Lucas PI

[Erik Quackenbush]

I've got a PI TR-6 and have considered doing something similar. Were you
planning to retain the existing control unit with the datum track, etc.?
I've though about keeping the metering distributor (rotor shuttle) but
gutting the control section and installing a computer controlled servo of
some kind inside it (I, um, have a spare). ;{)

-Erik


At 08:48 PM 6/3/00 -0600, you wrote:
>Ages ago, I submitted a question to the folks here about barometric
>compensation on an old Lucas mechanical PI (petrol injection) system,
>and I received a couple of helpful answers based on what I proposed to
>do to solve the problem (minor adjustment of the unit's datum cam to
>compensate for altitude differences).
>
>That was before I had the unit in hand, disassembled, and had
>considered, carefully, how the unit worked mechanically. It's a horribly
>and unnecessarily complicated arrangement of adjustments to do,
>essentially, one thing--move the metering unit pintle to adjust mixture.
>The only real variable input to the entire system is vacuum. (!)
>
>What I am contemplating is to do some sort of series loop input to add a
>MAP to calculate vacuum, air temp sensor, and perhaps include an engine
>speed signal on another line. If it turns out that barometric
>compensation isn't well-managed by the MAP, to also put a barometric
>compensator in the loop, then have a Motorola 68HC11 controller make the
>outputs to a servo motor actuating the pintle actuator barrel in the
>metering unit. 
>
>The question here is about compatible components. I'd like to be able to
>find sensors compatible with each other, i.e., all highs and lows
>compatible in the right direction--all devices register highs for more
>fuel, lows for less. It's just a way of minimizing inputs and making the
>system simpler, since there's not a lot of room in the RAM available on
>the board I found for fuel maps, nor enough clock rate for lots of
>calculations to compare input to a complicated fuel map.
>
>I could probably work out inverter circuits, etc., upstream of the A/D
>inputs, but it would be simpler to just input the loop result to the A/D
>port on the controller board (limited number of ports available) and
>program for just that, and, as said, perhaps an engine speed signal. 
>
>Funny thing. The control system on the Lucas system is a bit of a
>horror, but the metering unit itself is reliable and fairly free of
>difficulties, and the nice thing is that it is fully sequential,
>mechanically. 
>
>Yeah, I know, why not go digital, entirely? But, it's an older Triumph
>and I wanted it to be more stock-looking than not, and also wanted to
>take it to a show in 2001, which is being held at 9800 ft. altitude....
>(!) The car's younger sister (brother? niece? nephew?) is getting the
>digital treatment and a turbo.
>
>Cheers, and thanks.
>
>-- 
>
>Michael D. Porter
>Roswell, NM
>[mailto: mporter at zianet.com]
>
>`70 GT6+ (being refurbished, slowly)
>`71 GT6 Mk. III (organ donor)
>`72 GT6 Mk. III (daily driver)
>`64 TR4 (awaiting intensive care)
>----------------------------------------------------------------------------
>To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
>in the body of a message (not the subject) to majordomo at lists.diy-efi.org
>
>

--
Erik Quackenbush, V.P. Operations, Midwest Filter Corporation 
1-847-680-0566 fax: 1-847-680-0832 http://www.midwestfilter.com


----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org