enough to be a problem by the time I get to WOT and 5500+ RPM?

[ahlem]

How old is your fuel filter? How old is your O2 sensor? I believe 50,000 miles means time to change it. Does your check engine light come on at all? I went through the dying fuel pump issue a while back and had a rev limiter issue. I didn't get it to those revs very often. You may want to check your cap and rotor and the quality of the spark. If you get an air constriction of some sort it can act like that. At idle, you can pull the dipstick up an inch or so and see if there's a difference. You may have multiple issues and need new o rings on the dipstick plus something else. How often do you adjust your valves?

[Jeff N.]

Can you clarify what you mean when you say it clears up at 30 seconds or so?

[paanta]

Can you clarify what you mean when you say it clears up at 30 seconds or so?

After 30 seconds of running really poorly with zero power (as in 15 mph top speed), it will suddenly return to normal and run flawlessly.

My gut tells me it's not a fuel problem. The cutoff at high RPM is instant. Any time I've had fuel problems in the past, the problem has always been a slow dying of power at high RPM. I was testing fuel pressure/flow just to be sure gas wasn't the issue...but then the .9 instead of .93 quarts flow pops up. Is that bad enough to make a difference in how the car runs? Bentley doesn't give a range of acceptable flow rates.

Fuel filter is new. Valves are adjusted, I've ruled out vacuum leaks both with carb cleaner and because inducing just a tiny leak gives me an immediate crapo idle.

O2 sensor is of unknown age. The voltage on the sensor fluctuates normally (ie, between ~.3 and .85 volts) and very quickly. Creating a vacuum leak or clamping the fuel return line makes the sensor read lean or rich, as you'd expect. The odd part is the 1.2 volts reading that pops up just after startup. I didn't think O2 sensors could even read that high. I've got a new one on order, but I'd like to rule out other culprits before installing it. Fouling a new one with carbon from a poorly running car would be dumb.

I do get an error 1222 from the DME and a CEL that comes on periodically. That's 'lamda control out of range'. So either the O2 sensor is bad or, something else is throwing the mixture off.

Thanks for the help.

[Jeff N.]

Strange. I don't think it's your fuel flow.

Your fuel pressure does seem low. Mine - at the rail is about 40 to 42 PSI at idle raising to about 48+ when you increase engine speed. 2.5 bar is ...what... about 35?

But, I wouldn't think that would lead to the instant cutout you describe. Low fuel / lean will cause poor throttle tip in.

I'm wondering if you might not have two separate problems. Maybe low FP via a faulty FPR that's causing the low RPM problem and something else going on with the high RPM.

Strange.

[paanta]

FPR is brand new Sorensen (a brand that seems well trusted among the Volvo + Saab folks). I replaced the old one because of a slight gassy smell in the vacuum line, but it didn't change anything.

36 PSI at idle up to 44 PSI at tip in, then back down to 36 psi once it stabilizes at a higher engine speed. It seems low, but the FPR does limit it to precisely 3 bar with the fuel pump relay jumped and the car off, rising to 4.5 with the return line pinched, which is what the Bentley spec says. So I think the FPR is ok.

If the FPR were bad, I'd expect it to run poorly more consistently, not just for a few seconds after startup. Even running poorly at idle, it has a distinctly electrical problem-esque thing going on...if you can get the engine over about 2500 rpm, it'll clear its throat and suddenly run fine until the engine speed drops back down to near idle. And yes, I've tested the throttle position sensor. And the alternator+battery.

I could start throwing parts at the problem, but I've been there, done that. No desire to blow through $500 trying to track what's hopefully a simple issue.

I keep coming back to thinking it might be a problem with the ECU itself or maybe a bad coil, though neither one can be tested real easily. Maybe the AFM pot is dirty? That's one thing I haven't tested yet. Weird values could throw the mixture off and give me that error 1222 I suppose. I'll check that tonight...

[angusbn]

Faulty/Dirty Idle Control Valve?

[Bill R.]

vane on the AFM is binding or sticking shut... so that you have to create quite a bit of manifold vacum before it pulls open... Pull it off the car and let it sit overnight then push the vane open by hand when its cold and see if its stuck.





FPR is brand new Sorensen (a brand that seems well trusted among the Volvo + Saab folks). I replaced the old one because of a slight gassy smell in the vacuum line, but it didn't change anything.

36 PSI at idle up to 44 PSI at tip in, then back down to 36 psi once it stabilizes at a higher engine speed. It seems low, but the FPR does limit it to precisely 3 bar with the fuel pump relay jumped and the car off, rising to 4.5 with the return line pinched, which is what the Bentley spec says. So I think the FPR is ok.

If the FPR were bad, I'd expect it to run poorly more consistently, not just for a few seconds after startup. Even running poorly at idle, it has a distinctly electrical problem-esque thing going on...if you can get the engine over about 2500 rpm, it'll clear its throat and suddenly run fine until the engine speed drops back down to near idle. And yes, I've tested the throttle position sensor. And the alternator+battery.

I could start throwing parts at the problem, but I've been there, done that. No desire to blow through $500 trying to track what's hopefully a simple issue.

I keep coming back to thinking it might be a problem with the ECU itself or maybe a bad coil, though neither one can be tested real easily. Maybe the AFM pot is dirty? That's one thing I haven't tested yet. Weird values could throw the mixture off and give me that error 1222 I suppose. I'll check that tonight...

[ahlem]

If you exhaust all the other options, you can pull the fuel pump and check for bronze metal shavings associated with a fuel pump bearing failure in the bottom of the fuel tank. There are two tabs that hold the plastic ring that holds the fuel pump. You squeeze the tabs together and the pump and ring come out. I noticed that when I changed my fuel pump. I started getting hesitation at serious acceleration and it got progressively worse and the check engine light started coming on until I feathered the throttle. It was a bit hairy a couple times on freeway on-ramps. Checking the fuel pump like this is mostly time consuming until you have to buy one at $285.

[paanta]

AFM flap was definitely sticking a touch. It might be better now after a LOT of t-body cleaner. Before it'd kind of 'click' when it was pushed open, now it's smooth. I think it cleaned up the idle quite a bit. Definitely worth $4 in cleaner and 10 minutes. Took the time to clean the contacts inside the AFM as well.

Didn't help the high-rpm cutout at all, though. I've heard the AFM doesn't do anything at high rpm. Is that true?

Also, I noticed while reading the resistance values that they go from ~500 ohm (or whatever the scale was) to around 1000 ohm but then drop back down to 800 or so when I push the flap past about half way. Is that normal?

[shogun]

http://paulaxford.com/bmw730i/afm_service.htm

[paanta]

I cleaned the potentiometer track, which smoothed out the resistance values as I moved the flap, but I didn't see info on what the correct values should be. Flap closed, I got ~500 ohm, rising to 1000 with the flap mostly open, but then dropping off to ~800 as it got even farther open. Is that normal? I don't have a sense for how the numbers should look. Everything I've read just says they should fluctuate "smoothly".

I'm also a little fuzzy on my DME 1.3 theory. Does the car totally ignore the O2 sensor and AFM at WOT, as it does on older versions of Motronic? If it does, I know for sure that the AFM isn't the cause of my high-rpm cutout problem....

[Jeff N.]

IIR, you need to test the AFM with the +v/ground connected otherwise you can see strange values. Haven't checked Bentley for a while - does it have a procedure to check the AFM?

At WOT, the car does go open loop and ignore the sender. I'm less clear on how it treats the AFM reading.

I know on my porsche 951 (similar era motronic), it reads the AFM all the time regardless of throttle setting. Not sure you can interpolate across or not.