Need some help, boost dropping off

[newsh]

Well I thought I was over this but no.. Every now and then, at WOT the car gets to about 4000 RPMs and the boost backs off, doesn't really feel like a proper misfire, but feels like its holding back, will drive almost as fast but not do this at less than full throttle, which is where I drive it most of the time.
As you all know, revving one of these things out through the gears, with a full RS2 job and Mihnea mapping takes a decent bit of empty road so Its hard to repeat the fault, sometimes it seems to be ok, other times if I floor it in 5th it will hold back above 4k, sometimes it will hold back in 4th. Other times it will hold the boost and head for the horizon. The worst scenario is WOT to 5k+ though 2nd 3rd 4th 5th.

The only thing I can think of is that maybe the ABY airbox and stock filter will just not pass enough air to hold full boost in the higher gears or maybe the MAF hose is collapsing, still havent adjusted and fitted the Samco one. Need to get this modified and fitted, but I feel this is not the cause..

Anyone got a big filter going spare I could try out or RS2 box and hi flow panel filter??

I might try and run the car with no airbox at some point over the weekend to test this theory..

Pressure test holds up (except to blow oil into the exhaust and make a smoke screen on start up) I can get enough pressure to make the oil filler blow air and have to hold it down.

Could bypass valve be opening?? Bailey DV30? recirc type.

Entire PCV system replaced.
Multifunction sensor replaced.
Related small sensor and otter switch replaced.
No fault codes last time I blinked.
Plugs good but looked to be running rich last time I did a chop..??

I'm getting a bit cheesed off now and need some pointers.

[quattronutter]

dodgy air temp sensor maybe?or a coilpack on its way out possibly??

[AAH!]

knock sensor?

Mine would intermintantly, less often than not, hold back boost at around 4000rpm which occasionally, but not always, would kick back in at 5500rpm, it ended up being a dodgy knock sensor connector. A blink code did point me in this direction, but i originally thought it was running lean.

[Graham]

Before i left the Uk my problem was i would be going for top speed, it would get there then the boost would start backing off and backed off to 0.3-0.5bar max and the engine would loose its power. As a result the speed dropped right down, oil, water and all other readings were fine, oil at about 100-110. When i backed right off to legal speeds and nail it again it would only give me 0.4-0.6 bar boost in any gear at any speed. When i pulled up to look under the bonnett the turbo and first inch of downpipe was glowing bright red. Underfuelling???, changed the fuel pump for one from ECP but still believe i still have the problem-any suggestions anyone, i am just scratching my head, thinking clogged injectors, or dodgy FPR, the turbo is worn with quite a bit of play on the shaft. I have a spare S2 turbo, ( cheers Mark H ) which i plan to have re-conditioned, as if it is just the wear in the turbo causing excess friction and in-efficiency then the idea sounds good.
Any help would be mucho appreciated just so i can plan what to do next. So could i have my injectors cleaned, surely it wouldnt do any harm.

[newsh]

New motorsport fuel pump and filter fitted recently. Deffo not lean.

[porkkis]

Knock sensor is a good guess. ABY/ADU/AAN ecu has the ability to reduce boost if the engine is knocking. AFAIK 3B only redards the ignition.

So this could be false knock signal so that ECU thinks the engine knocks even if it doesn't.

I though about perhaps a bit agressive timing by Mihnea but not likely as AFAIK (again) the timing is first retarded and boost reduced only if it didn't remove the knocking.

[Rusty]

Dodgy batch of fuel could be the cause. Mine was runnig poorly even though the boost was at 0.8/1.0 bar depending on overboost. Turned out my old man had put 95 in it and hadn't told me. The knock sensor was obviously doing it's job as the car is always run on Optimx. Had the effect of reducing the ignition timing rather than boost though. Try some octate booster if you have similar symptoms??

[newsh]

Hmm, knock sensor or a fault which looks like knock to the ECU.
(ECU would throw in extra fuel if it senses knock too which would account for the rich looking plugs??)
No fault codes though.. This has been going on now for over a year, it even appeared when Mihnea was mapping it, then went away again, no fault codes were found then either..

[Rusty]

Running rich could indicate a small boost leak or a indeed a misread from the MAF. I am not familliar with the actions taken by the ECU to reduce knock so you'll have to wait for an answer to taht one. Have you replaced the small bore hoses? I was quite suprised at the poor state of many on mine when I replaced them! They were almost certainly leaking a small but important amount of boost pressure.

[newsh]

Went ot today for a test drive, ran perfectly, would not play up

This means, I think, that its an intermittant fault.. Will have a look at the knock sensor connections I think.

[newsh]

Spent today doing a pressure test which was ok, have a small leak from the throttle body which changes when you wiggle the butterfly, its leaking at the pivot. I guess the only way to fix that is a new TB, its really small though. I also fitted the Samco MAF to Turbo hose (finally) and cleaned the WGFV (again). Had to do a lot of trimming to get the Samco piece in there and still fit the elbow to the DV which now rests against the hose coming from the RS2 IC to the inlet tube. I see now why they changed it around on the ABY, it would make a lot of sense to use an ABY MAF hose and reposition the DV, I'd like to see if it could be done as I'm still not really happy with the clearance still..
Anyway the car seems to be running ok again will have to see if it stays like it, if so then I can only assume that the old MAF hose was collaping with the suction from the turbo, that or the WGFV was sticking and WD40 has freed it up.

[newsh]

Anyone had any experience of this leak at the throttle butterfly pivot?

Oh Yes, I read the blinks and no codes stored.

[newsh]

I hardly dare believe it, seems to be running ok, boost holding nice and steady.