Tracy,
Paul posted a picture of his sump with this message:
http://lancaironline.net/lists/flyrotary/Message/17071.html?Language=
I'm including it again here attached to this message.
Yes, the width looks like it could cause a problem in turns if it
wasn't full. Also, it's location just below the spar bothers me, it
appears to be way too high, meaning the main tanks may not drain to it
correctly. My EFI tank is much lower, at the very bottom of the
firewall where the gascolator is mounted in the Long-EZ. My tank was
made as small as possible, as small as it could be and still hold the
two Mazda pumps. It is cubical about 6 to 7 inches on a side.
Perry, what did you think of the physical configuration of
Paul's sump/header tank? (short, wide & narrow). Don't know
if it was a factor in the accident but another problem prone detail I
think.
Tracy
If that's the definition of
a sump system, then I do not have a sump system. I have a small EFI
tank with two internal EFI pumps, fuel return to this EFI tank. This
EFI tank is located where the gascolator is in the stock Long-EZ fuel
system. I have the complete stock Long-EZ fuel system otherwise with
L-R-OFF selector (no BOTH). I will not refer to my system as a sump
system in the future, to avoid misconception (and arguments).
You've said it yourself Jim in the past, that you've had many problems
with your sump system. Maybe sump systems and BOTH selectors just don't
belong on Rutan derivatives, as RAF itself has said. As you saw, a
Lycoming-powered EZ had an accident in 1995 just after takeoff (sounds
like Paul?) with such a system.
Perry
Jim S. wrote:
A sump tank renders L-R(-B) selector redundant and superfluous. L and
R selectors allow the fuel pump to "suck" fuel from the selected tank.
When the tank is empty, you select the other tank. If you have "Both"
selected and one goes empty, the pump sucks air through that line.
Either way, the engine quits when one tank goes dry. You get power
back by selecting the tank that has fuel in it. The main disadvantage
with a "Both" position has to do with distraction and delay - when the
engine quits, you have to figure out which tank is dry and which way to
turn the knob to select the tank that has fuel. Takes time and is a
definite failure mode. Gravity feeding a sump through big enough lines
eliminates the need for valves. That's the main purpose of a sump.
Selectors are for when you don't have a sump. They have no place on a
system that does. The Rutan derivatives with a sump have no selector;
those with a selector have no sump. Paul's setup was unique.
Just a theory ... Jim S.
rijakits wrote:
The only safe way to have a "Both"-selection for a low wing would
be (IMHO), to have a low pressure pump in each tank feed a fairly large
sump ( 2 gallons +) with a system as described by Georges below:
EFI pump picking up at the bottom of the sump, with a return to
the top and baffels, not just separating the top from the bottom, but
also side from side, so when a tank pump starts to pump air into the
sump, it can vent to the top of the sump and out through the vent,
which would have to go to both main tanks.
With some additional cross-feed line from one tank to the other
with a manual valve, one could use one tank pump as the back-up for the
other:
Open the crossfeed valve and slip the plane accordingly for a
while. That should push the fuel into the other tank, in case one of
the pump fails.
One question remains: How would you prevent or equalize overflow
from the sump to the tanks?
Thomas J.
----- Original Message -----
Sent: Sunday, May 29, 2005 8:07 PM
Subject: [FlyRotary] Re: will EFI pumps pump air was Re:
Fuel Tank Selection
Group FWIW
Call it a sump or header tank, if its large enough say 3 liters
oops! I mean quarts with both pumps picking up fuel at the bottom 1/4
of the tank with the fuel return pointing at the cover to avoid
aerating the fuel in the header (possibly a baffle below it & a
vent pipe to each tank) Gravity feed would work in a high wing , in a
low wing I think you still need a low pressure transfer pump to push
fuel to the header tank from the mains. I don't think it is safe to
have "Both" position in the fuel selector except in a gravity system
with carburetors, not with EFI, the possibility of "sucking air "is
unavoidable without check valves ( more weight, more cost & more
possibility of problems).
Georges B. ( Not really stirring the pot, maybe just a little)
-------Original Message-------
Date: 05/29/05 17:42:01
Subject: [FlyRotary] Re: will EFI pumps
pump air was Re: Fuel Tank Selection
Now I'm not so worried, because each high pressure
fuel
pump draws from it's own tank and the only point of inter- connection
is where the lines join at the fuel rail(s).
This
brings up a question that I've had before, and I'm not sure this is
exactly what anyone is doing, so it's not meant that way.
Say
you have two tanks, with an EFI pump for each tank. You then connect
the output of each pump together, feeding into one line running to the
fuel rail. The question is: What happens when one tank runs out of
gas? Will the EFI pump move enough air through it to disturb the fuel
rail pressure that's being delivered from the other pump, or would it
just stop pumping at that point, and do no harm (other than maybe
burning the pump up eventually)?
Yep, still trying to figure out
how to fix my fuel transfer system.
Cheers,
Rusty
leak test.jpg
|