chapman-era
[Top] [All Lists]

Re: elan lust abatement

To: Keith Gustafson <gusmach@shore.net>, Eric Camiel <ecamiel@i84.net>
Subject: Re: elan lust abatement
From: "Erik V. Berg" <erikb@elrond.sp.TRW.COM>
Date: Fri, 10 Apr 1998 14:29:31 -0700
Keith asked:
> EFI huh, sounds interesting. What system does it use?

Meanwhile, Ron Bennett pointed out (thanks Ron):
>> Car can be viewed at: http://idt.net/~lotusf1/elan.htm

Eric Camiel also responded:
> Nice pictures! Does anyone have info on the fuel injection 
system?

Well, it is an amalgamation of parts from various different
sources.  Fred Miranda designed the thing himself.  Fred has 
considerable hands-on experience with custom EFI development.  

It is a single throttle body type system, using a very nice 
looking one-off Aluminum plenum / manifold that Fred welded up 
from scratch.  The remaining list of components is by memory 
(my notes are at home) but I believe Fred said he used a Saab 
throttle body, Autronic computer, Toyota high pressure fuel 
pump, Mitsubishi ignitors, etc.  I forgot what type of 
injectors, which is kinda important, so (mental note) I need 
to remember to ask Fred this question.

The Autronic computer uses a MAP type algorithm, and it has a 
serial interface which can be used to tweak the look up table 
data, from a laptop computer.

It uses a crank fire ignition system, with no distributor.
There is a cam position sensor (which actually replaces the 
distributor) and a crank position sensor.

Using this system, the engine is quite flexible and smooth 
running at any RPM, for moderate all the way to WFO throttle 
conditions, but it does have a characteristic anomaly that 
Fred has yet to get ironed out... When warmed up, and at low 
RPM and light throttle, it exhibits a "hunting" mode, where 
it wants to cycle up and down in RPMs.  

This is a broad side of the barn guess, but I suspect that 
there may be a minor instability in some part of the control 
algorithm.  It could have something to do with the fact that 
this motor has an exceptionally light flyweel, along with the 
characteristic time constants of the manifold pressure (or 
other?) sensors. 

When I get a chance, I may to try to debug this problem
myself.

Regards,
Erik


<Prev in Thread] Current Thread [Next in Thread>