Hello Paul,
 I am trying to determine the size of the return lines and volume of fuel
returned as I plan for my EFI 13B.  I may have to create a header tank to
handle
the return and the question is, how big does it have to be?  Any rules of
thumb?
 Thanks,
 William Hagood

 Most people use AN6 lines for fuel.
 The volume of fuel returned will be inversely proportional to the HP
being generated. At idle and taxi it could be 80% of the fuel
pumped. So the header tank will fill rapidly. The question then
becomes what do you do with the header tank overflow?
 At take off and climb the fuel pumped is a max so the flow to the tank
may be as little as 10 to 20 percent of the fuel pumped.


Done wrong fuel can be pumped over board in short order through
the tank vents.
 The traditional and safest way is use 2 tanks and a  two level
valve and send the return fuel to the tank it is pumped from.


A header tank, if used should be able to flow fuel back
into the tank it came from. One gallon should be sufficient.
 Defective fuel systems are a major cause of engine
stoppage and crashes. Particularly in low wing airplanes.
They are tricky to design and they appear simple but are tricky
to do right. The lowly vent system design is very important.

 One of the big mistakes is pointing  the vents in the
direction of flight as dynamic pressure above 100 MPH
will push the fuel out of the tank in a direction
you may not want it to go. That is the same pressure fed
to the air speed indicator by the pitot tube.
 Another mistake is having the vent lines coming out below
the level of the fuel causing a siphon under certain
circumstances.

 A third mistake is using a LEFT, RIGHT, BOTH & OFF
fuel selector valve on a low wing airplane.
 Low wing airplanes should use only a LEFT, RIGHT & OFF
fuel selector valve.
 The best and safest way is to copy certified aircraft
fuel injection systems if possible.
 Paul Lamar

Paul,
The engine test stand you will get has an automotive tank with submerged
fuel pump, just as it came out of the car, no return, so you will get an
opportunity to see how that works. One of the last things that we were
adjusting was fuel pressure with the engine running, you may have to tweak
that a little. milo

OK Milo, We can handle it. No worries mate.

Paul Lamar

Returnless systems were designed to reduced evaporative emissions to pass EPA testing.

Two methods of dealing with this. Constant fuel pressure, and monitor the engine vacuum. You'll change the Duty cycle (one time) on the injectors to compensate. This is one of the functions why Ford uses a floating point processor for fast calculations as it has to be done in real time to meet emissions. Another method is to use a pressure sensor on the fuel rail and adjust on time of the fuel pump to keep it a constant above engine vacuum. Single rail doesn't allow the purge of air, so return system is best.

Simple return I used on the Mazda Rally Car.... "T" fitting. Brass T and put a 0.050" carb jet soldered in place. Purged air, no vapor lock issues, little fuel returned but worked like a charm.

Paul, with the returnless system you could use a summing amp and linear circuit to adjust PWM based on engine vacuum. Linear computers rule!  Yes, when I worked at Bosch we were still running early (analog) L-Jetronic systems on the line in Reutlingen!  If you would have told me 20 years later we would need floating point to calculate fuel and spark, I would have laughed my head odd. Grin!

Tom W