Duratec poor running

[Angus]

Car is a 2007 R400 duratec with standard CC 210hp engine, done 13000 miles, standard CC (locked) MBE ECU.

 

I am experiencing some very poor running that I am trying to sort out. The symptoms are serious hesitation when trickling along in traffic. Up to about 4000 rpm when pressing the accelerator there is a lot of hesitation, kangerooing and poor acceleration. Above 4000 its fine and pulls strongly. Also it is only evident when car is warm - from cold for the first mile or so it is fine. Also it idles fine, warm or cold. The plugs were really sooted up. Ironically I have just returned from 2000 miles through Spain and France without any problems at all. It all started on the first drive back in the UK.

 

As this was the first problem I have had with the engine, I thought I would have a go with the MBE diagnostics from SBD and at the same time try to learn something about modern engine control systems.

 

So I bought the connector box and downloaded the software. After a bit of trial and error trying to fathom the dozens of different parameters that one is presented with I have made some logs of some of the engine settings.

 

See this link here. It shows the lamba reading in volts (the sqiggly green line in the centre and right axis) for different engine speeds (purple line and left axis). The blue line is the ignition advance. The dense areas of blue squiggle are periods of idle.

 

Should the lambda reading be oscillating so much (between about 0.85v and 1.2v) My vague recollection from MoT emissions testing is that it doesn’t really vary.

 

Am I right to suspect the lambda sensor or should I be looking elsewhere ?

I have already checked the earths, checked all the engine loom connections, TPS wiring and changed the plugs – no effect.

 

Any ideas much appreciated.

 

Angus

[The Pikey]

Sounds just like the Lambda to me.

 

Jason

[Pat the Plumber]

do you start the car with a dab of throttle?

 

[Angus]

Pat

 

In answer to your question - not usually. Looking closely at the trace I posted I can see that it looks I may have in this instance. Either way the poor running problem is the same, and it only happens when the car is warm - never from cold.

 

Jason - thanks ! You say 'sounds like' - is that from reading my description of the problem, or looking at the trace, or both ?

 

Angus

[The Pikey]

Angus

 

just from your discription really. My understanding is that the ECU runs on the map after 3k ish and below that it runs on the lambda to get it through emmission tests. Also the sooty plugs could point to that too.

 

I think to pass a MOT I think your lambda reading needs to be somewhere between .97 to 1.03 and if this lambda voltage reading you are referring is the same as my Easymap lambda reading then you are too rich at times with a .85 reading.

 

Have you tried unplugging the Lambda? If I do this on mine the ECU runs on the map throughout the range.

 

 

Jason

 

 

Edited by - Jason Fletcher on 2 Oct 2011 20:24:53

[Paul Deslandes]

It would be worth checking the water and air temperature readings, as logged by Easimap, as well as Lambda, to make sure they're 'sensible'. You say it gets worse when its hot so it could be that the water temperature sender is telling the ECU to overfuel if it 'thinks' its cold. Also, check that your airfilter is clean, especially if you have a CC elephant's trunk with the inline air filter and three line below-radiator pickup that sucks all manner of crap in from the road.

[mic]

What is the TPS site with the ignition on?

[Angus]

Jason

I have not tried just unplugging the lambda. I guess I was thinking that it wouldn't start or just stop ! I will try it. If it works one wonders what is the benefit of a lambda ? Is it just emissions ?

 

And another thought - if the ECU really needs it why not feed 1v into the ECU direct from the battery ? Easily achieved with a resistor, and would remove a failure point.

 

Paul, I captured water and air temps on another plot I did. Both look sensible. Water gradually climbs to 90 or so after about 3 mins, and peaks at 100 at around 4 min 30. Air starts at around 50 and gradually declines and settles at around 40. Perhaps seems a bit high, but its smooth and not spiky so hard to see it causing the problem.

 

As regards that air filter I have the caterham single plenum with the conical filter at the rear near the battery. Certainly doesnt appear clogged but I havent tried running it without.

 

 

mic: I think from the plots I have (trace labelled 'Throttle Site 1') it is 1, but I cannot be absolutely certain as I may not have started the logging until the engine was running. I'll recheck that.

 

Great ideas guys - keep them coming !

 

Thanks

Angus

 

 

 

Edited by - Angus on 2 Oct 2011 23:39:35

[Paul Deslandes]

I'm guessing that you have a narrow band rather than wideband lambda sensor fitted. This works in a very different way to the w/b version and only controls AF ratio to approximately lambda=1 at and around idle. The voltage ramps up and down very quickly as the ratio 'hunts' around its correct value. It has no affect otherwise. The wideband sensor gives a voltage that changes more or less linearly with AFR and can be used to control it over the entire rev and load range - so-called closed loop control. Generally these are only used when mapping and are switched out once a decent map has been programmed into the ECU - running in 'open-loop' mode. So, if it is a narrow band sensor it should not be doing anything in the rev/load areas that you're having trouble with.

The air temperature readings are a bit odd but I wouldn't expect them to have too much impact. All it does is apply a fuelling correction depending on air temperature and it normally flattens out at higher temps. You could try disconnecting the air temp sensor as the ECU should default to a sensible value if the sensor fails.

Correct fuelling and ignition timing rely on all the bits working correctly and not straying too much from their spec. The ECU looks up the correct value for each in a table that holds the 'map' for each rev/load site. So things to look at are injectors, plugs, crank sensor, temp sensors, potentiometer throttle position sensor etc. Does the standard plenum set up use a MAP sensor or air flow rate sensor rather than a TPS. That would be causing a problem if faulty.

If you run out of ideas, I'm only in Gomshall so blatmail me your email address if you want to bring it over sometime. I'm around all day in the week as well as weekends.

Paul

 

Edited by - Paul Deslandes on 3 Oct 2011 11:05:16

[geko]

Hi,

 

two month ago I faced a similar problem with my CSR.

Finally it was the ignition. I changed all COPs and the problem was gone.

Why the ignition works fine over about 3.500 rpm...no idea.

 

 

Edited by - geko on 3 Oct 2011 19:06:52

[The Pikey]

I think Angus's engine will be running wasted spark.

 

Jason

[Dignity]

"The symptoms are serious hesitation when trickling along in traffic. Up to about 4000 rpm when pressing the accelerator there is a lot of hesitation, kangerooing and poor acceleration. Above 4000 its fine and pulls strongly. Also it is only evident when car is warm - from cold for the first mile or so it is fine. Also it idles fine, warm or cold. The plugs were really sooted up."

 

Probably not relevant but I had exactly these symptoms with my K series and it was the Fuel Pressure Regulator ?

[Paul Deslandes]

could certainly be fuel pressure too high if the regulators goosed. The injectors rely on the correct pressure to deliver the correct amount of fuel. Do we know what pressure it should be set to. e.g. 3 bar?

[Angus]

Hi all and thanks for the responses. I have not had a chance today to do any further investigation, but in answer the of points:

 

Paul - I think I do have a narrowband lambda, it is this one here, MG Rover part number MHK10006. So are you saying that the lambda and its oscillating voltage output visible on the trace is ignored by the ECU above idle speeds ? And therefore it cannot be causing the problem ?

 

The R400 D uses a MAP sensor - this one here.. It is located on the plenum betwee No1 and No2 inlets. What does the MAP sensor do, and what is the best easimap trace to examine it with ?

 

I have also suspected the TPS, which is this one here., but the traces look fine. I've also taken it off and using a multimeter checked for a smooth change in the resistance as it is rotated.

 

As this is a Caterham standard engine, I have individual coils per plug, mounted direct on top of each plug. As per this - here.

Is that COP ? However seems unlikely they would all fail at once. Is there a way of testing them, other than cranking with the plugs out, when they all work fine ?

 

On the R400 duratec there is a single fuel line, ie no return line. If there is a regulator (and I am not sure there is), it is all contained within the fuel module - here.

 

Geko - how did you trace your problem to the ignition ? Was it just by swapping components or did you use the diagnostic outputs from the ECU ?

 

Many thanks everyone

 

Angus

[Paul Deslandes]

Angus, you have mail.

 

[Paul Deslandes]

From the reference this looks like a narrow band lambda and they are readily available if it does prove faulty, although your traces, at first sight, look ok.

 

The MAP sensor is monitoring Manifold Pressure, which varies with engine revs and load, dropping when the throttle is opened and rising when its closed. They are sometimes used on their own or in tandem with a TPS which measures the degree of throttle opening.

 

The Coil-On-Plugs should be okay and usually either work or they don't. You'd most likely have a misfire if one was faulty and this would probably get worse at high loads/revs rather than at low revs per your symptoms.

 

That fuel module looks like a box of tricks . Is it pump, regulator and fuel level? Depending on the plumbing we might be able to measure the fuel rail pressure and see whether its roughly right and, more importantly, reasonably stable.

[geko]

Quoting Angus: 

Geko - how did you trace your problem to the ignition ? Was it just by swapping components or did you use the diagnostic outputs from the ECU ?

 

Hi,

 

the datalog from the ECU showed some high lambda peaks.

This point me to unburned fuel (= oxygen present)

Inexplicably I had not much more misfire then usually.

 

Therfore I changed the spark plugs = not better

and then the COPs (= Coil on Plug) = bingo

 

Why the COPs worked fine when engine was cold and/or above 4.000rpm?

I don´t know.

 

I must admit that damaged COPs are very rare...just one point for you to check after the other more probable points are not the cause.

 

By the the way I had also no problems when the engine was idling.