xantia- error code number 31

_kid_ · Unread post by **_kid_** » 03 Jun 2008, 16:30

My xantia is using to much fuel, i had it on a cit diagnostic tool and came up with code no 31- this has about ten items which could be the fault- from ignition system to wiring to air filter and spark plugs (which I have changed recently- so these are ok). one of the other options was lambda sensor.

The car drives absolutely fine, the last mot shows emissions fine- however the garage think based on the list of potential items on code 31 and the fact the car drives perfect the most likely fault item is the lambda sensor, does anyone have any experience or ideas regarding this code or lambda sensor theory?

Btw the engine management light use to take four seconds to go off when the engine was started and use to remain off. Now the error code has been removed the light goes off immediately- this may or may not be relevant.

All help appreciated

AndersDK · Unread post by **AndersDK** » 03 Jun 2008, 20:16

Cant be sure on the Lambda at fault or not.

But its a fact that the lambda sensor is at work during idle, low revs and overrun - ONLY.
When the engine is loaded - even lightly at cruise or just coasting - the engine management system is working in "open loop" - not limiting the fuel usage. That would make absolutely no sense if you need to compensate going uphill or overtaking.

iIf you have any experience with the earlier carburetted engines - and their typical problems - then think back to the type of faults that caused bad mileage.
It would be exactly the same on a management controlled engine, except the functions got new names as well as parts are different.

1) Choker constantly active = air or coolant sensor (or related cabling) defective.
2) vacuum ignition advance pod inactive = MAP sensor defective
3) carburettor float valve leaking = drooling injector(s)
4) idle mixture too rich = check 1) or Lambda sensor

Standard failures :
1) weak ignition coil causes intermittent ignition usually with a hot coil only.
2) clogged air intake filter (often soaked in engine oil by the sump fumes hoses)
3) bad engine oil - even wrong viscosity (10W40 is the factory recommandation, semisynt or mineral)
4) engine running too cool = defective thermostat (stuck open, common problem)
5) bad HT cabling arcing over = test engine compartment in complete darkness, look for arcing
6) driver is too brisk on his right foot. Tends to run the engine too high on revs. = ease your foot, keep revs within 2000-3800 at all times. learn how to coast/cruise.

MikeT · Unread post by **MikeT** » 03 Jun 2008, 20:56

I think the EML going off immediately is because no faults are recorded - due to them being cleared. If it doesn't return, then I would it ignore it for now.

But if you do want to know a little about the oxygen sensor (lambda is a misnomer because the sensor only measures oxygen content and nothing else). It compares the amount the oxygen in the exhaust with that in the atmosphere and creates a small voltage (once up to temperature) relatively which is sent to the engine management system all the time the engine is running or the ignition is on.

The engine management system will ignore it during wide open throttle and decceleration - perhaps at idle too.

They are easy to test if you have a multimeter but I wouldn't go chasing ghosts if I were you. If you're using too much fuel and can be sure it's not heavy-foot syndrome you're better off starting at the beginning and checking the basics as Anders lists - tyres, brakes, service items etc.

DickieG · Unread post by **DickieG** » 03 Jun 2008, 21:51

Fault code 31 is a 'catch all' code relating to poor fuel regulation without specifically pointing to any one item although in the Citroën diagnostic manual its main suggestion is the Lambda Sensor. Have a read through the folllowing link for some good info on the Lambda Sensor http://www.picotech.com/auto/lambda_sensor.html

_kid_ · Unread post by **_kid_** » 04 Jun 2008, 00:55

thanks for the replys

i should also add according to the reader there was a misfire above 2000rpm (although i have never noticed) and the mixture was too rich, however it has passed the mot emissions fine inc lambda.

i changed the cts last year and the ohms reading are line line with what to expect.

MikeT · Unread post by **MikeT** » 04 Jun 2008, 18:33

What condition were the old spark plugs in and have you checked the current ones?

_kid_ · Unread post by **_kid_** » 04 Jun 2008, 18:44

old sparks were tannish brown colour and current ones are similar, so i would think they are ok.

MikeT · Unread post by **MikeT** » 04 Jun 2008, 19:18

No obvious signs of carbon fouling then?

I've been outvoted on whether to suspect the oxygen sensor right now so maybe you should consider testing it? Using a multimeter, you can read the volatage inputs and outputs.

_kid_ · Unread post by **_kid_** » 05 Jun 2008, 00:26

i think thats a good place to start mike, what is the testing procedere? (i have the digi multimeter) just need the instructions

also one other question, the emissions on co and hc and lambda were all well within limits at the last mot, so is it possible to have a lambda sensor which operates satisfactory to pass mot, yet still not be fully efficient thus drinking more fuel? if this is the case where is all the fuel going?

MikeT · Unread post by **MikeT** » 05 Jun 2008, 12:39

There are differing methods for testing (DickieG's link is one such example) but I chose to tap into the output wiring (using scotchlocks and speaker wire) so that I could read the output while driving.

You'll have to do a little detective work to identify the wires but it's not difficult. Trace the wires from the sensor to the connector. Test the sensor input wire(s) for a 12V feed using 20VDC range and test the output wires for a low voltage range that fluctuates rapidly (0.0v to 1.0v), using the 2000mVDC range. During normal cruising or part throttle you should see the output rapidly switching around 0.2v to 0.8v

Meanwhile, deccelerating shows a constant 0.0v (very lean) and wide open throttle gives about 0.9v (very rich). The reason these two extremes do not fluctuate rapidly is because the ECU will not correct off-load or full-load requirements.

If there is no output or it doesn't fluctuate radpidly, it's either not reached operating temperature (more revs!) or it's faulty. They generally fail slowly, getting more and more sluggish in their ability to respond. If your engine is running rich the average reading would be above 0.45v. HTH but if it still leaves you with questions, just ask.

In answer to your second question, I couldn't say for sure as both the emissions test and your spark plugs don't show a rich mixture. Have you got the MOT emissions test printout?

_kid_ · Unread post by **_kid_** » 05 Jun 2008, 15:28

the results are as follows

07/2007

fast idle
co 0.034
hc 55ppm
lambda 0.999

natual idle
co 0.019

2006

fast idle
co 0.08
hc 31
lambda 1.00

natual idle
co 0.1

MikeT · Unread post by **MikeT** » 05 Jun 2008, 16:11

That does look marginally rich IMO.

If your MOT expires next month you might as well wait for the next test results - I think you can have it tested one month before expiry so you'll have at least a month to get any failures fixed.

DickieG · Unread post by **DickieG** » 07 Jun 2008, 00:48

I've been having a look through the 'Info rapid' (service bulletins) for Xantia's and discovered no 84 relates to such a problem as this, the solution is to download a software update from a dealer/someone with a Proxia/Lexia. Whereabouts are you?