MV - mis-fire - help!

Post by **Diamond Hell** » 01 Dec 2005, 23:22

Right i've just fitted a replacement silencer and J-pipe and mine were well beyond the end of their lives. The remaining new-ness was rubbed off them at Coney Farm.

The silencer is a DJ unit as is the J-pipe. I've cut a hole in the J-pipe and welded the lambda boss into it and hooked up the lambda as-was. The lambda is vertical in the pipe.

What happens is when I take the van out it starts fine, then as it starts to warm up it starts mis-firing and it won't idle. If I stop it and re-start it (or it stalls) then it runs OK for about 30 seconds, then everything goes titsup.

Have I installed the lambda in a bad way?

Should it be shielded from the passing gases in some way?

Is it just crap luck?

Could I have damaged it when I was tack-welding the boss in (I removed it when I welded it in properly). It was disconnected the whole time - I was on the work bench.

Can I run the MV without it?

Any thoughts on this greatly appreciated.

Post by **Diamond Hell** » 02 Dec 2005, 00:47

£10 says it's the lambda (which to be fair took a couple of heavy hammer blows on the ring spanner to crack it off the cat!).

Put £20 of fuel in it and took it out on the motorway quickly, thinking 'this'll flush the bugger out'. Nope

Not until I came OFF the motorway, onto THE SAME roundabout where the problem first surfaced did it re-appear. So I pulled in and disconnected the lambda, restarted, looped around again, no problems.

What I need to know now is this - could it be due to my mounting of the probe, or is it a faulty sensor? Your views on the mounting and any specific requirements of that first, please.

Any (helpful) thoughts appreciated.

Not exactly the weather for test driving - blowing a gale and hacking down with rain! I was struggling to detect mis-fires etc over the hiss of the spray while I was busy keeping the van on the road in horrendous side winds!

Anonymous · Post by **Anonymous** » 02 Dec 2005, 01:09

It may just be plugged up with crap which has wound onto the sensor as you've removed it from the cat. However you can check if its still working by measuring the voltage ouput.

This is a pretty good link for info on what to look for and these are the usual wire colours for VW stuff.

http://www.wps.com/LPG/o2sensor.html

2 White wires = positive & ground
Black wire = sender
Grey wire = earth

Oh, the lambda sensor on the MV engine is particular to the T25, nothing else VW or Audi uses it and its part number is 025 906 265 B. Not saying you can't get a cheaper relacement from somewhere else though... just what comes up on Etka.

MG

Post by **Mocki** » 02 Dec 2005, 08:12

TC is this not the same symptoms as terrys "DESPERATE" fiasco last year?

Post by **Ye Olde Syncrospares** » 02 Dec 2005, 08:57

if you can get it to me you can try a lambda probe from something here,cj.

Post by **syncroandy** » 03 Dec 2005, 08:50

Dh, I've very occasionally had very similar symptoms with my MV (misfiring, won't idle).

However mine only ever occur after a longish, fastish run. I would initially clear the fault by backing right off the speed, after another 20 miles the symptoms would then clear.

They started after I had the JP exhaust fitted, in which a cheap Thompson lambda probe was fitted.

I've been running with the probe disconnected for about a year. You can do this, the Digifant ECU will simply run in open-loop mode on it's standard map, as it does on A2 Golfs. There's no appreciable drop in economy.

I've since learned I can immediately clear the condition by switching off then disconnecting the battery pos, which clears the ECU memory.

A few weeks ago I used a borrowed Bosch probe on a longish fastish run, and no problems occurred. I've since ordered a new probe from VW (£83 + vat

).

My advice would be to run with the probe diconnected, if you still get the symptoms, then look elsewhere.

PS. The MV/Digifant workshop manual is here:
http://www.obsolyte.com/vanagon/Vanagon ... /DigiFant/

Post by **Louey** » 03 Dec 2005, 09:29

Harrymann off the old ez board

lets pretend the engine is your body and I'm a homeopathic doctor (adds a bit of interest anyway).

What away with words you have, sorry Tom I can't help

Post by **Diamond Hell** » 03 Dec 2005, 21:43

Cheers Andy, good to have advice from those experienced with this engine and its quirks.

I think I'll run without the lambda up to the MOT and see if fitting a new one around then cures it. It's another opportunity to try out Vee.W Services and see what they can do on price!

Post by **HarryMann** » 04 Dec 2005, 01:12

Many vanagon.com posts over the years have shed doubt on non-OEM lambda probes in these engines being far less reliable, some being as good as useless. They do need to be in good nick. Maybe its the ECU circuitry being fussy or positioning (hence the change in exhaust design affecting it as Andy suggests)
Also seem to think the wiring runs a shielded outer (earthed only at one end to avoid loops?).

And yes, lambda probes should really not be in the direct exhaust stream as are relatively delicate transducers, not supposed to even touch the tip of the thing some say.

Post by **Aidan** » 06 Dec 2005, 22:26

I had a similar problem with mine, on a warm start it would start then go erm no I don't want to run I don't like it, idle down to 500rpm, and then either stall (and be ok on restart) or catch itself and then run ok. Again it was the JP exhaust with boss in the pipe and a Thompson probe. When tested it could be seen that the signal from the probe was all over the place, eventually the ecu decides to ignore it and goes into overide mode and runs ok. Problem appears to have been solved by fitting a new genuine one - only available from VAG apparantly, my local Bosch main agents couldn't supply.
A similar misfire issue appeared on mine just before the internals of the disi tore themselves to bits this time last year - in that case the vane unit was catching the hall sensor we think, possibly due to wear on the disi shaft and particularly the wear on the fibre spacer washers in the assembly. When disembled there wasn't much left to see of the hall unit but the rotor arm and cap were in good condition.
If the van runs ok with lambda signal off, and very insulated (I cable tied it to the outside of the rubber insulator and there was enough signal trickle to confuse the ecu) then the problem must be that I guess. It should run ok without it, just a bit richer than optimal. You will still be able to adjust the idle and co before MOT and get it through without th cat.