Page 1 of 1

Lambda sensor- heater wiring...

Posted: Mon May 04, 2009 11:08 am
by Quagmire
Guys,

Am fitting the closed loop LPG kit to my v8 Defender and have a couple of questions about the lambda sensor... It is a VW/Audi bosch unit that looks like this:

Image

It has four wires, white, white, grey and black. From googling around it seems that the grey is likely to be the earth, the black the signal and the white wires the heater connections. Can anyone confirm this?

Also, i was planning to use a relay to switch the heater connection... will the N00AW do as listed here on the Maplin page? Looks ok to me. I will then have it switching on only when on LPG, or am I better having the heater on the whole time the car is running (whatever the fuel) to prevent fouling of the sensor?

Maplin

As a final question, is it worth installing the kit so that i can turn it on and check my petrol emissions too? How useful would it really be to have an indication of rich or lean when at idle- it could be doing all kinds of other things when i'm actually on the move...

Thanks! :D

Posted: Mon May 04, 2009 3:22 pm
by ChrisJC
As far as I know (and I do know for the Bosch LSU4.2 wideband sensor), you cannot just hook the heater wire up to +12V. What you are supposed to do is to monitor the sensor temperature (in some complicated way by measuring the AC resistance of the sensor element) and adjust the heater current accordingly.........

For my Landie, I purchased an Innovate LM1 which does all the clever stuff. There are various DIY circuit about as well, but I don't know how accurate they are.

Chris.

Posted: Mon May 04, 2009 3:29 pm
by bones
The 2 white wires are heater, Black is signal out ,and grey is signal common. :? rich

Posted: Mon May 04, 2009 5:17 pm
by katanaman
Being narrow band lambda bones has confirmed your wiring and that relay will be fine to power it although don't fuse it at 40A. I cant remember what they take but I don't think its anywhere near that. A gauge is probably not going to be much use to you being narrow band. Its probably going to be even less use running on LPG, if I remember right they run at different AFR's but I could be wrong on that.

Posted: Mon May 04, 2009 6:23 pm
by Quagmire
This is the place to post if you need answers! Have an identical post on LR4x4 and no-one has even touched it yet...
katanaman wrote:A gauge is probably not going to be much use to you being narrow band. Its probably going to be even less use running on LPG, if I remember right they run at different AFR's but I could be wrong on that.
The box is pretty primitive, just a couple of LED's for rich or lean. Was just wondering if it might be useful for giving me a rough pointer as to what mixture the petrol is running at- think its a little rich at the mo. Hadn't thought about the possibility of different AFR's, but assume that as the probe was supplied with the LPG closed loop kit that it must be up to the job and therefore LPG and Petrol have the same or similar AFR requirements...

**Update** LPG stoich is apparently 15.5:1 or thereabouts:


LPG info

Posted: Mon May 04, 2009 8:19 pm
by Eliot
ChrisJC wrote:As far as I know (and I do know for the Bosch LSU4.2 wideband sensor), you cannot just hook the heater wire up to +12V. What you are supposed to do is to monitor the sensor temperature (in some complicated way by measuring the AC resistance of the sensor element) and adjust the heater current accordingly.........

For my Landie, I purchased an Innovate LM1 which does all the clever stuff. There are various DIY circuit about as well, but I don't know how accurate they are.

Chris.
For a wideband - yes. For a narrowband, just shove 12v down whichever two wires are the same colour (assuming 4 wires).

Posted: Tue May 05, 2009 7:05 am
by ChrisJC
LPG stoichiometric is indeed different to petrol, but it doesn't matter. The sensor is measuring oxygen, and you are still trying to achieve the same oxygen content in the exhaust with either LPG or petrol.

So you just ignore whatever fuel you are running on, and pretend it's petrol. If the sensor readout is lambda, then it doesn't matter, if it's AFR then treat it as petrol.

Chris.

Posted: Tue May 05, 2009 1:18 pm
by Eliot
ChrisJC wrote:LPG stoichiometric is indeed different to petrol, but it doesn't matter. The sensor is measuring oxygen, and you are still trying to achieve the same oxygen content in the exhaust with either LPG or petrol.

So you just ignore whatever fuel you are running on, and pretend it's petrol. If the sensor readout is lambda, then it doesn't matter, if it's AFR then treat it as petrol.

Chris.
OP has a narrowband - which is tuned for a lambda of 1 (~14.7afr) - So i'm not sure if it will give meaningful results when running gas.

Posted: Tue May 05, 2009 2:00 pm
by ChrisJC
Doesn't matter. The narrowband is looking for a certain percentage of oxygen in the exhaust stream. This is the same regardless of fuel type being burned. It takes a bit less propane to get the same oxygen content in the exhaust compared to petrol.

Chris.

Posted: Tue May 05, 2009 2:14 pm
by Eliot
ok I get it now - I'm thinking wideband, but as you say its narrow band and just looking for Lambda - which is stoic for any fuel.

Posted: Tue May 05, 2009 2:28 pm
by ChrisJC
That's right. If lambda=1, then you've got a stochiometric ratio of fuel and air going in. What that ratio is doesn't actually matter.

Chris.

Posted: Tue May 05, 2009 4:30 pm
by katanaman
so 14.7 is the right mixture for both LPG and petrol? I know its that for petrol but thought it was different for LPG for some reason.

Posted: Tue May 05, 2009 6:54 pm
by ChrisJC
katanaman wrote:so 14.7 is the right mixture for both LPG and petrol? I know its that for petrol but thought it was different for LPG for some reason.
No. The sensor measures lambda, then multiplies the number by 14.7 to give you AFR on the display. It is not measuring AFR (that is an important point). It is measuring oxygen content of the exhaust and inferring AFR from that. If you changed the software so it multiplied by 15.5 (or whatever the number is), it would be 'calibrated for propane'.

Chris.

Posted: Wed May 06, 2009 4:28 pm
by katanaman
ok were talking different things, I was thinking just a simple gauge which cant be altered where your talking about something more complex.

Posted: Wed May 06, 2009 9:15 pm
by Quagmire
Well all the electrics are fitted and the box is working, just need to fit the stepper valve to the gas vapour line. Unfortunately ran out of daylight and then maglite went flat so had to pack it in for the evening. :cry:

Heater on the probe makes a big difference it seems, as without it connected the box always shows a green light (lean) no matter what happens- even with petrol AND lpg running through. :shock:

Hooked the lambda heater connections up via relay triggered by a switched live so that the probe stays clean whenever engine running no matter what fuel is being used, also means the probe is hot already when you change to LPG after starting on petrol.

With the stepper connected electrically but not installed you can watch it opening and closing as you vary the revs and it tries to alter the mixture. Idle shows as being rich (red), periodically going to lean (green) so must be pretty close to ideal anway. Was too late to try any revs without peeing off the neighbours :lol:

Tomorrow i'll get the valve in and see what happens!

Thanks for all the replies, didnt expect topic to generate as much conversation as it did :D