Author Topic: testing a lambda probe  (Read 6153 times)

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
testing a lambda probe
« on: June 01, 2009, 12:11:42 pm »
I'm having what appear to be some fuelling issues off-boost on my turbo and I'm suspecting the lambda probe is the issue. 

I do have an afrm on the car but I consider these a pretty inaccurate device at best.  However the behaviour does seem a bit odd.  When pulling up to a stop the engine returns to idle and normally I'd expect the afrm to cycle between rich and lean.  In my case the gauge reads full lean for a few seconds and then begins to slightly cycle between full lean and only a quarter of the way between rich and lean. 

When accelerating off, below about 2.5k rpm the car hesitates massively and you can hear lots of popping and banging from the exhaust which suggests overfuelling but the afrm reads lean.  Once boost builds up and revs get over 2.5k rpm it suddenly takes off like a rocket and the afrm reads about 14:1 which seems a bit lean. 

Hopefully I'm borrowing a spare probe from someone so I can give that a try rather than going out and just buying a new one.  The one on the car is only a couple of years old and has probably done less than 4k miles.  But the previous engine in the car was burning oil last time I used it so perhaps that's damaged it.

Is there any real way of testing a lambda probe other than simply swapping it out? Or has anyone got any other ideas?  Maybe I could see how it runs with the probe simply unplugged?

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
Re: testing a lambda probe
« Reply #1 on: June 01, 2009, 12:14:55 pm »
oh a couple more things,

When pulling to a stop if I suddenly lift off the accelerator the revs will go too low and the engine cuts right out although it always bump starts right away if I lift the clutch up with it still rolling.  If I use the throttle to make the revs fall slower then it wont cut out and idles ok.

I have put the lambda probe boss on the turbo downpipe in the engine bay probably less than a foot away from the turbo outlet, is it possible that the probe is overheating?

Offline Robin

  • Members
  • *****
  • Posts: 1178
  • 194bhp - 1341 Turbo
Re: testing a lambda probe
« Reply #2 on: June 01, 2009, 12:55:38 pm »
Hayes is your ARFM a wideband one?

If you need a spare probe give me a shout as i have one you can borrow at the unit.

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
Re: testing a lambda probe
« Reply #3 on: June 01, 2009, 01:02:10 pm »
no it's just a normal one connected to the normal probe.  I do intend on getting a wideband at some point but I don't have the money right now.


Offline Robin

  • Members
  • *****
  • Posts: 1178
  • 194bhp - 1341 Turbo
Re: testing a lambda probe
« Reply #4 on: June 01, 2009, 01:05:27 pm »
I would lend you my wideband to use but all the wiring is loom taped up to the car loom.

The one you got now running off the narrow band is very very hit and miss, get those pennies out and get a wideband!!!

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
Re: testing a lambda probe
« Reply #5 on: June 01, 2009, 01:35:05 pm »
yes indeed, I know the narrowband ones are hit & miss.  I really can't afford a wideband right now.  I'll try and borrow a normal narrowband one off someone which I know works and see what happens.

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
Re: testing a lambda probe
« Reply #6 on: June 01, 2009, 08:33:43 pm »
right then, seem to have got somewhere with this but the lambada probe isn't the issue.

DK kindly lent me a spare one he had and it made no difference.  Which lead me to believe that the afrm was right and it is lean, removed a plug which was white which also suggests lean running.

Then happened to be talking to yoof so mentioned it to him and he said it sounded like something that happened to him once and have I got other things t'd into the fuel pressure reg vacuum pipe. Well I had the dump valve pipe, boost gauges and the boost switch signal line both t'd into it.  So disconnected that lot and directly connected the fpr to the servo pipe.  Drove up and down the street keeping off boost (since the pressure switch isn't connected) and it seems loads better. 

So now need to re-plumb in the DV and boost switch and my boost gauge (although to be fair that's been t'd into that line for years).  It would appear the issue isn't electrical but vacuum related!

Offline Yoof

  • Traders
  • *****
  • Posts: 2355
  • I know naathing..
    • Polo Performance Parts
Re: testing a lambda probe
« Reply #7 on: June 01, 2009, 10:59:30 pm »
Easy when you know how  ;D


Offline mstar

  • Members
  • *****
  • Posts: 39
Re: testing a lambda probe
« Reply #8 on: July 07, 2009, 10:06:19 pm »
Hayesey, how you doing on with sorting out the problem in the vacuum circuit? What can be broken in the vacuum circuit besides a leaking pipe/hose?

My G40 failed Co2 testing (12%) and is running rich all the time.
As I understand, the lambda voltage is low i.e. 0,1 volt when running lean(much oxygen in exhaust) and up to 0,8 volt  when running rich(less oxygen in exhaust).

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
Re: testing a lambda probe
« Reply #9 on: July 08, 2009, 10:05:00 am »
our problems aren't going to be the same thing.  Mine is converted to turbo, I replumbed the vacuum pipes so that the dump valve isn't on the same bit of vacuum hose as the fuel pressure reg.

But I think most of my issues were down to needing remapping, we've done some basic remapping now and it's a lot better.  Hopefully doing more soon.


Offline bmxthrash

  • Members
  • *****
  • Posts: 37
Re: testing a lambda probe
« Reply #10 on: May 10, 2011, 03:57:28 pm »
i know yours in turbo'd hayesey but this is pretty much identical to what mines doing but running the charger!

was it just the vacuum that was the problem with this?! few years old i know but i hope you can remember  :P

cheers

Offline hayesey

  • Administrators
  • *****
  • Posts: 5824
  • The CV joint killer
    • ClubG40
Re: testing a lambda probe
« Reply #11 on: May 10, 2011, 04:21:40 pm »
no it was mainly (or possibly completely) down to needing remapping.  But I do have the DV plumbed into the bottom of the inlet manifold rather than sharing the vac pipe with the fuel pressure regulator.  None of these things are going to be your problem though I'm sure

Offline bmxthrash

  • Members
  • *****
  • Posts: 37
Re: testing a lambda probe
« Reply #12 on: May 10, 2011, 05:18:41 pm »
i hope so, cant really afford a remap! :-[