REFLECTOR: FUEL FEED

[Al Gietzen]

This issue of fuel flow and vents troubles me because it is difficult for me
to believe that there is simpler, more reliable fuel system setup than
gravity feed from two tanks to a sump tank, and a sample drain at the bottom
of the sump.  Gravity never fails.  If the system isn't working; we're doing
something wrong.  Here are my thoughts on conditions that must be met:

 

1)    Main tank vents from the highest point in the tank, continuous upward
slope to the manifold point that is well above the upper level of the tanks.
You do not want fuel in the vent lines.

 

2)    Feed from the bottom of the tank that slopes continuously downward to
the sump tank, with no restrictions, or screens finer than a screen door
screen with at least 2-3 sq. In of area.  The feed from the two tanks should
be close to symmetrical; same size, same length, so the flow characteristics
are the same.

 

3)    Fuel caps that don't leak.  

 

4)    Airplane trimmed to fly straight and level.

 

5)    Vent system exit in ambient (or higher) pressure area.  The exit of
the vent should probably extend at least 1 1/2 inches out the bottom of the
fuselage to get through some of the boundary layer, and be bent into the air
stream to be sure of positive pressure.  A 1/8" hole at the back of the bend
can help alleviate the concern of plugging, but if the check valve idea
works, go for it.

 

Further thoughts: Except in the case of a cap leaks, a 1/4" vent line to the
tanks is fine - the makeup air flow required for the fuel flow rates we're
talking about will easily flow through a 1/4" line.  Tank outlet of 3/8"
will handle something close 20 gph from each tank.  I know this because
during engine tests I did static runups on my engine at WOT running fuel
from one tank (with only 5 gal in the tank), and the low fuel warning light
never came on. I have 1/4" vent line to the tank. (Dyno tests at same
conditions showed 19.2 gph).

 

About the sump tank vent - 1.) the tank must be vented to allow it to fill,
2.) because of the low vapor pressure of fuel (particularly auto fuel) you
do not want to depend on "sucking" fuel anywhere because of the possibility
of forming vapor at the pump (vapor lock); 3.) if you have fuel injection
with recycle flow back to the sump, you want a sump vent for air or vapor
bubbles from the return line.

 

Further, it seems to me; if you vent the sump for it to fill, and then close
the vent, you eliminate gravity flow from the mains because you have fixed
the level in the sump (somewhere up the vent line) at whatever the level was
in the mains at that time, and you now are depending on the slight negative
pressure created by the fuel pump to draw the fuel from the tank. Not the
best situation.

 

If keeping the caps leak tight is so difficult, we must be using the wrong
caps.

 

I can think of no explanation for the problems the Jim S. experienced
because he tried every fix, but there is something wrong somewhere.
Possibly different flow characteristics from the two lines; and/or less than
ambient pressure at the vent exit.

 

The blinking of the low level warning float switch at high altitude is a
mystery to me.  For those of you who have observed this; is it more likely
with high, or low, levels in the main tanks.

 

As near as I can figure; it seems that if the stock system is properly
executed, and it is not working properly, something is wrong and we should
find out what it is.

 

FWIW,

 

Al

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.tvbf.org/mailman/private/reflector/attachments/20050219/634c1f66/attachment.htm