Page 1 of 1

lambda/oxygen sensor testing?

Posted: 04 Apr 2012, 20:59
by bodgit
can these be tested by any reasonable garage? my van-1987 2.1 mv, petrol/lpg seems to run and idle perfectly with the sensor disconnected but will barely run with it connected.i have replaced sensor twice, first time with cheapo from local factors and second time with pukka from reputable dealer. its starting to dawn on even me that replacing this one with another £65 part is a bit of a false economy so am hoping someone here may have have come across this before.the van has been decatted but has intermediate pipe with lambda boss. plugs all look fine so not over fuelling. any tips?

Re: lambda/oxygen sensor testing?

Posted: 04 Apr 2012, 21:37
by AngeloEvs
A narrow band Lambda sensor can be tested in various ways, any reputable garage should be able to do that. I did reply with various methods but re-reading your post suggests that you want it carried out by a qualified person rather than attempting to do it yourself. If you do wish to test it yourself then search google/youtube, you will find a wealth of videos and fact sheets regards testing lambda probes.

Re: lambda/oxygen sensor testing?

Posted: 05 Apr 2012, 06:25
by Aidan
dead simple to test you can pop voltmeter onto 2v dc range and pull back the rubber boot at the join and with your earth probe against a nice earth on the body or the engine (like the bolt the earth strap is connected to) and the other probe against the signal connection watch it vary from one side of 0.7v to the other
However I can say from experience that the MV ECU is very fussy, I don't know why but unless it is absolutely the correct Bosch lambda then it won't be happy, and if you have to cut and shut the wiring connections at all it probably won't be happy
When the s/s exhausts from JP were done as a club discounted batch job three of the syncros on MV's had them and we all fitted the sub £30 Thompson lambda from GSF and they all ran fine for about 750 miles and then the ecu got arsey
solution was genuine bosch and I can report mine is still happy 7 years later
The thompson one tested as working ok but something about the signal or the rate or something the ecu did not like but as soon as genuine one was fitted it was happy again


my subaru on the other hand was quite happy with a cheapo £15 one off ebay rather than the scooby £150 one, but the subaru doesn't go lambda watching till the engine is at full working temp. ie 5 minutes down the road whereas the MV goes lambda after 30 seconds so is effective over a much greater range of engine exhaust temps so I guess it may be something to do with the heating function on the genuine one and the way is accesses the clean air outside the exhaust

oh yes and the exhaust MUST be leak free otherwise the lambda will see a lean mix and richen it accordingly, on the MV exhaust there is a test point just before the cat (if fitted, I run the decat pipe) and the cap of that often corrodes away giving a nice 1cm square air inlet just before the mix is 'sniffed', and the cat can is notorious for leaking at the joins too, there will probably be nothing in the cat by now anyway, all the ones I have had have burnt out and dissappeared, but they pass mot happily without a cat as pre 1993 regs apply

Re: lambda/oxygen sensor testing?

Posted: 05 Apr 2012, 09:02
by California Dreamin
Aidan is right about the quality of any replacement being very important on an MV, I was just going to add that I have also had good experiences with NGK.
Incidently Bosch and NGK are the only two spark plug makes I would recommend as well.
And too also add that my NGK probe has been fitted for nearly 3 years now.

Martin