Emissions

[lee]

The continuing saga of this here donkey running too rich.

 

Tested on the new Carlos Fandango computer this morning.

 

At idle:           c/o  2.6

                    HC  1900  Unburnt fuel (Oh dear)

                Lamba  1.4

 

At 4000 rpm:    c/o  2.5

                     HC  330   Unburnt fuel

                Lamba  1    

 

Does anyone know what is normal? obviously the HC is way out. This is what causing me all this grief and whilst I sit here know I smell like it.

 

Thanks Guys

[CP62]

Lee,

 

Sounds like you need to get the car onto a rolling road and check a few things out.  Given that the PI system meters fuel based on vacuum versus the pressure from the MU springs, you need to know the vacuum at idle and at speed, both unloaded and under load (ie throttle open).  My knowledge of specialists with rolling roads is limited to the South West, but you could try Tony Lindsey-Dean at Kingston Sportscars.  He is near Norwich, which is a bit closer to you, and though he doesn't have his own rolling road, he knows a lot about these cars and I'm sure will have a tame rolling road that he uses.  (phone:  01359 269777)

 

The problem I had with my 5 was that my cam had more overlap than the standard one, leading to much less manifold vacuum once I opened the throttle.  With the MU adjusted to give the right mixture at high speed/full throttle, it over-fuelled virtually everywhere else in the engine's speed range (even with light springs).  End result:  15 MPG and black smoke/smells most of the time.  I have since fitted an EFI system which is much better at fuel metering ...

 

All the best,

 

CP62

[TimMills]

Lee my old chum!

 

Did you know there is a rolling road in Tonbridge at Classic Cars of Kent? They are in Medway Wharf Road.  01732 773330

 

They might be worth a try?

 

T

[lee]

Thanks me old fruit cake but are they not MG Specialists?

[JohnC]

It does look like your datum track (aka fuel cam) is miles out, or that the adjusting screws are set wrongly, or both. Which is odd, as I'm pretty sure you've said previously that you have had the metering unit calibrated. Is your camshaft standard?

 

Just for giggles, try inserting a feeler gauge between the cam follower & datum track roller in the metering unit while at idle (you'll have to take off the black plastic "Lucas" cover first). What gauge thickness gives you the fastest idle (without touching the throttle, dizzy etc.)?

 

You've banged your head against a brick wall on the fuel front; Is it possible that it's the breathing that's wrong? In other words, the right amount of fuel is getting metered, but the wrong amount of air is being allowed in. Have you checked your cam timing? Might there be a restriction in the air filter?

 

Just a couple of thoughts. I'd be interested in the results...

 

Cheers,

John

[BryanD]

Since there is a lot of unburnt fuel at idle, could the idle breather tube be blocked?  The rubber pipes between the Pi bodies have been known to collapse internally.

 

If the idle was particularly rough, or some plugs sootier than others it might support this theory?

 

Alternatively, do you have an injector (or two) dribbling through a leaky seal?  Easy to check the spay with a jam jar...

[lee]

Your all getting a bit technical now...If any of you knowledgable guys live in Kent and are bored I'd love you to meet my car.

 

I was just thinking that maybe the car has a different cam. This would explain the rich idle I'm sure. Not sure how i would know though.

 

Yes Brand spanking M/U from Malcolm.

 

Fitted new injectors at the start of this saga but it may well be worth checking them again since all the work done to the fuel system could have caused poo to partialy block them (think they call it constipation).

 

Idle air breather has been changed for a minature k&N since the standard air box and chamber has been launched.

 

As always I appreciate your help

 

lee

[Roger H]

Lee - a short answer to your question is....not much!

 

I would be wary about comparing current vehicle emission standards against a system that's over 40 years old. You will never achieve the current ideal figures in all aspects of operation. The Lucas system was designed for performance, not economy. Cam overlap results in misleading readings as fuel and air is allowed into the combustion chamber prior to the closing of the exhaust valves. This is worst at idle as there is minimal scavenging from the exhaust system and is better at higher revs. Because the system is mechanical and the datum track is linear, it is not possible to individually adjust settings for all situations. It will run richer at all revs on a hot day compared to a cold day. It will run richer at higher altitude etc.

 

If your unit has been set up "correctly" then don't touch it - and from all reports, Malcolm at Prestige has a reputation second to none.  Even the Lucas manual states a range of acceptable fuel "shots' at various vacuum levels so there will be some variation from car to car. With recon injectors your injection system is probably very good and you should look elsewhere for the problem, if there is one!  

 

It could be that incorrect timing is causing your problem.

 

My first sojourn in to the dyno pit left me depressed. I investigated everything possible withthe MU and even considered fiddling with the linear datum track profile. I used a real-time fuel/air meter in combination with 2 metering units. 4 'castle' top assemblies, red, blue and black spring sets and combinations of all. After over a hundred hours of logging painstaking statistics, I relented and sought professional advice!!

 

I suggest you try to find a dyno shop where someone has direct experience with Lucas PI - the shops that tune turbo rice-burners and EFI stuff will not be able to comfort you. They will also be able to (very accurately) get the timing right.

 

Hope this helps

 

R

[lee]

I'm aware that you cannot compare with modern emissions but the unburt fuel levels at idle is excessive i'm sure.

 

I would be interested to know what it should be. does anyone have this figure?

 

There is certainly something not right with this set up. Yesterday I completely sealed the boot lid with tape and plugged the rust at the back of the spare wheel well and I still get fumed out.

 

My next step is to extend the exhaust pipes upwards by about a metre, similar to that on a Massey.

[Schnippel]

hello lee,

your engine run at idle  too lean!  lambda 1,4 shows this clearly.

When the engine is an  143er  you must have an Lambda about  0,86- 0,90 .

CO about 4 – 5,5 %  . the high hc is normal at idle from a 143er tr engine !

when the engine run a few minutes at idle the hc exceeds over 4000 hc !

responsible is the duration 280 °   the 123er engine have better rate.

importantly to know is , becomes the mixture too lean rises the HC worth.

 

control of the datum track is important   0,3 mm 143er

0,05mm 123er.a very frequent error is that the range idle only with an attitude is made 0.05 mm for all engines .

 

regards

ralf

[JohnC]

Ralf has made me think (steady!). I'm afraid I dunno about these lambda thingies, but if Ralf's right, and lambda 1.4 is very lean, then it may be that you have a cam with loads (and I mean Stage 3) of overlap. With lots of overlap I understand that the intake valve opens well before the exhaust closes, and at low rpm this results in fresh air (relatively speaking) flowing from the inlet straight to the exhaust. As the lambda sensor measures oxygen levels, it tells you that you have a lean mix, whereas you really just have lots of unburnt mixture in the exhaust.

 

This would tally with a high unburnt HC level at idle, which you have.

 

If you have a non-standard cam, then you're wasting your time getting the PI calibrated on a bench. You need to get it to somebody with a rolling road who knows what they're doing with a Lucas PI. Perhaps it's time to get a dial gauge and check the cam profile?

 

I'm no expert, so anybody please chip in with corrections!

 

Cheers,

John

[Roger H]

John C - While we agree on the point of misleading readings, I would be very cautious about

If you have a non-standard cam, then you're wasting your time getting the PI calibrated on a bench.

I believe the only way to get the metering unit calibrated correctly is on a propery equipped test bench. The basis of Lee's concerns is the readings that he is getting on the rolling road. If the misleading readouts are not an accurate indicator of the combustion requirements of the engine (and we both agree on this) then it is not much use fiddling with the MU on a dyno.

 

If it were that simple, calibration of the MU would be done on dyno's regularly, in preference to test benches as you would be able to calibrate the MU to the individual characteristics of the car. The major reason for this is not done is that the MU is virtually inaccessible for the full range of adjustments that need to be carried out to calibrate it properly. It is virtually impossible to adjust the datum track while on the car and access to the full-fuel screw is nearly as bad, particularly with a hot engine!! The car's vacuum system is not controllable enough to enable the job to be done properly. The other thing is that it could cost a small fortune in time, assuming you can find the adequately trained personel.

 

Lee - I asked a similar question of the board a couple of years ago, but it seems there are no figures available. It was then I embarked on the DIY route. Exhaust gas analysis could not get me the results and I was only dealing with fuel/air ratio (lambda - unburnt oxygen).

 

My TR has a fast-road cam and hi-lift rockers - I have not had it dyno'd since the MU was done but I plan to do it soon. I will post the results.

 

If any others could also be posted, it could be an interesting and valuable database.

[JohnC]

Fair point Roger. Reading what I said, I didn't write what I meant to mean, etc...

 

IMO to calibrate a non-standard engine you need a combination of rolling road and[]/i] test bench. You're absolutely correct that the MU can't be fully set up on a dyno - in particular the max fuel stop & datum track really can't be accessed. The adjusting screws on the top can, but they're only part of the story.

 

Where you do need a dyno is in determining what adjustments you're aiming for on the test bench. Not if the engine is standard of course, as the manual & experts' experience would be fine.

 

Cheers,

John

[CP62]

Roger and John,

 

Sorry, but I have to disagree on the value of bench settings for modified engines.  The only way to assess the engine's fuel demand is to run it at a range of speeds and loads, which you can only realistically do on a rolling road.  

 

A few years ago, I fitted a hot cam, gas flowed big valve head and tubular exhaust, all from the same supplier.  He also re-calibrated my metering unit, but advised me to get it set up finally on a rolling road as soon as I could.  When I had this done, it became clear that the full load setting was a little lean for comfort, which meant moving the fixed stop and recalibrating the rest of the range (all done on the spot).  Without doing this, I would have risked overheating and holing a piston.  

 

The accepted wisdom that the PI experts can calibrate a MU accurately on the bench just doesn't cover modified engines adequately, though it is fine for standard engines whose charateristics are well known and hence are more predictable.  

 

This all became clearer to me as I went through the process of fitting the EFI and mapping it; once you have built/modified the engine, it basically acts as an air pump, drawing gas in and expelling it according to (primarily) air density, speed of rotation and throttle opening.  It will only burn as much fuel as the airflow will permit, but it's crucial to get the fuelling right to avoid too lean or too rich a mixture.  The only way to do this with any degree of accuracy is on a rolling road, with appropriate sensors in the exhaust gas stream, and to use the road to simulate as much of the operating range of the engine (loads and speeds) as you can.

 

Hope this helps, Lee - my bottom line is that you need to know whether your engine is modified or not, and what it's really doing in terms of air flow.  A rolling road session with someone who knows what they are doing may be the only answer.  All the best with it!

 

CP62:)

[lee]

I think that with all your help and suggestions that a rolling road set up is the only way.

 

With all the difficulties I have been experiencing the car has probably got a hotter cam. Is there any way that I can know for sure?

 

Failing that and a box of swan-vestas,

EFI...what is the best system to go for and at what cost?

 

thanks

lee

[CP62]

Lee,

 

You can get a reasonably good idea of your cam's characteristics with a dial test indicator (aka clock gauge), a timing protractor and some patience.   This will allow you to measure the lift, position of full inlet and exhaust lift wrt the crank and also to get a rough idea of the duration/overlap.  However, if you aren't particularly experienced in engine building/diagnosis, this may be wasted time as it isn't that straightforward.  It might be better to engage a specialist like Tony L-D (mentioned above) or others who know the PI system and have access to a rolling road, and spend an hour or two diagnosing and hopefully sorting it out.  The guys at Sigma Engineering in Dorset who set up my 5 could do everything it needed and had spare springs etc to hand.  

 

On EFI, my experience is limited to the system I fitted myself, which is based on the Lucas throttle bodies bored out to take modern injectors, and uses an OMEX ECU.  I did all the 'fitting' in its fullest sense myself, including a toothed wheel + mag pickup on the crank nose to provide timing info, air and coolant temp sensors, throttle position and lambda sensors.  The result is functionally very good but I suspect could be bettered in terms of facilities by the kits available from the likes of Revingtons.  I fitted EFI to my 5 because I just wasn't happy with the smelly, thirsty compromise which seemed the best available from PI on my modified engine.  I couldn't afford - or justify - the high cost of these complete kits, so decided to DIY it.  I believe OMEX are introducing a TR kit too, so they may be worth a try (01242 681044).

 

All that said, there are lots of people who still use the PI system on non-standard engines and are happy with it.  I just couldn't get the characteristics I want and moderate economy at the same time, which is why I went that way.  My car does a lot of miles, so economy is probably more of an issue to me than to some.  

 

Once again, hope it helps - if you want to discuss any of the above, send me a PM and I will email you a phone number.

 

CP62

[Roger H]

Guys - I think were all standing on the same side of the bar here....

 

FWIW, this is the method used to get the thing right for my car.

 

Using a portable fuel/air analysis meter (via exhaust), found the maximum fuel requirements for the car (at 4000+ rpm in 3rd gear, pedal to the floor).  When this theoretical point was established, adjusted the max fuel up and down a bit to see how the car responded. (Ideally this would be done on a dyno but the costs were a factor and this is only the beginning of the tale.)

When happy with flat out performance, the MU was removed from the car and sent to a calibration bench, together with a note of the manifold vacuum figure at idle (and the cam specs). The 'vac at idle' figure is needed to set the bottom point of the fuel curve. The max fuel output of the MU was measured on the bench and used to set the top point.  [As it turned out, the max fuel flow for the car was only slightly higher than the Lucas spec** A lot more fuel is ususally only needed if the engine is oversized using the mazda pistons to achieve a 2700cc capacity].

 

The fuel curve is plotted on a graph and the the MU calibrated at various vacuum levels/rpm to suit.

 

MU fitted to the car and test driven. Because each car spec varies, I was advised some hesitation may be apparent under light acceleration, but fortunately this was not the case.

 

I suggest that if you are to use a dyno to tune the car, then it is done based on power output, not emissions. I don't think it is possible to tune Lucas PI on a dyno for the subtlties of everyday use. Adjusting fuel delivery for light load, medium load would be a nightmare as you can only check the emissions, not the power output for these conditions. The only way to really do this is by driving the car after the MU has been set to close to factory specs.

 

(**FYI the car has a stage 2 head, fast-road cam, 1.65:1 rockers and custom-made interference header).

 

cheers

R

[JohnC]

Guys - I think were all standing on the same side of the bar here....

Totally agree! Didn't quite understand your method for determining max fuel requirement, but otherwise your approach sounds exactly like what I'm in the middle of...

 

Lee, if you want to check the characterists of your cam, it can indeed be done with some patience & a dial indicator. If you use a protractor, I understand you will need to remove fan, extension etc in order to put the protractor on the pulley. I may be wrong in that, but there is another approach: Stick a strip of paper on the pulley, marked up with degree increments. You can download a graphic which will print to the right size from here. Print, cut out & stick onto your pulley. Voilá - you can read the crank degrees accurately Plot your findings on this timing disk, and let's see what you've got.

 

Cheers,

John

[CP62]

Roger,

 

I'd love to talk this through over a pint sometime...

 

I think you may be underestimating the capability of a good dyno though.  The 2 I've been to could both run at set speeds, while measuring the power output at varying throttle settings, and the diagnostic systems they both had could monitor the emissions at the same time.  I know my view is influenced by my own experiences with PI and EFI, but there's a lot they can do.  

 

I appreciate it is more difficult to assess intermediate power settings with PI as you don't have an accurate indication of throttle position, but manifold vacuum is a useful indicator on PI cars.  This is what pinpointed the problem I had with my engine on PI - the vacuum was OK with the throttle closed or just open, but fell away to nothing as soon as the throttle was opened any more than a tiny bit.  This meant that the MU went to full-load fuel as soon as you put your foot down at all, even with the lightest springs fitted and adjusted so far that the plastic cover wouldn't fit.  The engine was healthy; it just had more cam overlap than PI seemed able to deal with.  

 

(The sensible answer would have been to change the cam for a more conventional one, but to be honest I liked its power delivery and relished the challenge of fitting an EFI system, and hoped I would get a big enough improvement in economy that it would pay for itself - the jury's still out on that though)

 

Cheers,

CP62

[CP62]

John,

 

Just read your reply - what a great site!  There looks to be lots of useful stuff there.

 

One word of caution though, having read the cam timing check instructions:  You should only turn the crank 'forwards' ie clockwise if looking from the front when you are taking measurements.  If you go the other way on any of the measurements, you will be out by the equivalent of any slack in the timing chain.  You need to keep the tension in the 'tension side' of the chain.  (sorry if that's teaching anyone to suck eggs...)

 

All the best,

CP62

[jonlar]

Yes, brilliant sites John - have saved them in Favourites (must remember to save them to FD, before I lose them again!).

[Tom Fremont]

What's the meaning of the curved arrows on the timing disc? They seem to be pointing in the direction opposite to the normal sequence ??? I can't help but associate them with crankshaft rotation

[jonlar]

Seeing your comments Tom, I'd also noticed the arrows pointing the 'wrong' way, and the fact that the disc is 720º in total (each quadrant represents 180º) - trying to get my head around that as well.....................?

[Tom Fremont]

(each quadrant represents 180º)

That way all (4) strokes are represented which corresponds to (1) revolution of the cam - does take a little study I grant you

[jonlar]

(each quadrant represents 180º)

That way all (4) strokes are represented which corresponds to (1) revolution of the cam - does take a little study I grant you

That's the same conclusion as I'd come to - it should be used on the camshaft not crankshaft, glad to say I've got my head around this disc now and how to use it!