Jump to content
JohnD

Driving Lucas Pi by direct throttle control

Recommended Posts

Posted (edited)

Lucas Pi works via the vacuum in the inlet tract, that moves a 'cam', a wedge really, on the shuttle in the metering unit.

Lucas%20page%2011%20fig%2010.jpg

But Kas Kastner developed a circular cam system, opertaed diretcly from the throttle.   That wqas described in Triumph World. some time ago - see below.

Kastner's direct throttle control M-u triumphworld_apr2004.pdf

Over on the Friends of Triumph in the US, a couple of FoTs are asking for anyone who has done this.  One of them has some nice looking kit to do it:

1278173233_LucasPidirectthrottlecontrolM-u.png.bf9e5c3ddf22365de2a67587276316ca.png

  I haven't, but I'd be as keen as them to learn more.     Please post here if you can offer any advice, so I can earb and pass it on.

Thnaks,

John

 

Edited by JohnD

Share this post


Link to post
Share on other sites

Very interesting article John - thanks for posting.

Interesting in that the cam controlled system is the historical equivalent of alpha N (or use of a throttle position sensor as the primary load sensing) as opposed to manifold vacuum as the standard PI or MAP in a modern ECU.  And for the same reason - lack of a decent manifold vacuum signal.

He makes it sound quite straight forward - but the devil is in the detail - and there's alot of detail I reckon.  He talks of difficulty off-idle and part throttle mapping...... not helped in any way by 46mm throttles IMO.  Even on a full race 2.5 I reckon you'll get all the flow you need somewhere between 1/2 and 3/4 throttle.

Regret I don't have anything useful to add.  I think mechanically it's a reasonably straight-forward thing to do if you have a lathe and a mill and actually know how to use them with precision.  The setting-up part though....... pretty difficult, and time consuming.

Share this post


Link to post
Share on other sites

Kas had a bunch of those cams at a race one time.  I can't recall if he brought them or someone else had them.  He was explaining how difficult it was to get the profiles.  

Share this post


Link to post
Share on other sites

Hello all,

This is how all racing versions of the Lucas P.I. are controlled, but of course they spend their time either off or high throttle openings. I noticed that Ric Wood's three litre racing Capri is using Lucas injection. I don't know who supplies spares for all these rare Lucas systems, but relatively there are a lot of them on the historic racing schene, Can Ams for one and of course any Cosworth DFV powered cars.

I bought a spare injection system with the idea of fitting it to my Mk 2 Jagaur sometime,  but think it will need a larger diameter shuttle to give enough fuel. I have looked online to see if theer are any suppliers but all that seems to come up are Triumph injection parts suppliers? Does any one know differently?

For the Triumph, using the figures from tthe manual physical measurements would give a basis for a cam profile, but I don't see any real benefit over the basic vacuum control for a road car?

Alec

 

Share this post


Link to post
Share on other sites
Posted (edited)

Pi was originally developed on Jaguars, of course, so why not!    I posted here earlier this year about an article I had bought from "Motor Trader" magazine in 1956 (!)  That described the 'new' Jaguar-developed Pi system, but it wasn't the one we know.    It had TWO shuttles for six cylinders, and an integral oil pump!    If you can't find the post, let me know and I'll send a copy.

Kinsler Fuel Systems supplied Lucas Pi mods, I think. https://kinsler.com/Shop/

I get the impression that Kastner was alway fiddling with his system.  He would take a box of cams to an event and file another to fit if not satisfied!    Maybe that it was easy (Easier!) to take out the circular cam and fettle it than fiddle with the complicated three nut arrangement on the top of the Lucas/Triumph device, was better for him?

John

Edited by JohnD

Share this post


Link to post
Share on other sites

I'm not really missing the point (I don't think, I like clever stuff like this), but these days you'd need a very good reason to beat yourself with this thorny branch when even the cheapest and most basic of electronic systems do the control side so much better and easier.

Really though, so much of the controllability issues on PI, standard and PI modified, stem from having throttle bodies that are FAR TOO BIG so most the action takes place over the first few degrees of movement.  My PI made just shy of 150bhp on a single 60mm throttle (28cm2), and that is probably bigger than it needed to be.  Factory PI has 6 x 46mm (100cm2).  Enough for 600bhp....?

 

 

Share this post


Link to post
Share on other sites

Don't forget Poiseille, Nick!

Flow varies as the fourth power of the radius. The resistance to flow in each 4.6cm throttle will be almost twice that in your single 6cm throttle.

JOhn

Share this post


Link to post
Share on other sites

mechanical fuel control was standard on the Lucas race units, commonly seen from the mid 60's onwards in everything from 1300cc Fords to 6000cc Chevs. Lucas made 4, 6 , 8 and 12 cylinder metering distribution units, with both 6 and 8mm shuttles. ive worked on many having attended a Lucas training course in the 1970's . There used to be a wide selection of mechanical cams to choose from, but they were all focussed on full-power race engines, with getting them 'accurate" below about 50% load, a major challenge. My race 2.5 still uses this system

 

Share this post


Link to post
Share on other sites

Welcome, Terryo!

Please tell us more, perhaps starting another thread for the purpose, about your race car?

Mine uses standard Pi vacuum control.

Share this post


Link to post
Share on other sites

I've always wondered if controlling the old PI system with a simple electronic one would be an option. Some kind of IAC type server controlling the thing rather than the cam etc

 

Share this post


Link to post
Share on other sites
23 hours ago, James said:

I've always wondered if controlling the old PI system with a simple electronic one would be an option. Some kind of IAC type server controlling the thing rather than the cam etc

 

I reckon it could be done - solenoid or servo actuating via the diaphragm link... to give the load signal and control the fuelling but retaining the mechanical drive to keep the sequential element.  Simple enough in principle but as with the cam method, the devil is in the detail - the calibration and characterisation - probably actually very hard to get right, with the additional challenge of speed of response, overshoot etc.....  Might be possible to apply the mechanicals from a modern fly-by-wire throttle......

A lot easier just to go fully electronic.....

Share this post


Link to post
Share on other sites

Am I being a killjoy, If I ask why use electronic control of a Lucas Pi?    When electronic control of fuel injecors is so sophisticated?

I can see that its like climbing Mt.Everest, "Because its there",  you can and it's fun to try,  but in pursuit of performance???

 

Edited by JohnD

Share this post


Link to post
Share on other sites

Well, the PI does give timing-accurate sequential injection and a good high pressure spray.........

But I agree with you John - an interesting technical exercise maybe, but I won't be trying it myself (not clever enough for one thing!)

Share this post


Link to post
Share on other sites
On 5/8/2020 at 12:49 PM, JohnD said:

Don't forget Poiseille, Nick!

Flow varies as the fourth power of the radius. The resistance to flow in each 4.6cm throttle will be almost twice that in your single 6cm throttle.

JOhn

Poiseille only applies to laminar flow, and air in the intake duct will be turbulent. Maybe Darcy-Weisbach would be closer. And then theres the butterflies... Peter

Share this post


Link to post
Share on other sites

IIRC the Kinsler site says tick-over must be set smoky-rich to cover throttle closing transients and pinking due to fuel octane fractionation upon wottiing the thottle from closed. Lack of a thottle pump is the Lucas Achilles heel

Peter

Share this post


Link to post
Share on other sites

Indeed, which I think contributes to Pi running lean at WoT, another reason for a rich setting at idle.

In Darcy-Weisbach, turbulence,  flow varies as the  square of diameter!    Double the diameter, and flow goes up four times!    Nice work if you can get it! And you can get it if you try!

 

Share this post


Link to post
Share on other sites
17 hours ago, JohnD said:

Indeed, which I think contributes to Pi running lean at WoT, another reason for a rich setting at idle.

In Darcy-Weisbach, turbulence,  flow varies as the  square of diameter!    Double the diameter, and flow goes up four times!    Nice work if you can get it! And you can get it if you try!

 

The lack of a throttle pump leads to transinet rattling of pistons the second the throttle in floored from closed. The fuel octane at the inlet valve briefly plummets unless the closed-throttle mix is rich enough to keep the  manifold walls wetted:   https://supertrarged.wordpress.com/2013/06/12/the-lucas-pi-lean-spike/

 

 

Share this post


Link to post
Share on other sites

after road and race tuning many of these and some of the mechanical control versions on non-triumphs, i offer these comments..... 

yes, rich at idle etc is essential on all of them. 

the entire system is really only 1 or 2 steps ahead of a good carby. At best its an approximation device, rarely perfect, but, with enough time, you can tune/drive around most of its shortcomings. it is all-but impossible to get a Triumph PI system mixture absolutley perfect across the entire load range on a dyno, but in use on the road, that does not really matter, and the average owner would never know how "bad" it is at some load points. on the race track, its largely immaterial as you are working mostly at 60% or greater throttle.

the 8mm shuttle version is easier to work with on race engines due to greater fuel volume per metering unit shuttle stroke. BUT, filing/making/swapping the cams, is a truly terrible job. Lucas provided a chart of theoretical volume flow per stroke per rotation of the cam, so when new, you got some clues to work with

anyone quoting theoretical air flow issues on a triumph cyl head is both missing the point and has never had to modify one. If annything, the intake ports are marginally too big for the valve sizes from factory.  The "mk2" head has so many flow issues, is not funny. i'll happily swap real flow data with other genuine racers/tuners

 

Share this post


Link to post
Share on other sites
11 hours ago, terryo said:

 If annything, the intake ports are marginally too big for the valve sizes from factory.  

 

Hi Terry, as I'm getting ready to do some basic DIY head work on my mk2 head to gain a few more horses on a road engine.Attention particularly to the inlet SSR, what size inlet valves would you say are right for the ports which the factory have given us. On the same note the exhaust throat is slightly too large if it should be about 87% of valve size, in this case would you use an seat insert to get to the right ratio?

john

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...