Rough running at light throttle- Update

[DW199]

Sigma engine in 2019 Academy car.  The car runs very poorly on light throttle at around 3500 rpm.  It has done this ever since I bought it but during my last track day at Goodwood, in the wet the car was almost impossible to drive.  It feels as though the throttle is pressed and released once or twice a second. 

The traces above are from an OBD scanner and logged on my phone.  In the centre of the  trace the lambda sensor voltage goes low without any obvious fuel trim response.  The tps position seems quite noisy. Is this behaviour as expected when the rpm is betweem 3200 and 3800? 

There is a very slight exhaust leak at the 4-1 collector.  Could that account for the behaviour? 

Any suggestions welcome.

Chart below shows the traces with the car stationary  at different rpms

 

[Tom_Arundel]

Sounds a bit like drive train shunt i.e. the rubber/flex in the drivetrain (engine mtgs..gearbox..diff. A frame etc) winding up and releasing.. .. sometimes making your right foot bounce on the pedal. Usual cause is TPS setting error....check voltage and position at closed throttle......but could be many other interesting things!

[GulfSeven]

Could it be the immobiliser relays vibrating and causing the engine to cut out? I had this on my new 310R at 5,000 rpm. For me it felt like losing power briefly before continuing to accelerate. I think Williams changed how they were mounted or added some foam or something to fix it.

[aerobod - near CYYC]

I would ensure that the 4 into 1 collector is properly sealed, as a continuous low voltage from the lambda sensor (shown in the graph) either indicates too much air in the exhaust giving a lean reading, too little fuel, or a faulty sensor as likely causes. In the mid range you could get the most relative air leakage with the exhaust starting to scavenge well, but not too much flow through it to make an air leak negligible.

[GulfSeven]

Out of interest, how is the collector sealed? Mine seems to blow a bit when cold and I assume seals when it expands with heat? It doesn't seem to blow when hot.

[StevehS3]

As an over-winter job I will be using this which was recommended by others. 
 

https://www.ebay.co.uk/itm/251646614287?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=Jf6wyxtgQu6&sssrc=2349624&ssuid=N1mAWR0DS2q&var=&widget_ver=artemis&media=COPY

[DW199]

Tom_Arundel - The excessive driveline shunt that afflicts these cars certainly exacerbates the fault but I don't think it is the root cause.  I was careful to brace firmly against the side  of the footwell to minimise that effect.  The tps has been reset several times by Caterham Gatwick which solved the poor idling and immediate off idle progression problem.  I think this is a different fault.

GulfSeven - That's interesting.  There is a slight hum from under the dash when the ignition is on.  I will investigate further.  The collector seems to be an interference fit.  I can't see any evidence of sealant.

aerobod - The thing that is puzzling me is that the short term fuel trim does not seem to have responded to the low lambda voltage.  Is that what you would expect with a faulty lambda sensor/exhaust leak? 

StevehS3 - I have a tube on order!

 

[aerobod - near CYYC]

The lambda voltage stuck at the minimum voltage for a narrowband sensor of around 0.1V may indicate that it is not operating closed loop, even though the revs and throttle position would point to it being in a suitable state. The lambda control status would be needed to determine that. The trim would not be applied if it is open loop or if the correction value is outside the maximum allowed amount. If it is due to lack of fuel from a faulty injector the plugs should be mismatched from a colour perspective, although a TPS value that is too far out could cause an overall lean condition.

There may be an underlying condition that is not due to lambda control, but if the base map is always out when not under lambda control it could cause rough running under partial throttle, to eliminate lambda control as the problem I would ensure air flow wise the lambda sensor is receiving the right exhaust flow and there isn't an intermittent signal (although 0.1V would point to the sensor just retuning a maximum lean value).

[John Vine]

#8: A couple of (dumb?) questions:

Can lambda control status be determined with an OBD scanner, or will the OP need to call on Easimap? 

The voltage oscillation rate in the second O2S1 trace looks very low (maybe 8 switches per 20 secs, rather than, say, 2/sec).  Is that simply down to the scanner sampling rate?

(As you can probably tell, I know nothing about OBD scanners!)

JV

[aerobod - near CYYC]

Hi John, the Fuel System Status PID 0x03 should show Lambda control status, but it may depend on the scanner capabilities whether this can be displayed or not (https://en.wikipedia.org/wiki/OBD-II_PIDs#Service_01_PID_03)

I think the oscillation at 0.4Hz or so is on the low side, but not unusual from other vehicles I've seen and is going to be as much dependent on ECU response to the sensor input as anything else.

[DW199]

Fuel system status can be displayed in real time on the phone.  I'm not sure whether the software I am using (Car Scanner ELM OBD2 for ios) allows this to be logged.

The right hand side of the lower trace where the revs are held constant at around 2500 shows an odd shaped trace for the O2 volts and trim.  I am reasonably confident this is a sampling frequency artefact.  

The car is going to be returned to Caterham Gatwick next week for investigation and fix under warranty.  Previously they couldn't replicate the fault and I didn't realise that so much useful information was available and could be logged with cheap OBD tools.  At least I now have some objective evidence of a fault rather than it being caused by my incompetent driving.  

[John Vine]

#10:

Thanks, James.

For us MBE folks, does an OBD scanner offer anything that Easimap doesn't? 

JV

[aerobod - near CYYC]

Hi John, I don't think an OBD scanner will give anything from the MBE 9A4 or 992 ECUs that Easimap can't display.

[Tom_Arundel]

Talk to Lewis Hamilton, he had the same problem in Mexico and said `The track was much higher up than we expected, we thought it was at sea level but it wasn`t and it made the motor misfire`.. .....Sorry!

Is yours a variable or fixed cam engine?

Is the air filter clean?

Any manifold air leaks ?

cold and wet needs more fuel so if it`s lean to start with it will get leaner.....chug...chug!

3500 is where it starts to `get on the cam` 

 

 

[John Vine]

Thanks, James.

JV

[DW199]

The car went back to Caterham Gatwick under warranty last week.  Apparently the problem is due to the catalytic converter breaking up.  The car has only done 3.5k miles so I am slightly surprised.  It will be interesting to see how it runs after the cat is replaced.  

[aerobod - near CYYC]

Usually a cat break-up is due to running too rich of a mixture for too long. I would ensure that the cause of the break-up is identified, as it is very uncommon at such a low mileage unless the converter has suffered a very significant mechanical shock.

[John Vine]

I agree with James. 

Did CC offer any comment or explanation about what may have caused the cat damage?  Cats aren't cheap, so I'd hope that they didn't simply replace it without a lot of investigation.  For example, did they say anything about the lambda sensor?

Do you happen to know whether the cat was fitted continuously throughout the car's Academy life, and does the 3.5K miles include all the track usage?  Indeed, was the cat the same age as the car, I wonder?  I'm even wondering whether the lambda sensor has been faulty throughout your ownership, leading to continuous over-fuelling.

JV

[DW199]

You are right, cats aren't cheap looking at the price on the Caterham parts site.  I'm glad they have agreed to cover this under warranty.  The car did the Academy followed by a few track days so I think the mileage is about right.  I don't know whether the cat is the original.  I will ask CC the relevant questions when they confirm the availability or otherwise of a replacement.

[DW199]

CC have no cats or decat pipes in stock and don't know when they will receive any. Apparently it is not totally unheard of for a cat to fail in a race/ex-race car due to 'hard use'.  They will look for any underlying cause when the car has a properly functioning exhaust system. 

All a bit frustrating.

[DW199]

To update this thread, Caterham replaced the cat and air filter and I collected the car.  It seemed the same on the road but I had a track day booked at Goodwood so took it there for more thorough testing.  It was possibly slightly better but still rough at light to moderate throttle around 3 - 4000 rpm. I was a bit disillusioned with it and Caterham at this point and contemplated selling the car.  It would have to go to a dealer as I couldn't sell the car privately in this state with a clear conscience.  

Then I spoke to Steve Greenald of Two Steves fame.  He seemed interested in the problem and after I had purchased the MBE can adapter and installed easimap on my laptop provided me with an ec2 setup file that was appropriate for the 959 ecu fitted to my car. He looked at the log files and made a few suggestions.  It was at this stage when I was operating the throttle pedal by hand, crouched over the side of the car peering at my laptop, that I realised the throttle was quite stiff and "sticky".  I disconnected the cable and operated the mechanism at the throttle body, which was very stiff.  I loosened the tps and the action of the throttle suddenly felt free and normal.  Clearly the tps had been overtightened or misaligned.  The cover plate had been bent by the screw heads.

I fitted a new tps, being very careful to get correct alignment and not overtighten. New spark plugs went in, I sealed the leak at the 4:1 collector with Permatex Ultra Copper sensor safe sealant and replaced the lambda sensor.  I checked for vacuum leaks on the induction side but couldn't find any.  I then spent ages setting up the throttle stop and tps position.  

On the road the car has been transformed.  It now works as I would expect.  There is no hint of the rough running, the kangaroo effect has gone, idle is steady and overall I am really pleased with it.  It is a shame it didn't run like this when I bought it.  I said at the time that I couldn't believe they all ran this badly but I was assured that they do.  It's not true.  I am glad I persevered.

****

Notes

Caterham were out of stock of the tps, plugs and lambda sensor so I purchased from alternative sources for around half the price.

Plugs NGK TR5B-13 from Euro car parts £12.36 for the set.
TPS came in a bag labelled Jenvey TP1.  It is a Colvern CP17 and appears identical to the original.
The lambda sensor was a problem.  The Bosch catalogue doesn't include 0258006600 which is the type no on the original sensor. I fitted a Bosch 0258006155  This has a similar body to the 600 but slightly fewer larger holes at the business end. The strain relief is slightly different, the lead is about 2cm longer but the wiring and connector are the same.  It was about 1/3 of the cost of the Caterham specified part and seems to work.

[John Vine]

Well, that's what I call a good-news story.  Thanks for the update.

But I'm surprised CC didn't spot an over-tightened TPS (or the stiff throttle action).

JV 

[DW199]

To be fair to them, they did replace the catalyst and the throttle stiffness was quite subtle. I can't be absolutely certain which of my changes solved the problem.  Possibly more than one could have contributed.  I was contemplating spending money on a rolling road session and re-map, so decided to fix all the simple things that could be the source of the issue before going to that expense.  It was also an interesting educational exercise to use easimap and log the engine data.  

[oilyhands]

It is not uncommon for a TPS to bind on the end of a throttle shaft, it happens on just about every set of Jenvey TBs I fit which retain the OE TPS, so I always space them on the mountings to ensure this does not happen.

Oily

[AlastairA]

Wrong thread! Apologies.