K series emissions shocker [again] - please help

[YellowPeril]

Yesterday I had a pre-MOT emissions test. The car is an R300 (K-series, EU2) so I was not expecting great results but these seemed rather extreme :

 

Fast idle:

CO: 1.5 – 2.5% (limit is 0.02%)

HC: 2000ppm (limit is 200ppm)

Lambda: 0.92 (limit is 1 +/- 0.03)

 

Within the last 18 months the car has had verniers fitted (DVA), RBTBs balanced, new plugs and TPS, compression tested (all good) and the MBE ECU has been remapped by The Two Steve’s. I have also attempted to better seal the exhaust joints as these were leaking slightly. Up until recently the car has been running really well, if still a little thirsty (26mph), but far more driveable than before the verniers & remap.

 

I am confident that the map is good as emission readings were at the time within the limits of an MOT.

 

The current CO and HC results suggest something has changed. Over-fuelling? The car has also recently started kangarooing badly at low speeds.

 

I am at a loss as to where to start checking (I am not that handy with a spanner). I suppose I should check the TPS again and possibly the coolant temp sensor. I don’t want to start randomly replacing parts so is there anything else that might cause such poor emissions and is easy to check? 🤔

 

[Grim Reaper]

It's running rich, get the mixture leaned off a bit and you'll be suprised how the co & hc figures drop like the proverbial stones.

Check air filter is clean, lambda sensor is ok and if it's still too high you're looking at remapping the necessary points to get it to pass.

You could go down the wideband lambda controller route where you can offset the emulated narrowband output response to fool the ecu into thinking its rich enough when it really isn't.

[Jerry Auton]

Regret don't have any suggestions for you Stephen, however, I beleive I have a spare TPS if you want to swop your's out to eliminate it.

[YellowPeril]

It's good to know that the CO and HC readings can be reduced significantly. The car was remapped to pass an MOT 18 months ago and all was fine.

My concern is that something has changed to cause the rich running. I am open to all ideas as to what might have caused this change as unfortunately it is a 150+ mile round trip to return to Track 'n' Road to get the ECU connected up to see what might be happening. Tis a shame so few places can read an MBE ECU – if only I had an Emerald

[YellowPeril]

Hi Jerry. The offer of borrowing a TPS is appreciated - I may take you up on that please. It would save me replacing it again unnecessarily

[Stationary M25 Traveller]

Is the exhaust tailpipe clean inside, or black and sooty ?

[Jerry Auton]

Quoting YellowPeril: 

Hi Jerry. The offer of borrowing a TPS is appreciated - I may take you up on that please. It would save me replacing it again unnecessarily

I intend to be at the ReHaB meet on Thursday, will take it with me in case our paths cross. Otherwise give me a call when you are this way.

[Born2Run]

Intake manifold air leak would be my long shot. More air going in compensates with additional fuel and hence increased readings. With it running you can test for leaks with a liquid sealing product (it is stuff in a spray can that is not flamable but I do not have the name) If there is a leak you will hear the engine note change temporarily as the leak is filled. A small leak can make a big difference to the readings. Experienced this on a suzuki van I had.

[shn7]

Was the test done with it all hot hot hot? Could be rich readings due to being cold though this wouldn't account for the change in behaviour emulating the bouncing marsupial.

[Grim Reaper]

I notice you have Roller Barrel TB's too. I believe these are noted for being difficult to set up correctly for emissions tests.

Even more likely is an exhaust manifold leak upstream of the sensor, air getting in makes ecu think it's too lean so richens mixture up, result is obvious.

 

Edited by - Grim Reaper on 23 Jul 2013 15:46:37

[Ian Mackenzie]

Hi there

 

Any exhaust leaks can effect the emissions so I would make sure these are all fixed. As others have said check all the senders / connectors, if there is a bad connector this may also upset things as the ECU will see a different readings.

 

Another clue is the idle test (Normally the ECU will not be closed loop and will only use the fuel map without any correction), was this OK?

 

Cheers

 

Ian

[YellowPeril]

Thanks for all your thoughts and suggestions

 

Forgive my ignorance, but I am trying to think this through methodically…

 

Air leaks have been suggested to me before:

If the ECU adjusts the fuelling based on information it is sent, it must get this information from the lambda sensor? Is this correct? If so, would disconnecting the lambda sensor result in fuelling as per the map rather than adjusted to suit a reading caused by an air leak?

 

Roller barrel throttle bodies:

Yes, these can be difficult to set-up but they have been done. Perhaps they need doing again – do they typically go out of adjustment much over 12 months?

 

Senders/sensors/wiring:

These definitely need checking. The TPS has been replaced recently and may be faulty again, plus the connections to the TPS have previously failed so I could recheck this.

The coolant temp sensor is a cheap part so perhaps I should just replace?

Are there any others to check?

 

To answer a couple of question:

I don’t have a copy of the idle test results but I believe they were similarly as bad.

The car had a good blast shortly before the emissions test, plus the test was done at length twice in the hope that we could get the temp up further.

 

[Birkin S3 ZA.]

If you checked the voltage on the TPS and recorded it, then change the voltage to give the right readings to run 1,00 lambda ?

You might have to disconnect the exhaust sensor, its only a quick fix that might work to reduce the emissions.

But you need to find the main fault that's causing the problem sooner or later.

 

My TPS has to be set after every gymkhana, the sudden opening and shutting the throttle seems to move something as the volts drop from 0,75/76 to 0,74 and it makes a difference to the idle mixture.

just 0,01/02 and it all runs normal again, if it gets set to 0,77 it runs to rich.

Chris.

[Ian Mackenzie]

Air leaks have been suggested to me before:

If the ECU adjusts the fuelling based on information it is sent, it must get this information from the lambda sensor? Is this correct?

 

ANS – This will be dependant on the rev range programmed to enable closed loop –normally closed loop will not be used for idle or for high load areas of the fuel map (especially if a narrow band sensor is used) but Two Steve’s will be able to advice here. Air leaks will still provide poor emissions as ECU will not be able to correct this even when in closed loop.

 

If so, would disconnecting the lambda sensor result in fuelling as per the map rather than adjusted to suit a reading caused by an air leak?

 

ANS – Potentially yes, but thus will be dependant on the ECU – I’ve not used an MBE, I built a megasquirt for my car

 

Roller barrel throttle bodies:

Yes, these can be difficult to set-up but they have been done. Perhaps they need doing again – do they typically go out of adjustment much over 12 months?

 

ANS – Potentially yes, as an aside where is your O2 sensor, is it in No4 primary or after the collector?

 

Senders/sensors/wiring:

These definitely need checking. The TPS has been replaced recently and may be faulty again, plus the connections to the TPS have previously failed so I could recheck this.

The coolant temp sensor is a cheap part so perhaps I should just replace?

Are there any others to check?

 

ANS – there should be an air pressure sensor and inlet air temperature sensor as well – this may be the same unit – for the roller barrel set up U think this is on the air filter back plate.

 

To answer a couple of question:

I don’t have a copy of the idle test results but I believe they were similarly as bad.

 

ANS– This is good so at least there is consistency here!!

 

As you said, start with the basics, as the TPS was changed I start here along with making sure there are no air leaks.

 

Hope this helps?

 

[oilyhands]

An air leak will make a difference for the testing equipment as it will see free oxygen, it will only make a difference for the ECU if the lambda sensor is downwind of the leak, I.E. in the collector of the exhaust. I would check your TPS resistance across pins 1 & 2 and make sure it is 0.43 ohms or close, if it has been replaced it may be incorrectly aligned.

 

Oily

[Paul Deslandes]

Do you mean 430 Ohms Oily? .43 would be too low to measure accurately. The TPS has a total resistance of about 5000 Ohms so 430 at idle seems about right.

[oilyhands]

Sorry, scaling error..

 

Oily

[Fil M]

Hello Yellow peril .... I am in your county and have a K series 1800 on RBTBs with an MBE ECU (full R conversion by CC). To get the car through an mot on that setup requires it to be adjusted spot on with regard to the TPS resistance and the balance and draw of air on the throttle bodies as specified by CC. The throttle bodies do need checking now and then and it is very likely yours need a 'tweak'. Mine are checked at six months and before MOT and they generally need a tweak. If they are not spot on you will run rich.. Essentialy my car is an R300 under the bonnett. I adjust mine and have the tools to do it. If you want to use my kit or for me to dart over and bring it with me to show you how its done just blat mail me. I feel sure the adjusting of the RBTBs and TPS is your first place to check..... it only takes 5 minutes to check them and perhaps twenty minutes to adjust if needed. I too considered a re map etc due to MOT issues so have experienced the same problem with the same set up. If you want me to help please give the two steves a bell and ask them what level of air draw on the RTBS they set on your car in case it is different to those set by CC. CC state 5kg/hr on each pot or averaging 10kg/hr for each pair ..this is at an idle of 900-1100 rpm. The TPS should be .43-.45.. If they set it differently with their mapping we would need to check against their settings and not CCs to get her back to MOT standard

[Piers Rendell]

I also have an R300(K). I drive the car to the MoT station and the checks are done on and off the ramp + brake tests. Then I thrash the car around the block and get it really hot, as it is the only way it passes the emmission test. I have been going to the same friendly MoT station for the last 27 years, but only in the last four years with the R300.

 

Piers

R300KWK