BMW M54B30 - Specific high fuel trim/misfire condition

2001 Z3, 160k miles

I did a CCV replacement, intake gasket, every rubber piece I came across. I've got a problem now that I can't figure out

Start up - sounds like a misfire, not BMW smooth. Hesitation in throttle response from off idle.
Sit in driveway at 1500 rpm and let the coolant temp come up to normal
Driveability is pretty good. Hesitation with a sharp press of the throttle, then it picks up and goes.

Short term fuel trims are pegged at 27%. If I am steady state in 3rd gear and let off the gas pedal, the STFT drops to 0% until I push the clutch in, then right back up to 27%. The data stream shows solid 27% except for the shift points.

Out on a 50MPH road, if I get to steady cruising at 50MPH and then just slightly let off the throttle to let the car coast down to approx 46-47 MPH, all of a sudden the STFT will drop straight to 0% and it will start to misfire pretty bad. If I add throttle it'll clear up, STFT jumps back to 27% and the driveability returns to normal-ish. Still doesn't have the immediate throttle tip-in I would expect.

I smoke tested it once and didn't see any leaks. I can do this again to see if I missed something.
Fuel pressure was 45+ psi key on engine off. Idling is 50 psi and it stays steady to 2500 rpm - no drops.

I've driven it about 40 miles since the engine work. I don't get a CEL. The DME doesn't light the EML lamp but it does store 10 codes - misfire 1-6, short term fuel trim maxed out and permissable range exceeded for both banks.

Are there other data streams I should be looking at to help figure this out?

I would put a vacuum gage on it. I would also be looking at the AFM/MAF. Maybe a wire is broken or a ground is bad or a plug is loose. I assume it worked before you "fixed" it. ;-). 

Like he said ^^^

That sure sounds like unmetered (or incorrectly metered) air...

I have never had any luck finding vacuum leaks with homemade smoke machines or carb cleaner, etc. I am cursed as an ex BMW guy with this affliction. Seems u may be as well.....

I wonder if it might have something to do with the throttle body itself, or something in the DBW system. I don't have my Bentley covering these engines any more or I'd give you something more specific to look at.

Also, while it would be the apparent opposite of the issue you're having... Should the fuel pressure be that high at idle? Does this have a plain old return type fuel pressure regulator?

Just wandered back to this thread and thought I'd ask whether there were any developments...

You've got a hose off.  Either one of the crankcase vent hoses, or the throttle body boot is not securely in place (or is cracked), or something.

What everyone else has said, you have a vacuum leak. It is also unmetered or your stft would be sub 5-10% all the time.

What are your long term trims? Those really tell you what is happening. Your o2's are pegged lean because of the leak so, the ecu is dumping all the fuel it can to make it stoich. 
You will see a zero on stft on decel as there is not any fuel being injected on coastdown.

Played a bit and bought OBD Fusion to do some real data logging.  If anyone is good at understanding these I'd love some guidance.

Still very misfire-y

LTFT is 0.0% on both banks, even after 50 miles of driving.

I smoke tested it and I don't see any leaks.  When I remove the oil cap I can tell pressure was built up and I get smoke out the oil fill cap.

Brake boosters are becoming known to be another lean source, but I did the multiple brake pump and the STFT never wavered.

I think the O2 sensors aren't responding correctly.  I'm going to data log a functioning car to compare.  Then need to try to figure out if it's in the wiring or something else.

How thorough were you with the CCV job? Did you do the valve, all four hoses, and clean the distribution piece and replace the O-rings? Did you check that the drain in the dipstick guide tube is clear?

O2 sensors and brake boosters seem like long shots given the stuff that was just apart.

Stuff does coincidentally break, but it doesn't seem like the likeliest culprit candidates.

Speaking as someone who has absolutely convinced myself to chase red herrings too many times.

In reply to Jesse Ransom (FFS) :

The CCV hoses are notorious for cracking when disturbed. I wouldn't consider that job done unless all four hoses were replaced along with the valve.

In reply to 02Pilot :

Oh jeez, I wouldn't have even considered the possibility that the CCV would have been replaced without doing the hoses.  You have to break the hoses to get the old CCV out!

I'm curious why the intake manifold was removed, too.

Sounds like a mild version of when I forgot to plug in the MAF on my E36 after some work. 

Yes I did all of the CCV hoses and the valve itself.  On the Z3 the CCV is completely under the intake.  Maybe I could have done it, but it was much easier with the intake off.  I've done the job before on an E46 - this time I could hardly see the valve and couldn't even see the connections before disassembly.

All four O2 sensors aren't acting right.  I don't think I had to disconnect them but I need to spend some time tracing the wiring from the sensor connector back to wherever those go.  I put the car off to the side for a couple of weeks and then will get back to it.

Maybe the valve is bad new out of the box, but I don't know how to check that.  I am trying not to stray far and chase things not related to the parts I actually touched.

In reply to glueguy (Forum Supporter) :

The M54 CCV is awkward to change, but if you do it often enough you can complete it blind in about 45 minutes - I can't tell you how many times I had mine out trying to solve my oil consumption issue. A little dish soap on the O-rings helps to seat them more easily.

You can measure the CCV's function with a slack tube manometer. Sounds fancy, but you can make your own with a board and some clear tubing. Spec is 4-6in H2O, which is a really tiny amount of vacuum. I modified my system (due to the above mentioned oil consumption) to run a lot more vacuum in the crankcase, which had to additional side benefit of keeping the CCV lines totally clear of oil-water emulsion, which can build up and cause problems.

In reply to 02Pilot :

If I remove the oil cap while running and there is definite vacuum and it runs much worse, is this a good-enough sign that it's functioning?

In reply to glueguy (Forum Supporter) :

That is a sign that it's defective.  Should barely be able to suck a latex glove to the fill hole.

or get/make a manometer to be able to measure 4-6 inches of water, which is like 0.3 inches of mercury.

It runs worse because you're introducing a huge amount of unmetered air, but yeah, there shouldn't be that much vacuum on an unmodified system. The spec is so low there's really no way to tell if it's right without a manometer.