[Diy_efi] Several basic questions from a newcomer to EFI

Tlsalt at aol.com Tlsalt at aol.com
Thu Oct 10 00:57:27 GMT 2002


Adnan wrote,

For a while now I've toyed with the idea of building an EFI system for my
Austin-Healey 3000, a vehicle that I'm doing a complete restoration on.   It
has a 3-liter straight six with twin semi-downdraught SU carbs.   While the
engine is a terrific one, with loads of low-end torque, the intake manifold
design leaves a lot to be desired.   Mixture uniformity is poor as there are
a few right-angle turns that the mixture has to endure.

I am leaning towards using a MAP sensor to measure load.   While I'd love to
go MAF, I don't really want to get into redesigning the intake manifold
right now.   With injectors and the existing plenum, I think most of the
mixture issues will go away.   Now the questions:

1)  Is fuel metering to MAP purely an empirical process?   How would I
go about making at least first-approximation calculations to determine fuel
needs based on MAP, RPM and inlet air temp?

2)  Do I need an ambient pressure sensor (BAP)?   Would a Gauge Pressure
Sensor do the job of MAP + BAP?   Is it best to plumb the MAP directly into
the intake or keep it remote?

3)  Is there any value in adding TPS?   Would it allow for better
enrichment algorithms or is MAP quick enough?

I also have a few ignition-related questions:

4)  I see that some off-the-shelf systems use a Bosch ignition module.
If I use a darlington switch, can I not perform all dwell control through
the software?

5)  With a distributor system, dwell is angle dependent; what's the rule
of thumb for dwell time as with a processor I can now have fixed dwell,
regardless of RPM?   I plan on using a direct ignition system with three
coils and wasted-spark.

6)  What's the best (and easiest) way to figure out the advance curve
for the existing distributor?   The vacuum advance is probably going to be
more difficult, I'd guess.

Is there a fuel pump system that does not need a return line to the tank?

Hello Adnan,

I converted a Jaguar 3.8 six to EFI and can give you a few thoughts. For 
manifold/throttle bodies you can use SU bolt pattern TB's from Jenvey that 
hold two injectors each, machine your manifold for six port injectors, or get 
a Weber manifold and use (3) twin TB's from Jenvey or TWM.  Six 19 lb/hr 
injectors should be about right.  Use a Bosch pump from an L-jet motor and 
the adjustable Bosch fuel pressure regulator, it is much cheaper and easier 
to use existing parts that to attempt a returnless system.  I used a Haltech 
ECU with datalogging.  The initial fuel map was pretty easy.  You can roughly 
calculate pulse width at peak torque and shape the fuel curve at each 500 rpm 
interval for 32 load points (MAP).  I am using a GM 1 bar MAP sensor plugged 
into the manifold.  TPS is used for acceleration enrichment, the Haltech has 
adjustable enrichment and sustain.  This really helps if you have a 
dentonation prone motor.  You should use an advance timing light to plot the 
centrifugal and vacuum advance curves of your existing distributor.  Look 
around and try to find all of the existing factory power, torque, BSFC and 
ignition timing charts for your engine, they will give you a big head start.

Paul

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi



More information about the Diy_efi mailing list