Severe hesitation when accelerating

[vladness]

This is my first post, so hopefully I'm not breaking protocol in some way.

I picked up a used 330xi (manual transmission) a while back, currently at about 180k miles. It's accumulated a few issues, but the most severe issue is what i'm trying to address now (I'll list others for completeness, in case they could be contributing factors).

The issue:
When accelerating, there's severe hesitation and power drop. More pedal == more hesitation, most pronounced at low RPM.

Previously:
This past month, my car developed a rough idle, and a week or two ago, a rattling sound started coming from around the engine. The other day, I hit random misfires (P0300--P0306, inclusive). I replaced the full gamut of spark plugs and ignition coils, and I cleaned the MAF sensor (it looked clean, but I cleaned it anyway). No more rough idle, no more rattle, no more codes. The car performed well on a short test drive. Next day, I hit that acceleration issue, starting just a couple of blocks out of the driveway.

Possibly relevant history:

• I replaced the fuel pump about a year ago (OEM), which was my first real DIY job on the car.
• I replaced the fuel filter about 6 months ago (Mann). The old one was pretty dirty, but I guess they always are.
• Replaced the intake air filter about 6 months ago (Mann). Of course, we've had a lot of smoke in Seattle since then, but the new one still looks pretty clean.
• All new spark plugs (NGK Laser Platinum 3199).
• New ignition coils (a couple of Delphi, but mostly OEM after swearing off Delphi).
• Last tank of gas (not the current one), I used Techron fuel injector cleaner. I put a full bottle in, filled up the tank, waited until it was pretty low, and refilled with fresh gas. The random misfires started about 30 miles into the fresh tank.

Diagnostics:

• No CEL anymore; no codes or pending codes that i can read with an OBD2.
• Plugs all had oil on the threads and were somewhat fouled.
• Fuel pressure:
  
  • ~2.6 bar at pump prime
  • ~3.25 bar at idle
  • After turning off engine, pressure remains within spec (~0.3 bar loss after 20m)
  
  
• Air filter looks good.

Probably not relevant:

• For several months, there's been a soft heartbeat that can be heard and felt through the steering wheel at idle. I think it's the VANOS, and it's on my list of things to address when I get around to it.

Current plans:

• Rebuild DISA valve. I'm waiting on a kit from G.A.S. to arrive. I'm thinking that's the rattling sound that I heard, but unfortunately, without the rough idle, I can't duplicate it for testing (I guess I could just remove an ignition coil to incite rough idle again).
• Get INPA running on an old XP VM. I dug one out of my old backups, but I don't have it running yet. I did pick up a K+CAN/DCAN cable. I only have Linux on my laptops, so hopefully a VM will work well enough if I enable USB passthrough for the device. From what I've read, there may be some codes or problems that INPA can see, but that aren't being picked up via the OBD2 device and mobile app.

Other thoughts:

• Should I replace the fuel filter again? Is it possible that the fuel injector cleaner dislodged a lot of gunk, and now I'm getting poor flow through the filter?
• Does that fuel pressure seem appropriate? According to some specs I read, it seems a tad on the low side, considering that the pump was replaced a year ago.

Any other likely candidates that I'm missing?

Thanks much!

[vladness]

Some further diagnostics:

• There doesn't seem to be any hesitation when the engine is cold, but it recurs once it starts warming up.
• I performed a throttle adaptation reset to no effect.

[MarcoZandrini]

Have the intake smoke tested for vacuum leaks. Why? The engine runs better when it’s cold. When the engine is cold the engine computer (DME) runs a rich fuel/air mixture to get the cats up to temperature quickly. Has the VANOS ever been rebuilt?

Remove the fuse for the fuel pump with the engine running and allow the engine to shut off. Remove the fuses for the DME. Perform a compression test, both dry and after adding a teaspoon of oil into each cylinder. After putting oil in the cylinders (leaving the spark plugs out!) try to “start” the engine to distribute the oil. Then perform a wet compression check.

[johnjmurray]

Vlad: Sounds like you have a good mechanical background, at 180K you're gonna need it. I had a similar issue with acceleration. Turned out to be an error in the DME. I had to update the software to get my throttle back. I think you're on a good track getting INPA going. A regular scanner doesn't see everything... The hard part for you is gonna be getting the VM to recognize the cable on the laptop. You may want to dedicate a spare as the car computer.
I don't recall the spec numbers but your fuel pressure seems ok to me. Oil in the plug wells means it's time to change the valve cover gasket and grommets. A leaking VCG can also throw excess air codes and induce other gremlins. I think the spec on fuel filters is something like every 30K but few of us do that.

[vladness]

Marco and John - thanks for your input so far.

Another useful detail: The problem only seems to occur south of 3000 rpm.

I've gotten XP, INPA and WinKFP configured appropriately in a VM. No problems so far using the cable with the VM; I enabled a filter for the USB-FTDI device in VirtualBox so that it's always claimed by the VM automatically.

INPA:
I found a bunch of old/inactive codes related to my previous cylinder misfires, but after clearing the codes and driving for a couple of days, INPA still shows no errors. Shadow memory reveals codes 90/91/93 (TEG_up_1/TEG_down_1/TEG_down_2) that have not cleared, but my understanding is that this is a red herring. The analytics in INPA look okay to me. Everything's in the green. VANOS positions match target positions, and i think that the lambda/O2 sensor voltages look good, but I'm not really sure that I know what I'm looking for there. I used INPA to clear all adaptation settings as well, but this didn't have any effect on my problem.

WinKFP:
This was a pain. My old VM was XP SP2, but WinKFP didn't work properly for me until I installed SP3.

Updates:

Have the intake smoke tested for vacuum leaks.

I don't have the materials to do a smoke test yet, but I'll work on it, especially if my immediate plans don't give me results.

Perform a compression test, both dry and after adding a teaspoon of oil into each cylinder.

This sounds like a good idea to do regardless of my problem, so I'll try to do it soon. I should have the necessary tools, though I'll have to buy a cylinder leak tester if I also want to test leakage (which I probably should do while I'm at it, though I naively suspect that leakage wouldn't manifest as this particular issue without also showing up as a problem in compression tests).

I had a similar issue with acceleration. Turned out to be an error in the DME. I had to update the software to get my throttle back.

John, after reading your comment, I did some research and found several posts with similar-sounding problems that were solved by this. Now that WinKFP is working, I'll try to update the DME software tonight. I'm going to be (happy but also) annoyed if this solves the problem.

Oil in the plug wells means it's time to change the valve cover gasket and grommets.

Sounds reasonable; I'll add it to my to-do list.


Immediate plans:

• Update DME software.
• Do the DISA valve rebuild (my rebuild kit arrived).
• Dry/wet compression test, maybe also test for cylinder leaks.
• If the above don't solve the issue, change the fuel filter again (just to rule it out).

Future plans so far, regardless of whether the above solve the problem (but sooner if they don't):

• Intake smoke test.
• VANOS rebuild.
• Change valve cover gaskets/grommets.

I'll report back with status updates, but in the meantime, let me know if any other promising leads come to mind. Thanks!

[horishima]

I may have the same problem please update thanks

[MIKYZZ4]

If only occurring above 3000rpm, I would focus on results of a:
Fuel pressure test(psi) at > idle-1500-2500-3500(rpm).
Exhaust backpressure test.

[vladness]

Solved now.

(mikyzz4 - It mostly seemed to be occurring below 3k rpm, and only under load. Or maybe the real number was closer to 2700 rpm.)

The DME update didn't help. Funny enough, it seems that flashing the DME is a rather power-hungry operation. In the 14 minutes it took to successfully flash, my battery drained to the point that it was only supplying ~8.5V, and I needed to find someone awake in the middle of the night to jump my car. So, for posterity, make sure you do your DME flashing with a heavy duty charger connected (or another car, idling and hooked up with jumper cables) unless you're _very_ confident in your battery.

But the actual issue was one of the new ignition coils. I removed them all, preparing to do a compression test, and one of them simply fell apart, with a smell of electric smoke, as i removed it from the engine. I looked at the spark plug in the cylinder, and the top of it was tarnished, as if there was a poor electrical connection (which i imagine was in fact the case). I replaced it with a good coil, put everything back together, and the car ran perfectly.

I'm curious as to why it didn't throw a code.

Anyway, thanks to everyone who offered suggestions.





^^ Does that look like an old, crusty most-of-coil, missing its rubber boot? That coil was brand new a week and a half ago. The rubber boot practically disintegrated.

[MIKYZZ4]

Where did you source your coils, along with the other replacement parts?,
also are they BMW OE or OEM Replacements?
Purchasing the cheapest replica-Chinese counterfeit parts would lead to current problem.
Btw, FCP offers OE & OEM parts with lifetime replacement warranty.

[vladness]

Where did you source your coils, along with the other replacement parts?

This coil was O'Reilly Auto's "Import Direct" brand, made in Poland. O'Reilly claims that these parts are at least as good as and possibly better than OEM. I replaced it with Bosch, which I think is OEM, albeit a newer model.

[MIKYZZ4]

NEVER purchase BMW replacement parts, especially sensors or electrical,
from O'Reilly, Advance...etc.
The claim of as good or better than OEM...LOL!, what a joke.

Stick with a source such as FCP, or one of the other BMW sites posted throughout this forum.
Here is a link to the options available at FCP:https://www.fcpeuro.com/BMW-parts/33...004&m=20&e=180

[johnjmurray]

Vlad: Glad you finally found the (literally) smoking gun. What Miky sez is true. One of the things I've learned working on these car: If it goes in or on the engine, make sure it has a Roundel on it.

[vladness]

Supposedly, O'Reilly also has a lifetime warranty on these parts. But, I agree, that doesn't mean they're necessarily as good as OEM (though I think that in this case, both are just rebranded Bosch?). The only reason I got that coil from O'Reilly --- and paid more than necessary --- is that I needed the part same-day. Unfortunately, I didn't have a lot of options for that.

In retrospect, I think that the part didn't just malfunction; I think they accidentally sold me a part that was bad and had been replaced under warranty. When I examined it more closely, there were all kinds of signs on it that it was old and had been thoroughly used before I got it.

[MarcoZandrini]

A lifetime warranty won’t get you to your destination if the part fails!!! Parts like sensors, coils and cooling system components need to be original manufacturer. Period.