Lambda Sensor replacement - is it the right route?

Hi All
My first post here, I have been skulking in the background for a while now and found some great support and answers already.

I have tried to search for this but I cannot find the right answer so here goes

I have a P38 2001 4.6l Which has StarGas Multipoint LPG conversion.
A few weeks back the car started to run rough while on the motorway going from 50 to 70mph - it felt like I had suddenly picked up some dirty fuel.
SInce then it has got worse on Petrol
Now when I start it up on petrol it idles rough, apart from the minute or so from cold (it is currrently freezing here in Barnsley), When you apply some revs and set of the engine pops and backfires and lacks power - the only way around it is to floor the accelerator pedal but it is still rough. When the LPG flicks over it is instantly better but it does feel jittery at speed as if stuttering.
When the car has been running for about half an hour the LPG flicks off with a red flashing light and we are back to petrol.

Initially I was advised to test the MAF and indeed it was dirty so I cleaned it. This made the LPG feel better but still dog rough on the petrol.
I have had the HT leads changed and they were good quality - no benefit sadly.

I invested in the Faultmate FCR and have constituently got the same errors (Petrol or LPG )
Here are the errors
Bosch Motronics
P1300 Catalyst damaging misfire on multiple cylinders Drive Cycle C - Signal too high , Currently present, Intermittent, causes MIL light Activation
P1000 Catalyst damaging misfire on cylinder 8 Drive Cycle C - Signal too high , Currently present, Intermittant, causes MIL light Activation
P0304 Catalyst damaging misfire on cylinder 4 Drive Cycle C - Signal too high , Currently present, Intermittant, causes MIL light Activation
P1000 Catalyst damaging misfire on cylinder 6 Drive Cycle C - Signal too high , Currently present, Intermittent, causes MIL light Activation
P0302 Catalyst damaging misfire on cylinder 2 Drive Cycle C - Signal too high , Currently present, Intermittent, causes MIL light Activation
P1315 Misfire with low / empty Fuel drive cycle C - signal too high , currently present , persistently present
P0300 Emission relative misfire multiple cylinders drive cycle c, signal too high , currently present , persistently present
P0308 Emission relative misfire cylinder 8 drive cycle c, signal too high , currently present , persistently present
P1000 Lambda Heater Upstream catalyst bank 2 Drive cycle C, Signal too high , signal missing, currently present , intermittent, causing MIL Lamp ACtivation

no other errors on the other modules - thank god

So from the above am I right in thinking that the even cylinders are infact all on the same side of the engine and that the Lambda sensor to bank 2 is the same side, and would the above suggest that it is the sensor that is failing?

If anyone can shed light are suggest my next step (so long as it doesn't include a match and a can of petrol.... although )
:doh:

I'd be changing the coil pack for that side first. Or even both as they aren't that dear. I don't have lpg but did have the cat damaging misfire faults, changed one coil then three months later did the other one. You have the Thor engine with platinum plugs which can take their toll on the leads however as your prob is down the one side look to the coil first. Eliminate faults as you go. There are three different kinds of lambda sensor for the P38 range, make sure you get the right one if you need to do them too.

hippolover said:

I'd be changing the coil pack for that side first. Or even both as they aren't that dear. I don't have lpg but did have the cat damaging misfire faults, changed one coil then three months later did the other one. You have the Thor engine with platinum plugs which can take their toll on the leads however as your prob is down the one side look to the coil first. Eliminate faults as you go. There are three different kinds of lambda sensor for the P38 range, make sure you get the right one if you need to do them too.

Click to expand...

Thanks for the answer.
So it would be ok to replace the Lambda sensor and coil pack on the one side in one go as they are not that expensive?
Is it a simply case of finding them on the car and a couple of bolts out etc and replace in reverse (I am a DIY disaster - enthusiastic but usually starts with one job and ends up with two more)

The coils on your car are at the back of the V and actually down the back of the engine. (The slightly older Gems model leads poke out the top at the back).They are difficult to get at and what you must not do is take off all the leads at once - even to one of the coils. Note down which one goes where or draw a picture as its critical they go back in the right order. The lambda sensor is on the exhaust down pipe near the cat. Leave that til last and see how the coil goes first. The sensors are dear unless you land lucky at a breakers. (The coil packs are square units rather than what you normally know as a coil which is the long round thing on older cars!)

First off ignore that platinum plug gumf, you need NGK BPR6ES plugs, nice and cheap and plenty good enough for your engine. The platinums may have had some benefit on the older RR with coil ignition, but not on your P38. The coils are a bitch to change and I would recommend doing both once you get to them. It is not helped by having LPG either as I imagine there is plenty of pipes in the way on top of the plenum (that has to come off) I did a write up on my LPG install and there are asome good pictures of the process to get the plenum off and get to the coils. Have a look here.
The O2 sensors are easier to do, but go prepared with a Stillson wrench.

Dont buy cheap anything from eBay !!!!!!! Use OE parts.

I wouldnt worry about the coils or oxy sensor,look at the leads/plugs first.I've given a bit more detail on the other forum you posted this thread on.

Bit of a coincidence that it's all down the one side though, like I say I'd suspect leads and plugs only if both sides were showing faults. By all means do the plugs (platigumfs or standard) and leads too so it's entirely personal choice. When mine went all misfiring I changed the plugs to (BPR6ES) and put on new leads but the problem persisted. Re-fitted the platinums, original leads checked again for faults which were still there then changed the coils. The new BPR6s went in the classic so no waste. One thing though in concession I think the coils do two cylinders from each side so it could be something to do with the lpg delivery instead.

Thanks guys
I have already changed the leads with good ones but the spark plugs looked good when I removed them (can't remember what they were mind)

I will try the coil packs first as they are inexpensive- if it is a dog to get to I will do both of them coz I know the other one will go next week.

I will let you know what happens.

Lambda wise is there any benefit from disconnecting it ? What effect would it have?

As I said in the last post I could could be leaning towards fuel delivery as the coils I'm sure each go to two cylinders on either side so if your prob is all down the one side then one side is not getting fuel. The sensors each control a bank of cylinders left and right so what you could do is swop them round (careful not to break them) and see if the prob migrates to 1,3,5 and 7. If it does then you've found it.

AckAckFlack said:

Thanks guys
I have already changed the leads with good ones but the spark plugs looked good when I removed them (can't remember what they were mind)

I will try the coil packs first as they are inexpensive- if it is a dog to get to I will do both of them coz I know the other one will go next week.

I will let you know what happens.

Lambda wise is there any benefit from disconnecting it ? What effect would it have?

Click to expand...

So did you read my post on RRnet ? Visual inspection of plugs can mean very little,at least I'd change the plugs on the rh bank.One plug on one bank with a partial misfire is enough to cause your problem.Can you view misfire counts as live data with your faultmate ? If you can then you are halfway there,a fuel issue is VERY unlikely and the Bosch coilpacks are reliable unless the tops have burnt of as a result of tracking because of a failed plug or lead.
Does the oxy sensor switch at all in live data ?

eightinavee said:

So did you read my post on RRnet ? Visual inspection of plugs can mean very little,at least I'd change the plugs on the rh bank.One plug on one bank with a partial misfire is enough to cause your problem.Can you view misfire counts as live data with your faultmate ? If you can then you are halfway there,a fuel issue is VERY unlikely and the Bosch coilpacks are reliable unless the tops have burnt of as a result of tracking because of a failed plug or lead.
Does the oxy sensor switch at all in live data ?

Click to expand...

Late last night I change 2 of the plugs over from left to right ie plug 8 to 7 and plug 6 to 5
The passenger side plugs (even ones) tips looked the same - nice and sandy colour but the lead edge of the threads on the even ones looked blacker but not a big difference.
I cleared the faults and fired it up- it was already warm. And let it idle fo a few minutes - checked with faultmate and no errors.... Was getting giddy at this point but when I took it for a run on petrol it was Miss firing and when I got back the errors as before where there. As it was 12:30 in the morning and -2 I decided to call it a day.

I found the coil packs - they look a begged to remove because of the popes for the LPG so I am not looking forward to changing these- I was told that the coil packs don't control one side of the engine but more of
Coil pack 1 = cyl 1 ,2 ,5,6
Coil pack 2 = 3,4 , 7,8
Not sure ifthat is correct.

I plan to change over the other plugs this evening just incase it is one of the remaining ones causing problems.
As it was dark I did have a look for sparking from the leads but I couldnt see anything (they were changed when this happened first so the leads are new)

The problem of misfiring is definaty worse when cold and gets less noticeable when warmer , doesn't seem to happen when warm and idleing
Always gives the errors in the same(ish) order on faultmate - not sure if the car is showing me in time order.

Faultmate doesn't show me live data I am afraid- I might have to get someone with textbook to sus that

You really need to see that live data,watching the oxy sensor outputs,misfire counts and short term trims would be a good start.Will your Faultmate do an adaptive reset ? If the fuel trims have been upset from the misfire it could just be trimmed to lean for it to work properly - even if the leads and plugs are now sorted.

eightinavee said:

You really need to see that live data,watching the oxy sensor outputs,misfire counts and short term trims would be a good start.Will your Faultmate do an adaptive reset ? If the fuel trims have been upset from the misfire it could just be trimmed to lean for it to work properly - even if the leads and plugs are now sorted.

Click to expand...

Sadly the faultmate won't do that... I know now I should have saved up a bit longer and got the msv version which you connect to the computer

Would the adaptive values go off on ome side of the engine?

I think it will be a trip to the testbook techie guy soon (that will use up the extra £200 I should have spent on the msv

Blackbox Solutions Ltd. -- Products Page

Like the bike btw

Its always a compromise with diagnostics.....
Yes the trims will go off on one side,a decent live data session should give some answers fairly quickly,I usually say that most get sorted in a 1/2 hr session.Few get into a second hour with a problem like yours inc replacing leads/plugs etc.Shouldnt really cost too much.
The bikes are my main interest,I'll stick a pic of my HT5 Ariel on me avatar soon.BTW,the xrv750 is abit of a favourite of mine,I like v twins,Vincent,Brough,Matchless and BSA etc - but cant justify one.Just agreed a price to sell my MHR Ducati.

eightinavee said:

Its always a compromise with diagnostics.....
Yes the trims will go off on one side,a decent live data session should give some answers fairly quickly,I usually say that most get sorted in a 1/2 hr session.Few get into a second hour with a problem like yours inc replacing leads/plugs etc.Shouldnt really cost too much.
The bikes are my main interest,I'll stick a pic of my HT5 Ariel on me avatar soon.BTW,the xrv750 is abit of a favourite of mine,I like v twins,Vincent,Brough,Matchless and BSA etc - but cant justify one.Just agreed a price to sell my MHR Ducati.

Click to expand...

Certainly can't harm getting some data even if it is to eliminate it from my enquries

The logic to the values being out and effecting one side makes sense along with the fact it runs fine when cold rich idling.
I love my bikes too. I do about 15k miles a year between them.
I am loving the Dakar at the moment and currently scheming a way of getting there next year - on the Africa twin offourse

Just an update.
Finally found someone locally to help with a diagnosis. They had a snap on diagnostic kit.

It brought up the errors that I had just about and when he showed the live feed while the engine was running the only thing that appeared out of the ordinary was the Lambda values

From memory Bank one around 0.95 Bank two 5.01 (not sure what the units where)
When he revved the engine the Bank one values changed but the bank two one remained at the same value - I presume this means it is knackered so I am going to get one and change it.

Am I right in thinking that bank two is the passenger side of the car?

Thanks for the help guys

So it was the lambda then. Bank 1 is the left hand or nearside. Whichever one of the sensors is not fluctuating is the defective one. Change it and re run the diagnostic. Remember there are different sensors depending on the car's age.

hippolover said:

So it was the lambda then. Bank 1 is the left hand or nearside. Whichever one of the sensors is not fluctuating is the defective one. Change it and re run the diagnostic. Remember there are different sensors depending on the car's age.

Click to expand...

I have just ordered one from Quality Land Rover Parts Spares and Accessories Online
I think I have the right one but there were 2 listed for mine

Now a bit confused about bank and cyl numbers

If I am stood infront of the car and looking towards the engine (so my offside is to my left) is it true that cyl 1 is furthest away and left. And then cyl 2 is furthest and right? So all odd cyl on left and even on right.

Also bank 1 is odd ones and bank 2 even

As my problem is even then I need to change the right one - nearside?

Sorry if this is a dense question..... It's just my character

You're kind of there.

Cylinders are numbered from the front to back.

1,3,5,7 are on the EAS block side - Bank 1
2,4,6,8 are on the Expansion tank side - Bank 2.

Whichever sensor is giving a fixed reading needs changed and corresponds to the same as the cylinder bank i.e. 1 or 2.

Be prepared for the ssame thing happening again in time though as experience shows that if you change one of a pair then sure as fate the other one will follow shortly afterwards. I had this on a 1997 P38. Granted I put on a second hand sensor the first time but I took the pair from the donor car when the problem first showed up. It took a further two years for the second one to go but I still had the spare.

hippolover said:

You're kind of there.

Cylinders are numbered from the front to back.

1,3,5,7 are on the EAS block side - Bank 1
2,4,6,8 are on the Expansion tank side - Bank 2.

Whichever sensor is giving a fixed reading needs changed and corresponds to the same as the cylinder bank i.e. 1 or 2.

Be prepared for the ssame thing happening again in time though as experience shows that if you change one of a pair then sure as fate the other one will follow shortly afterwards. I had this on a 1997 P38. Granted I put on a second hand sensor the first time but I took the pair from the donor car when the problem first showed up. It took a further two years for the second one to go but I still had the spare.

Click to expand...

Thanks for the input.
I nearly did order both sensors as I agree with your logic but as I havnt fitted one yet to say for certain that is what it was then I decided to save the money. At least I know what it is if it happens again and can always get another - hopefully 2 years from now.
I will update once it's solved thanks again