Non-GT4 misfire question, looking for ideas

[Coyoteboy]

Not sure if mods want this in the tech section or if it should stay out as its technically off topic but I thought I'd post it up ...

BMW 535i. strt 6, SOHC. Recent HG, valve guides,full intake rework all with genuine gaskets etc. Tested, so far, all elec items (sensors etc) all read normal, codes on ECU - reads fine, identifies intentional sensor pulling so its working ok as far as we can see, swapped entire ignition setup from dizzy to plugs,compression tested - reads bang on perfect across the cyls with +/- 2.5psi. Sparks strong and regular on all cyls. Searched the intake for leaks, cant find any, including detatching the brake servo and blocking it up.Checked idle control system works, seems to move just fine. No problems with anything attached on vac lines. Swapped out the air flow meter twice and adjusted the idle bypass with no real effect.

99.998% sure the static timing is correct and the chain is tensioned fine, yet still she barely starts when cold (on LPG or petrol) and runs like a lumpy lump-based thing, you need to give it ~1500rpm to keep it running until its warm, when it just overcomes the misfiring and settles at normal revs but with misfires. Just cant figure out what the hell else it can be. Posted to BMW forums and they came back blank so i thought id see if anyone here had an inkling or could see anything ive forgotten. Its symptomatic of an intake leak yet it has all new injector seals and was assembled with good care. It has a very poor idle vacuum, yet the exhaust if very free-flowing so its not an exhaust leak choking it. Its not using water/oil.

Anyway, bit of a poor show considering I'm the technical spanner monkey but ive just about got sick of it and to put the nail in the coffin we fired it up with the inspection lamp laying across the viscous fan which made a nasty mess and took 30 minutes to retrieve glass items and plastic cage buried in parts of the alternator, then find the fusebox in the dark as the torch batteries were going flat.

Bad day!

[Driveby]

Does it have egr? possible egr valve not closing or seating correctly or possible split internal diaphragm if vacuum actuated, when you say all sensors reading correctly Im assuming your using a snap-on scanner or similar code reader for live data? Im sure your aware and Ill apologise now if Im teaching (you) my granny to suck eggs but a sensor will only flag a TC if it runs out of its normal parameters, saying that it can still be duff but not throw a code up, ie coolant temp sensor reading 100c on a cold engine or -10 on a hot engine, neither outside its parameters but both wrong for the current state!.  Are the plugs wet when checked? whats the mixture like? have you had the injectors tested? checked your fuel pressure? blocked return maybe? can you isolate which cylinder the miss is on? infrared thermo on the manifold runners? 
Inlet leak will usually show on idle worse but be masked when the rpm's are raised to offset this.
If you have live data, check the idle air control valve readings, if you have an air leak this will read 0 steps (or volts which ever the system displays) as it will close off as the engine is getting air from elsewhere and does not need to use the regulator valve.

Youve probably checked all of the above but if I can be of any help just ask

[Coyoteboy]

Some interesting points there, I'll start at the top...

No EGR that i can find, i will double check as Im doubting myself now!
All sensors tested in situ and read correct values for temperatures/positions including all coil/VR type sensors having correct resistances and signal output waveform (checked on oscilloscope).
All plugs vaguely damp looking. Baked them, re-insterted - all vaguely damp again - presume misfire is not cyl specific. Dont have a thermo camera handy, but my mobile does have a strange habit of picking up IR as bright blue so will test that, as I say though -it seems to be a non-cyl-specific miss so might not show up. Not checked fuel pressure but as it is petrol/LPG I've run it on both and neither runs better. Possible, i suppose, that a petrol injector is leaking but I would expect one cyl to be wet and the others dry. The LPG injectors have a secondary sealing valve that cuts them all off when the kit is off, plus you can hear them leaking and they dont seem to be.

The problem is more apparent when a)cold b)idling, revving it clears it a little but it is still present and drastically reduces power. It does seem to fall into intake territory but ive been over and over the hoses with no findings Half of them are BMW specific, £100 a time hoses - grrr. I cant watch the idle control valve DC, but i did watch it directly and immitated it with my thumb as the valve. It operates fine and my thumb does too, but neither stop the misfiring . I've yet to hook the LC1 up, the normal O2 sensor shows lean as hell (normal misfire reading) and the exhaust stinks of whichever fuel it is running on.

[Driveby]

Have you tried disconnecting the exhaust at the front pipe to completely eliminate the exhaust sytem? yeah itll be loud but you can mark it off as 100% ok (I know you said it had good flow at the back box but this will check the system for deffo)
What did you use to check the intake for leaks? brake/carb cleaner or acetylene work great (dont tell the health and safety dude )
Does the TPS have an idle contact? Ive had people mess with throttle stops to try and offset idle problems and this can have knock on effects to other systems.
Does it have an external crank sensor? if so whats the air gap like from the pick up? (I know you said youve checked the wave form on the scope so that should clear that one up)
When did the fault start? was it right after the HG was done?
Dammit itll put me boots on and come down and have a look

[Coyoteboy]

Not disconnected the exhaust, its a bit of a nightmare to get to and rusted to hell, will certainly give it a go soon thought as im running out of ideas! It always had a sniff of a misfire but after the HG/valve job it was very consistent and got progressively worse. The problem is that all of these things were done at once in an attempt to bring it back up to scratch and it only got worse!

Checked intake with butane torch but im going to go over the whole lot again tomorrow if i get the chance. TPS was never altered, nor the throttle stop - that had been my first culprit but it tested identical to 2 years ago before the problems. Crank sensor is giving a good signal and has never been shifted - technically no reason for it to move but i'll double check despite the good wave shape and size.

Its such a simple engine im really struggling to see where it could be going wrong! If the timing had jumped 1 tooth I cant see it misfiring to hell (1 tooth ~ 5 degrees), its not impossible though. Quite odd. I've been over the car 3 times now, initially as a stab in the dark process, secondarily as a systematic approach replacing parts with older parts, now going over every last setting, continuity test and resistance/signal check. Car is my dads, its been off the road for a month now because we cant cure it!
  :'(

[Driveby]

Lack of power and poor vacuum points to timing but that shouldnt really give a misfire :/ Has it got an external coil trigger/amp setup (I know you said youd swapped the dizzy etc but an under performing amp can give similar probs)
Did you do the HG? What if 2 vac hoses have been mixed up? I do a lot of jobs that have come from other garages that dont run right since xyz and people getting things mixed up is a first port of call, also never trust anything they say is ok, I ALWAYS check for myself.
Have you tested the battery condition and alternator output? the ecu etc all have good clean earths (jump lead from the block to the battery neg for a quick check), no water ingress anywhere?
Have you had the ecu tested? can be done for about £40 for piece of mind.
Good luck mate, Ill have a ponder today if we're quiet.

[Coyoteboy]

Ign amp is in the ECU, coil readings are fine and the spark is clear and strong while cranking, havent checked it while running. Amp is in the ECU unfortunately. Only has (would you believe it) 2 vacuum hoses on the intake! They cant get mixed up, but i havent tested the fuel pressure reg diaphragm yet, will do that in am mo.

Battery/alt are a contentious issue - it /was/ fine, ages of sitting dead and cranking/restarting seem to have made it unhappy - ive never had a poor battery/alt cause serious misfires though so i hadnt suspected it as a problem. Will transplant a battery and test the alt too.

[Coyoteboy]

Cured the major misfire, it was a leak between AFM and throttle which only opened up when the idle control valve was inserted into the hose. hence multiple removes, stretches and checks over didn't find it! Its still missing a little but its perfectly drivable now, i can work on the other mis in due time!

[Driveby]

Glad you got to the bottom of it mate

[Coyoteboy]

Ta for your time and effort!