Paul,

<snip
With the discussion on-going about in tank fuel pumps and fuel
return lines, I am asking if I need to make accomodation for
anything special. Presently, my tanks have no provision for a fuel
return or in tank pump. Would it be necessary to design in any of
this stuff?

Thanks, C. Smith

C. Smith -

I have offered to build and test an EFI PWM Pump design utilizing in-
tank, submerged pumps, but I'd be more than happy to test an external-
pump design afterwards, probably using the Walbro pumps that Tracy
sells.  (He and many others have proved their reliability.)  I assume
that would work for you.

One goal of this project is to design a RETURNLESS EFI pump system.
Other goals are reliability, redundancy, and low cost.

The usual caveats apply for EFI pump placement.  You will want to
place them AS LOW and AS CLOSE as possible to the fuel tank output
line.  NOTE: EFI pumps DO NOT LIKE TO SUCK - they will simply create
fuel vapors and bubbles, lose prime, and spin to destruction.  You
might hear the pump suddenly "spin up" when this happens.  Even a rise
of many inches can cause this effect with warm fuel (ask me how I know).

Needless to say, I disagree with Paul about putting the pumps on the
firewall.  That can create too much "sucking" during climb out - just
when you need absolutely reliable fuel delivery.  (Hence, my in-tank
submerged preference.)

Regarding you "designing in any of this stuff" - that's up to you.
The RV series has a wet wing tank with a round access port on the
inboard side.  I plan to mount the EFI pump on the inside of that
removable access panel.  We'll see what EFI pump parts are available
as the project progresses ...

-Vince Orton

Some suck better than others. The MSD is positive displacement
while most are centrifugal (Walbro) and don't suck worth a damn.

Paul Lamar


If you are going to run a returnless system with fuel injection you are
going to want to test pressure spikes and drops. It is almost standard
to run a pulsation damper in the rail to make sure the pressure loss in
the rail due to injection stays consistent for consistent pulse width
delivery. I dealt with this on Korean cars we tuned since they like to
use cheap stuff. In really cold environments you can run into weird issues.

Rob Woods