Mot emissions failure
Discussion
450 failed by some way the CO and HC test today,
The tester said average required v actual were;
CO 0.3 v 8.6
HC 200 v 2640
Lambda 0.9 to 1.1 v 1.6
As mentioned earlier a new stealth cam is in and due for a remap/rr test next week with jools but wondering if its worth holding back until I've fixed this.
Tester thinks its ecu fuel management mix "and i used to build rally cars"
I've had the car 2 years and I'm curious to know if the main cat is in (update: it pushed some bike brake sheath in there and stops at the cat area)! I know exhaust cats are in. Last year the tester was more sympathetic of tvrs.
Fwiw- new sparks are in
So could the ecu or lambda's be misbehaving?
The tester said average required v actual were;
CO 0.3 v 8.6
HC 200 v 2640
Lambda 0.9 to 1.1 v 1.6
As mentioned earlier a new stealth cam is in and due for a remap/rr test next week with jools but wondering if its worth holding back until I've fixed this.
Tester thinks its ecu fuel management mix "and i used to build rally cars"
I've had the car 2 years and I'm curious to know if the main cat is in (update: it pushed some bike brake sheath in there and stops at the cat area)! I know exhaust cats are in. Last year the tester was more sympathetic of tvrs.
Fwiw- new sparks are in
So could the ecu or lambda's be misbehaving?
Edited by Chimp871 on Saturday 23 July 12:17
Chimp871 said:
tested NS lamdba voltage and got a constant 1.0 volt at idle and fluctuating when revving between 0.0 and 0.9 volts, no other numbers.
I'll test OS one tomorrow got cut zip tie first.
What does 1.0 volt mean at idle, rich fueling-knackered probe?
Yes Rich all the time, it wont be a knackered probe, that would be 0 voltsI'll test OS one tomorrow got cut zip tie first.
What does 1.0 volt mean at idle, rich fueling-knackered probe?
Test the other side and if its the same check the AFM output:
Peel back the rubber boot on the airflow meter connector and leave it plugged in to the airflow meter. Set up the digital multimeter to read voltage. Insert the negative probe into the Red/Black wire (sensor ground), and the positive into the Blue/Green wire (Airflow signal).
Turn on the ignition, but do not start the engine. The meter should immediately indicate a reading of approximately 0.3-0.34 Volts after the initial "warm up" spike. Most defective airflow meters will overshoot to 0.8 Volts or higher, and take at least 2 seconds to come down to the correct voltage.
Now start the car- you should get 1.6 to 1.7 volts or there about depending on engine size and idle rpm.
meanwhile..... plugged in rovergauge to check if there's any faults.
The lambda odd is fluctuating around zero to -70 but the lambda even is a constant -100%.
Here's the RG data link and pic below.
https://www.dropbox.com/s/bzte6fs1scfkhdf/2016-07-...
The lambda odd is fluctuating around zero to -70 but the lambda even is a constant -100%.
Here's the RG data link and pic below.
https://www.dropbox.com/s/bzte6fs1scfkhdf/2016-07-...
Thats an odd one seeing as both sides get the same air and same sensor signals. Drop the plugs out and check to see if they are all sooty on one sides- try swapping probes if you can get them out- Possibly a leaky injector on one side- but Ive not seen this fault before. This is worth a read.
http://www.g33.co.uk/images/PDFS/14cux%20faulot%20...
http://www.g33.co.uk/images/PDFS/14cux%20faulot%20...
interesting. Engine has been rebuilt so senors should drop out easily enough. So in theory the issue could be:
Dodgy lambda sensor (but RG shows it does ramp up from 0 to 256)
Misfire - help explain the high HC. FWIW - all plugs are new so possibly injector problem, injector wiring, spark wire (magnatec but don't know age) or extender.
Anything else to add?
Jools has been helping me out saying a high lambda reading of 1.6 doesn't make sense as this is a lean setting not rich. And pointed towards misfire or sensor.
I'll read the even bank sensor tomorrow and then swap it over.
Dodgy lambda sensor (but RG shows it does ramp up from 0 to 256)
Misfire - help explain the high HC. FWIW - all plugs are new so possibly injector problem, injector wiring, spark wire (magnatec but don't know age) or extender.
Anything else to add?
Jools has been helping me out saying a high lambda reading of 1.6 doesn't make sense as this is a lean setting not rich. And pointed towards misfire or sensor.
I'll read the even bank sensor tomorrow and then swap it over.
The screen shot was showing the engine quite cold so I would not put too much trust in the lambda readings until the engine is up to temp with the fans cutting in and out.
On the leaking injector comment I have twice seen the Cyl. 2 injector wiring shorted to the coil mounting bracket so all the injectors on that side were open almost full time as the short was not quite continuous.
Steve
On the leaking injector comment I have twice seen the Cyl. 2 injector wiring shorted to the coil mounting bracket so all the injectors on that side were open almost full time as the short was not quite continuous.
Steve
For the lambdas to give an accurate reading the engine should be above 80c with no protracted idling to ensure the temp is high.
The emissions readings are saying the car is massively overfuelling and 1.6 indicates a very rich mixture, not lean.
With the work you have had done, and to eliminate any possible component failure, I'd have the map and rolling road done first. If you have a dodgy injector, extender (a real possibility - best to get rid if you can) or plug lead, then this will be found during the mapping and tuning.
The emissions readings are saying the car is massively overfuelling and 1.6 indicates a very rich mixture, not lean.
With the work you have had done, and to eliminate any possible component failure, I'd have the map and rolling road done first. If you have a dodgy injector, extender (a real possibility - best to get rid if you can) or plug lead, then this will be found during the mapping and tuning.
I had a similar MOT failure, with the car running very rich.
Turned out to be no spark to plug 8, caused by a coil pack failing. Even on 7 working cylinders the car was perfectly drivable, and it did not sound like it was missing at all. The ECU was over fuelling even more than it needed to as the lambda sensor was picking up the unburnt oxygen, tricking the ECU in to thinking it was running lean. The HC value was high because the test rig at the MOT centre was picking up all the unburnt fuel. I can only assume the coil pack had been failing for some time and I had got used to the slight drop in power. However, putting a new coil pack in transformed the car.
One other point to note. I have a wide band lambda sensor and MS2 ECU. I recently changed the lambda sensor and controller (Innovate MTXL) from a bosch 4.2 to a 4.9, and used the same boss welded on to the manifold. I have been getting lean readings at idle for a while. I cut the boss down by about 3mm, making it a bit shorter, and the readings returned to normal and held steady at idle. The 4.9 sensor is slightly shorter than the 4.2 and I assume the new shorter sensor was not far enough in to the air flow. The 4.2 must have been right on the edge of the limit.
Turned out to be no spark to plug 8, caused by a coil pack failing. Even on 7 working cylinders the car was perfectly drivable, and it did not sound like it was missing at all. The ECU was over fuelling even more than it needed to as the lambda sensor was picking up the unburnt oxygen, tricking the ECU in to thinking it was running lean. The HC value was high because the test rig at the MOT centre was picking up all the unburnt fuel. I can only assume the coil pack had been failing for some time and I had got used to the slight drop in power. However, putting a new coil pack in transformed the car.
One other point to note. I have a wide band lambda sensor and MS2 ECU. I recently changed the lambda sensor and controller (Innovate MTXL) from a bosch 4.2 to a 4.9, and used the same boss welded on to the manifold. I have been getting lean readings at idle for a while. I cut the boss down by about 3mm, making it a bit shorter, and the readings returned to normal and held steady at idle. The 4.9 sensor is slightly shorter than the 4.2 and I assume the new shorter sensor was not far enough in to the air flow. The 4.2 must have been right on the edge of the limit.
Edited by Smokey Boyer on Sunday 24th July 11:37
If you are coming near Jools again (I am 35 miles from Jools) and think your car is de-catted, I can lend you a catted Y piece for the MOT if it would help.
Do you have the standard (rough metal) Y, or the Clive F (shiny stainless) Y?
It does sound to me like some sort of over-fuelling, and all the causes I can think of have been covered above.
I had endless trouble with plug extenders causing misfires last year. I can confirm that the car will run fine on 6 working Cylinders on the open road, you only really notice the 2 misfires when you get awesome pops and bangs in town traffic.
Do you have the standard (rough metal) Y, or the Clive F (shiny stainless) Y?
It does sound to me like some sort of over-fuelling, and all the causes I can think of have been covered above.
I had endless trouble with plug extenders causing misfires last year. I can confirm that the car will run fine on 6 working Cylinders on the open road, you only really notice the 2 misfires when you get awesome pops and bangs in town traffic.
The lambda probes are resistive type, so they take a voltage from the heater supply to produce the output. As the output never goes anywhere near 12 volts, there must be some sort of potential divider inside to limit it to around 1.5 volts when very rich. I suppose if part of the resistve element failed it would be possible to get a rich reading all the time.
update: so removed the both lambda sensors and swapped them over.
End result is they both now trim, hopefully as they should. When I removed the O/S sensor the red/white/black wires on the sensor side were not matched up with the red/white/black wires on the ecu side. I assume this is why the sensor readings were weird.
Took the advice of going for a long run and logged it with roverguage (see link below), from line 100 you'll see the sensor trimming on both sides at speed but have no idea if it's as expected. But I'll the rule out a faulty lambda sensor for now, but still would appreciate your feedback on whether the MOT HC fail would be caused by a non-working lambda sensor (due to incorrect wiring)?
Qbee many thanks for the offer for putting in a y piece. I've the standard one. I'll see what jools says and go from there. BTW - can he do MOT tests?
PS - on RG really weird is only shows road speed at around 33mph when I defo was moving faster.
https://www.dropbox.com/s/uvdmp3z56sogmti/2016-07-...
End result is they both now trim, hopefully as they should. When I removed the O/S sensor the red/white/black wires on the sensor side were not matched up with the red/white/black wires on the ecu side. I assume this is why the sensor readings were weird.
Took the advice of going for a long run and logged it with roverguage (see link below), from line 100 you'll see the sensor trimming on both sides at speed but have no idea if it's as expected. But I'll the rule out a faulty lambda sensor for now, but still would appreciate your feedback on whether the MOT HC fail would be caused by a non-working lambda sensor (due to incorrect wiring)?
Qbee many thanks for the offer for putting in a y piece. I've the standard one. I'll see what jools says and go from there. BTW - can he do MOT tests?
PS - on RG really weird is only shows road speed at around 33mph when I defo was moving faster.
https://www.dropbox.com/s/uvdmp3z56sogmti/2016-07-...
Edited by Chimp871 on Sunday 24th July 22:02
Chimp871 said:
update: so removed the both lambda sensors and swapped them over.
PS - on RG really weird is only shows road speed at around 33mph when I defo was moving faster.
iirc there is a TVR interface box which just tells the ECU if the car is moving or not. It doesn't need any more data than that.PS - on RG really weird is only shows road speed at around 33mph when I defo was moving faster.
Edited by Chimp871 on Sunday 24th July 22:02
Had an emission failure once, then the tester tried the other exhaust and passed it.
He told me there must be something wrong with the RH bank on the V8 !
He obviously hadn't noticed that the manifolds merge into a central pipe before splitting at the back ha ha.
Lesson learnt is that the test equipment isn't always that reliable !
He told me there must be something wrong with the RH bank on the V8 !
He obviously hadn't noticed that the manifolds merge into a central pipe before splitting at the back ha ha.
Lesson learnt is that the test equipment isn't always that reliable !
From the rg screenshot it looks like the ecu is running fuel map 1. From what i understand this is an open loop map - the ecu will ignore any lambda inputs. Standard tvr map for uk car should be 5.
i had exactly the same situstion with my '98 450. Repeated mot probs with emissions. Replaced afm, lots of cataclean pre-mot, and usual faffing around. Finally got sorted with rovergauge and found ecu was running map 1. After much tinkering, and completely by chance, managed to track it down to fuel pump wiring. Previous owner had wired fuel pump to live feed from ignition, bypassing relay, and disconnected the original wiring at the pump. When i put everything back to original the ecu flipped onto map 5 which is the tvr closed loop map where lambda input is used to tune fuel/air mix. No idea why changing the wiring should impact the ecu - lambda heaters are wired on same circuit as the fuel pump, controlled by same relay, so presumably something to do with that. Given up worrying about why - sailed through mot last time round.
i had exactly the same situstion with my '98 450. Repeated mot probs with emissions. Replaced afm, lots of cataclean pre-mot, and usual faffing around. Finally got sorted with rovergauge and found ecu was running map 1. After much tinkering, and completely by chance, managed to track it down to fuel pump wiring. Previous owner had wired fuel pump to live feed from ignition, bypassing relay, and disconnected the original wiring at the pump. When i put everything back to original the ecu flipped onto map 5 which is the tvr closed loop map where lambda input is used to tune fuel/air mix. No idea why changing the wiring should impact the ecu - lambda heaters are wired on same circuit as the fuel pump, controlled by same relay, so presumably something to do with that. Given up worrying about why - sailed through mot last time round.
Gassing Station | Chimaera | Top of Page | What's New | My Stuff