limp mode but no error codes

[S2central.net]

Its gonna be something stupid / unexpected / overlooked...

[20Vcqdriver]

Found the N80 valve in the garage (lucky it didn't get chucked out) so going to connect it up under the dash as cut the wiring in the engine bay for this but kept the plug. So let you know.
Weird how it cuts out to limp when the VDMS page of VAGCOM says all the sensors are spot on within the ranges allowed...
Will check all wiring too and have re instated the breather system that appears to have helped the smoke issue noticably. Now goes via a catch tank before reaching the MAF to turbo hose as oil was getting into intercooler last year via breather system...
So just got to get this limp mode sussed.

[S2central.net]

Just for the removal of doubt... I have never ran a 3B with the carbon canister solenoid (N80 - not 'G80' as stated in Post #1) disconnected... I know the ECU will raise a fault for this, but I wonder does it also operate differently because of this....

I've just walked over to the bookshelf and picked out an Audi training book for the 3B engine and it doesn't specifically mention this scenario for N80 failure.
However it does say that if the fuse protecting N80 (and also ISV & WGFV) fails that the engine will go into limp-home mode... Most likely a result of inoperative WGFV, but it is possible at least that open circuit on N80 gets interpreted as a serious fault by the ECU.

So two things to try then - someone with a healthy 3B to disconnect N80 and observe any differences whilst 20VCQdriver simulates the presence of N80 with a resistance of 40-50 ohms. Important - This resistor needs to have a power rating of 3W to cater for 100% EVAP cycle from a factory map inside the ECU. Or alternatively - just plug in the old N80 with no pipes connected.

[20Vcqdriver]

4th gear, foot down at 2000rpm and 2+ secs later I see 1.2 bar at 2600rpm, by the time I get to 5000rpm its in limp mode whichever gear I'm in. With the K24 thought it would jump up to over a bar in under a second, but it takes a second before anything happens. Is this normal?! The 350bhp Engined s2 I drove last year seemed to have less lag...

[Rusty]

What gear? In second I'm at 3500 by full boost on a run from 15mph

[20Vcqdriver]

Well! I stripped all the hoses off the back of the manifold and tested each one and found the one to the ECU and pressure reg had a split in it. Not very big but trimmed it back re-fitted and thought that had fixed the limp mode, but er NO. Still the same,
Going for a proper boost leak test next week and managed to locate a working K24 that has no smoke and has previously run on a re-chip without issues. But if non of these 2 issues can sort it, its into the garage and close the door until next year when hopefully I have saved up enough to carry on finding the issue.
One thing that is odd and that Tim is baffled about is why it takes so long to spool up. A K24 should jump to max boost from 2000rpm very quickly but mine takes around 2+ seconds by which time 2600rpm is showing...

[20Vcqdriver]

Still think its an air leak, though MFTS is worth trying just to rule it out.

Rather than visually inspecting the diaphragm - you need to apply vacuum to the upper WG cap. That should be completely air tight for many minutes. A MItyvac is the weapon of choice for this, but you could probably use a compressed air line (with a guage) to pressurise the upper chamber - not too much though - and see if it is airtight.

What about your DV/TBV - checked it for air leaks too ?
Or any auxiliary hoses linked back to manifold.

Yes DV checked out ok. Would a large boost leak not cause a code of one sort or another?
All manifold small bore hoses replaced after finding split in one last year.

[S2central.net]

Still think its an air leak, though MFTS is worth trying just to rule it out.

Rather than visually inspecting the diaphragm - you need to apply vacuum to the upper WG cap. That should be completely air tight for many minutes. A MItyvac is the weapon of choice for this, but you could probably use a compressed air line (with a guage) to pressurise the upper chamber - not too much though - and see if it is airtight.

What about your DV/TBV - checked it for air leaks too ?
Or any auxiliary hoses linked back to manifold.

[20Vcqdriver]

Good point - hadn't thought of MFTS but yes - its a candidate... though doesn't explain oily zorst.

Unmetered air is simply a leak in any pipes after the MAF that can be a source of excess air under vacuum. The normal candidate for that one is a damaged MAF hose, leaky PCV pipes or a split diaphragm in the wastegate.

Do you get the G28 error when the engine is running or not - THAT is important !

-Surely a dodgy MFTS would place in limp mode all the time not when you floor it past 4000rpm.
-Do not get the G28 code or any code if ECU codes reset whilst engine running and then drive off to get limp mode and recheck codes when I get back.
Also checked the wastegate working properly. All seems fine (cap actually came off easily..)

[S2central.net]

Good point - hadn't thought of MFTS but yes - its a candidate... though doesn't explain oily zorst.

Unmetered air is simply a leak in any pipes after the MAF that can be a source of excess air under vacuum. The normal candidate for that one is a damaged MAF hose, leaky PCV pipes or a split diaphragm in the wastegate.

Do you get the G28 error when the engine is running or not - THAT is important !

[AndyP]

Try unplugging the MFTS, had this on my brothers car recently. It would only boost to 0.4bar, the MFTS was faulty and it does not throw up an error code.

[20Vcqdriver]

Rather than a 2nd recon'd turbo passing oil, could have a mistake in the PCV plumbing such that you are venting excessive blowby gasses ?

G28 intermittent would be a worry - thats a very important sensor for engine RPM so I would look to change that if you have bad contacts or a dicky sensor.

Limp home can be a right beeatch alright and it doesn't always give codes. IME this is normally just a good old air leak somewhere - unmetered air.

Have you checked the WG diaphragm is airtight ?

- Venting excess blowby gases...I no longer have the PCV system feeding back to the inlet, but into a small catch tank and out through a filter
-Changed G28 without any effect.
-unmetered air- where do you start with that one?

David

[EspenW]

You will trip a code on G28 if you read codes with the engine off.

[S2central.net]

Just had the codes read on my 3B as it keeps going into limp mode. Only came up with G28 intermittent fault and G80 carbon canister recirc valve (which I never connected or used during the conversion). No errors regarding boost or why it trips to limp mode.
Reset the faults whilst engine running and shot off down the road, straight into limp mode so kept it running and connected VAG-COM, No errors...HELP!!! Clamped the hose to the wastegate and I shot off down the road again. over 1 bar straight away but no codes on returning. Baffled. Turbo seems to be passing oil again too, Doh!

Rather than a 2nd recon'd turbo passing oil, could have a mistake in the PCV plumbing such that you are venting excessive blowby gasses ?

G28 intermittent would be a worry - thats a very important sensor for engine RPM so I would look to change that if you have bad contacts or a dicky sensor.

Limp home can be a right beeatch alright and it doesn't always give codes. IME this is normally just a good old air leak somewhere - unmetered air.

Have you checked the WG diaphragm is airtight ?

[Rusty]

I know that if the ECU loses the crank or cam position sensors then it goes to mechanical boost only, as for anything else, you'll have to ask someone else I'm afraid.