Running rich or lean?

[mrgraphics]

I've got this problem that keeps rearing it's ugly head - 1986 325es

Symptom: Driving on the highway, I'll lose power when going up an incline hill trying to maintain 70. It'll start bucking, but if I let off the gas it'll start running ok, albeit with less power and it slows down. On a decline I get power again. It's almost like if I keep pressing down on the gas it is flooding?

Other symptom - Then starting out from a dead stop, it'll sort of lose power, semi stall, then pick right up with a surge.

I've fiddled with the MAF, and the semi-stalling from stop issue would get less pronounced if I opened up (counter clockwise) the air bypass screw on the MAF.

Then if I move the needle (un-doing the screw that holds the arm on the MAF) and position it further up the track, things smooth out. An unmodified MAF runs like crap.

Idle is fine.

New plugs, new injectors (flow-matched), rebuilt engine, new cap and rotor, new CPS and speed sensors, new O2, new temperature sensors, new fuel pump, new fuel filter, no rust in tank.

Only thing I could think of is a failing fuel pressure regulator?

[5Toes]

FPR is $50 new for autozone brand, worth a try?

[jlevie]

Tee a pressure gauge into the rail supply line, zip tie it to a wiper, and see what the fuel pressure does while driving. That will tell if you have a fuel delivery problem.

[7pilot]

Follow post #3.
If the tach slams down to zero revs and back up to the correct reading during the bucking, it may be an ignition problem.
A fuel pressure test will reveal a possible culprit in the fuel system... A weak Fuel pump or a clogged filter at the foot of the in tank pump would then need to be investigated.

m

[mrgraphics]

It's not the FPR, because I had an extra one and it's still behaving just the same.

Pressure gauge is on order.

Checked the fuel tank and filter - it's completely clean and clear. No rust. I doubt that it is the fuel pump since it's a brand new Tre 255 LPH.


What about this theory: If I have a big intake leak, that means that there is more air entering the system unmetered. So if I move the needle higher on the MAF, that is essentially adjusting the MAF value, telling the computer that I have the "correct" amount of air entering into the system. Could it be as simple as that?

[Quinthirty]

Option 1: Intake vacuum leak?

Option 2: Wrong AFM values? Try an untampered 2nd hand AFM just for kicks...Reset ECU.

I'm unsure about what you mean higher or lower on the MAF. Are these the "wiper needles of the MAF/AFM?" There is a corresponding ohm(?) signal being thrown to the ECU, so when you move it (manual says north or south referring to north of your original "wipe lines," or otherwise), the ECU gets a different reading and correspondingly throws the wrong message for air / fuel mixture.

The Bentley has a recommended test for the AFM. Better yet, ask Greg///M. Hope this helps.

[mrgraphics]

Quinthirty: I'm talking East and West. When air enters the MAF, the wiper moves to the East. So I am undoing the screw, rotating the wiper east, and locking it back down. I have also done North and South to have it wipe along a fresh section of track.

I have tried another unadulterated MAF, and the car ran like crap. In fact, exactly like my current MAF with the wiper in the correct East / West position.

And the tach does not dip or change when this issue happens. The econo-meter does go crazy.

[mrgraphics]

I had a full diagnostic run on the car (first time I ever paid anyone else to touch the car) - and they indeed found a vacuum leak. I would have never found it, which shows why the carb cleaner trick isn't enough.

It is the bitch tube, and the leak is at the top where it meets the intake manifold. It's a really big one too. This would explain why it would run fine if I "richened" the mixture at the MAF. I was tricking the MAF into telling the ECU to calculate fuel mixture for the extra air entering into the manifold.

Any tips on how to fix the bitch tube without having to completely remove the manifold again?

[jlevie]

If only the upper o-ring of the oil return tube is bad, you can push the tube down and replace that seal. But if a new seal was not used on the bottom of the tube, get new intake gaskets and both o-rings, pull the intake, and do the job properly.

This is exactly why my first suggestion in many cases is a smoke test.

[603Racing]

Sounds like you found you cause.

Just in case that doesn't fix it, you could be experiencing a fuel pump issue. When one of the in-tank pumps goes bad you can see the symptoms you describe when the fuel level is lower. Friend of mine just replaced a fuel pump for this exact reason... any time he was lower than 1/2 tank of gas car would have less power on inclines, try to stall when going from a stop, all that.

[mrgraphics]

Almost seemed like things were ok, but then it started acting up again. Then I thought maybe I'm running the engine too cool as well (I have a 71 degree thermostat). Put an 80 degree in and it seems to be running more stable. Perhaps I was driving around on the warm up / rich settings all the time. Could also explain why the MAF liked being set higher / richer.

Kept the negative off the battery all night to reset the ECU. Cold, it started right up and ran like a champ all the way to work.

LATEST UPDATE:

Let the car sit for 40-50 minutes after driving into work, engine was still warm and it started and it was sputtering and choking, until it died. Went to the MAF, turned the needle richer (undoing the screw, moving the default needle starting position higher) car started and ran.

WTF?

[603Racing]

Go back to the basics. I know intermittent issues are a giant PITA, but you have to stick to the basics.

Test fuel pressure. Just like Jlevie stated. He knows what he's talking about.

Does a full tank of gas have an affect on the problem, like making it run better? If so, it's probably a fuel pump problem. Maybe an air leak at the pump, like a seal or a cracked pump/pickup.

You should also do a noid light test on the inectors, specifically when it's running bad, to see if the injectos are firing properly.

In short: Setup a fuel pressure gauge, do a noid light test. This will either confirm fuel issues or eliminate a lot of causes.

[mrgraphics]

Fuel pressure is good - a little over 2.5 bar.

Injectors are good since they were just rebuilt and flowmatched.

I isolated the problem: If I disconnect the coolant temperature sensor, it runs perfect. This is with an un-adulterated MAF. So something is happening with that sensor that is making the computer not do the correct fuel mappings at temperature. With the sensor disconnected, it's running on default startup mappings, which the car likes.

The sensor is new - in fact I have 2 that I bought from Pelican, a Bosch and a FAS. Neither make a difference. So it's in the wiring or the ECU itself is bad.

How horrible is it to run the car on the startup mappings . . . yea, it'll run a bit rich, but then it also feels a bit more powerful. And I don't care about a few less MPG.

[jlevie]

Having the engine run better in open loop mode (rich) suggests that something more fundamental is wrong. If the smoke test was done properly (exhaust plugged, AFM replaced by a plug or adapter for smoke injection, and 2-4psi of smoke pressure held for at least five minutes) then intake leaks can be ruled out. In a like manner the injectors should be okay and 2.5 bar of rail pressure when the FPR vacuum line is disconnected would be correct and you were using what should have been a good AFM.

At this point I'd want to see what a known good DME does and what the rail fuel pressure is when the symptoms are present.

[mrgraphics]

I watched the guy do the smoke test, and he did it as you describe. That's how I found the bitch tube leak. No other leaks exist.

I have a spare ECU, but I'm not sure of its working condition. I bought it for the performance chip it contained. I'll give it a try to see if it runs any differently and report back.