Yowee these valves are expensive.  But, this is a seriously important
item, so good quality is important.  I assume the '6-port' means you will
route the return fuel back to the same tank?  If so, this makes fuel mgmt
pretty simple.  Is this a manual valve or electically operated?  A/C or
auto?  ( I assume A/C).  Sounds like a good solution if I understand this
correctly.

Bill
Long EZ builder & pilot
mechanical engineer

Bill,

  The discussion on fuel system design is starting to make me a little
nervous about my planned system. My planned design evolved following
previous discussions here about fuel systems. The airplane is a low wing
(RV4) with 1 tank per wing and EFI. It is different from the norm (and
unproven) and relies on automotive style electric selectors. This has been
discussed before but I would like to revisit it just to get a warm fuzzy
feeling about it.

 I've got to say that I mistrust electric selectors.  With a manual selector as
the valve develops stickiness (and most seem to with time) you can  (1) feel
it and decide if it is getting serious and fix or lube the valve  (2) add more
force above the normal to make it work even if deteriorated a bit.  Electric
acuators do not give feedback on valve condition and cannot overcome a
bit of extra resistance.  They work until the don't.  Binary.

  The plan is to use one hi press pump per tank located as close as
possible to the tank pickup. I have been told by a knowledgeable mechanic
who works on EFI systems that the typical pumps used arent designed to draw
fuel when mounted any appreciable distance from the tank and as a result
should be as close to the tank as possible. Ive seen the same
recommendation in the installation instructions for the 2 aftermarket EFI
systems Ive installed in cars so this seems to be a wise choice.

These are often referred to as 'flooded inlet' pumps as they are not designed
to pull any vacuum on the inlet.  Pulling vacuum is very bad with fuel as this
is a primary cause of vapor locking.  The auto mfgrs have almost all gone to
pumps in the tank (the ultimate flooded inlet) so ALL the fuel system is under
pressure to reduce vapor locking.  In the bad old days the fuel pump was at
the engine and sucked the fuel from the tank.  Vapor locking was common
on hot weather in traffic with low fuel flows.

 I am

planning to "T" the outlets of the pumps together and allow the built in
pump check valves to prevent cross feeding from 1 tank to the other. From
the "T" the fuel goes thru the fuel rail and then to an automotive style
electric selector valve for return to the proper tank. Tank selection is
accomplished by a single switch which turns on the selected pump and
properly positions the selector valve.

Sounds OK, but what about having the return go to a T at tank top level and
then let it go to both tanks.  The C185 does this, and apparently it works,
altho
I'm not sure of the relative qty of bypass fuel between the Continental mech
FI and the EFI's being ginned up by homebuilders.  I suppose it depends a lot
on what fuel pump is selected.  If the return lines are large I don't see why
it wouldn't work, even if one tank fills more (likely), it should send all the
fuel to the other once the low resistance tank is full (resistance goes up).

  An alternative to this would be to use a ganged selector (electric or
manual) rather than relying on the check valves. An electric valve has been
identified here previously and there is a good manual valve available from
Andair (in the ridiculous price range mentioned previously).

  I will need to have an additional line between tanks to allow for
crossfeed in case of pump or selector failure.

Hmm.  Is this only for the alternative or for both systems?  Seems like you
are adding a lot of complexity with the crossfeed system.  Personally, I'd
rather plan on landing if a pump fails and fix it while keeping the fuel system
simpler.  Burt Rutan says "Simplicate and add lightness"  - I agree.

  Going with all electric valves in such a system makes it possible to keep
all fuel system components out of the cockpit which to me would seem to be
a worthwhile goal. I know of 2 fatalities in 2 seperate incidents in the RV
community in the past year or so that were due to fire in the cockpit. It
also makes it possible to automate fuel management (using a simple micro
controller) at some future date to reduce pilot workload.

Fuel out of the cockpit is a laudable goal.  Unfortunately, you can cause
problems by sticking to this to the extreme.  John Denver's crash was
(apparently) partially due to the builder insisting on having the fuel selector
valve and lines aft of the firewall.  This led to the selector being at the left
shoulder level to enable the long torque tube to operate the valve.  When
Mr. Denver ran a tank dry, he had to release the stick and reach to an
awkward position to try to change tanks.  Reports are that the valve was
sticky, too.  Probably rolled over while working on the valve and hit at
high speed. Of course running a tank dry at low alt is bad business anytime.

  A more conventional system would bring both tanks together through the
Andair selector, then to 2 paralleled hi press pumps (for redundancy) then
on to the fuel rail, back through the selector and to the selected tank. Of
course this forces the pumps quite a ways downstream of the tank and makes
it impossible to keep the fuel system components out of the cockpit. This
would seem to be my fall back design if some one can point out a
significant flaw in the system described above.

Sounds OK, except for the electric selector, which (as I said) can probably
be replaced with a T. I'd probably want to return to the outboard upper
edge of the tanks to minimize flow resistance for the return fuel & vapor.

Are you sure the pumps can't reverse flow?

  So what is your opinion on this? All opinions welcome.

Mike Wills

Long EZ builder & pilot
mechanical engineer

 
The Aircraft Rotary Engine Newsletter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org