 |
Matronics Email Lists Web Forum Interface to the Matronics Email Lists
|
View previous topic :: View next topic |
Author |
Message |
frank.hinde(at)hp.com Guest
|
|
Back to top |
|
 |
glcasey(at)adelphia.net Guest
|
Posted: Tue Sep 12, 2006 6:18 am Post subject: Fuel Pump Switch(es) |
|
|
I might have missed some of the discussion, but the comment below seems to be directed at the tank valving method - specifically whether to use two separate fuel valves or one, with the recommendation to run all the fuel through one valve. I chose to put two valves on mine for the following reasons:
1. A single valve is a single point of failure - what if the handle breaks off or the valve sticks? The most likely time for that to happen is when one port is being closed off before the other opens (apparently most or all fuel selectors go to "off" between right and left positions). If it jams in this position all fuel is shut off. My on-off valves are completely independent and controlled from their own lever and cable. If one jams either open or shut the other is unaffected.
2. The fuel can be shut off at the tank, not in the middle of the cockpit. When both valves are shut off no fuel can enter the cockpit, which would be reassuring in case of an off-airport landing.
3. It reduces the number of fuel fittings and therefore the number of potential leak paths, especially in the cockpit.
4. It allows the backup electric pump to be placed at the lowest point in the fuel system which should provide the best protection against vapor lock.
5. When switching tanks the new tank can be turned on before the old tank is shut off, guaranteeing a continuous flow of fuel.
6. A trivial one: If the plane is landed with an unbalanced fuel load, both valves can be left on, equalizing the fuel load before the next flight.
The disadvantage is that both valves could be inadvertently turned on, which on a low-wing plane means that if one tank is run dry in that condition the engine could draw air. In that case one fuel gage would read empty and changing the fuel valve positions (shutting off the empty tank) would correct the condition. Interestingly, my DAR was skeptical about the arrangement, but was satisfied by the placement of a placard "continuous operation on both tanks prohibited." Conversely, my test pilot like the arrangement, saying that for at least the first half tank the engine could be operated on both, keeping the fuel load balanced (well, anyone that has had a Cardinal would not be convinced that would happen). Why do high-wing planes often have a "both" position? Probably because the fuel is joined together at the bottom of the plane, several feet below the tanks. One tank would have to run dry with the other pulling enough vacuum to overcome maybe 4 feet of head pressure - very unlikely. A low wing plane has nowhere to connect the tanks together except essentially even with the bottom of the tanks. When one runs dry air would immediately enter the engine.
Just my nickel's worth (inflation, you know)
Gary Casey
On Sep 11, 2006, at 11:55 PM, AeroElectric-List Digest Server wrote:
[quote]
George- "... an on/off valve on each
tank introduces problems and complexity..."
Boy, am I glad you wrote that! I completely missed the part about
independent fuel valves. Kevin, besides all the good points George makes,
and as someone pointed out a while back on one of these lists, you will
only find a "BOTH" position on a fuel selector in a high wing single engine
airplane. The low wing planes have left, right, and off. With a high wing
plane, you will sometimes see quite an imbalance develop when drawing fuel
from both tanks simultaneously, even with the head pressure caused by the
tank location. This effect could be much greater in a low wing
installation. If you had the ability to have both tanks plumbed to the
engine at the same time, when one tank emptied you could suck air into the
system. Obviously, this would not be good. I would definitely recommend
against two fuel valves.
Sorry for not catching that sooner-
glen matejcek [b]
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
frank.hinde(at)hp.com Guest
|
Posted: Tue Sep 12, 2006 6:46 am Post subject: Fuel Pump Switch(es) |
|
|
Not switching the valve is also one of the advantages of the fuel pumps in the wingroot method. In this case the lines are simply joined together in a tee fitting before going to the on /off valve.
In this installation the valve is only turned off in the event of an emergency, never in normal operation.
True there is pressurised fuel in the lines in the cockpit...Hopefully in an off field landing one would remember to turn off the pumps.
With dead (roller vane FI pumps) pumps the amount of fuel that could trickle through under gravity would be very small, certainly less than the standard arrangement.
Incidently my FAA inspector looked very skeptical for a few minutes. Not surpring this guy had built like 5 RV's and this was very different. Thought my "goose was cooked" for a moment but he signed it off once he understood the system.
Frank
From: owner-aeroelectric-list-server(at)matronics.com [mailto:owner-aeroelectric-list-server(at)matronics.com] On Behalf Of Gary Casey
Sent: Tuesday, September 12, 2006 7:16 AM
To: aeroelectric-list(at)matronics.com
Subject: Re: Fuel Pump Switch(es)
I might have missed some of the discussion, but the comment below seems to be directed at the tank valving method - specifically whether to use two separate fuel valves or one, with the recommendation to run all the fuel through one valve. I chose to put two valves on mine for the following reasons:
1. A single valve is a single point of failure - what if the handle breaks off or the valve sticks? The most likely time for that to happen is when one port is being closed off before the other opens (apparently most or all fuel selectors go to "off" between right and left positions). If it jams in this position all fuel is shut off. My on-off valves are completely independent and controlled from their own lever and cable. If one jams either open or shut the other is unaffected.
2. The fuel can be shut off at the tank, not in the middle of the cockpit. When both valves are shut off no fuel can enter the cockpit, which would be reassuring in case of an off-airport landing.
3. It reduces the number of fuel fittings and therefore the number of potential leak paths, especially in the cockpit.
4. It allows the backup electric pump to be placed at the lowest point in the fuel system which should provide the best protection against vapor lock.
5. When switching tanks the new tank can be turned on before the old tank is shut off, guaranteeing a continuous flow of fuel.
6. A trivial one: If the plane is landed with an unbalanced fuel load, both valves can be left on, equalizing the fuel load before the next flight.
[quote][b]
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
khorton01(at)rogers.com Guest
|
Posted: Tue Sep 12, 2006 5:18 pm Post subject: Fuel Pump Switch(es) |
|
|
On 12 Sep 2006, at 10:15, Gary Casey wrote:
Quote: | I might have missed some of the discussion, but the comment below
seems to be directed at the tank valving method - specifically
whether to use two separate fuel valves or one, with the
recommendation to run all the fuel through one valve. I chose to
put two valves on mine for the following reasons:
1. A single valve is a single point of failure - what if the
handle breaks off or the valve sticks? The most likely time for
that to happen is when one port is being closed off before the
other opens (apparently most or all fuel selectors go to "off"
between right and left positions). If it jams in this position all
fuel is shut off. My on-off valves are completely independent and
controlled from their own lever and cable. If one jams either open
or shut the other is unaffected.
2. The fuel can be shut off at the tank, not in the middle of the
cockpit. When both valves are shut off no fuel can enter the
cockpit, which would be reassuring in case of an off-airport landing.
3. It reduces the number of fuel fittings and therefore the number
of potential leak paths, especially in the cockpit.
4. It allows the backup electric pump to be placed at the lowest
point in the fuel system which should provide the best protection
against vapor lock.
5. When switching tanks the new tank can be turned on before the
old tank is shut off, guaranteeing a continuous flow of fuel.
6. A trivial one: If the plane is landed with an unbalanced fuel
load, both valves can be left on, equalizing the fuel load before
the next flight.
The disadvantage is that both valves could be inadvertently turned
on, which on a low-wing plane means that if one tank is run dry in
that condition the engine could draw air. In that case one fuel
gage would read empty and changing the fuel valve positions
(shutting off the empty tank) would correct the condition.
Interestingly, my DAR was skeptical about the arrangement, but was
satisfied by the placement of a placard "continuous operation on
both tanks prohibited." Conversely, my test pilot like the
arrangement, saying that for at least the first half tank the
engine could be operated on both, keeping the fuel load balanced
(well, anyone that has had a Cardinal would not be convinced that
would happen). Why do high-wing planes often have a "both"
position? Probably because the fuel is joined together at the
bottom of the plane, several feet below the tanks. One tank would
have to run dry with the other pulling enough vacuum to overcome
maybe 4 feet of head pressure - very unlikely. A low wing plane
has nowhere to connect the tanks together except essentially even
with the bottom of the tanks. When one runs dry air would
immediately enter the engine.
|
You are worried about fuel valve handles breaking. With your design,
what happens if you have run one tank down low, want to switch to the
other tank, and the valve handle on the tank that is feeding breaks
so you can't close that valve? Won't the engine quit once that tank
is empty and the engine starts sucking air? How is this better than
a valve handle failure with the normal design?
Kevin Horton RV-8 (finishing kit)
Ottawa, Canada
http://www.kilohotel.com/rv8
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
glcasey(at)adelphia.net Guest
|
Posted: Wed Sep 13, 2006 4:52 am Post subject: Fuel Pump Switch(es) |
|
|
Yes, there are still a number of failure modes that could create a problem, even with two independent fuel valves. In the case mentioned below, the low tank valve breaks and won't shut off. In that case the (presumably fuller) tank could be turned on, feeding fuel from both tanks and the plane landed at the earliest convenience. The fuller tank would probably back-feed into the lower tank, extending the range compared to just feeding off the lowest tank, depending on the level of fuel in the fuller tank. In any event one has to assume there is only the fuel left in the lowest tank. If the valve for the "new" tank failed to open (broke in the closed position) the response is simple - land before the low tank goes empty. Not a perfect condition, but there are still options available that might not be available with a single valve. If a single valve broke in the "low" tank position fuel in the fuller tank is completely unavailable. If it sticks and breaks between tanks (what I think is a more likely failure) neither tank is available and the engine quits right away. The normal operating procedure with two valves is to open the valve for the new tank, confirming that it actually moved, and then shutting off the valve from the old tank. Never shut the old one off before turning the new one on. Also, I know that there have been some postings advocating, or at least acknowledging, running a tank dry as a routine operating mode. I suppose if there were more than two tanks this isn't so bad, but with only two tanks doing that gives up one of the redundancies in the system - two independent sources of fuel. I personally don't like the idea of running a tank dry. My own standard is to never run either tank lower than what is required to get to the nearest airport. FWIW.
Gary Casey
Lancair ES
On Sep 12, 2006, at 11:55 PM, AeroElectric-List Digest Server wrote:
[quote]
You are worried about fuel valve handles breaking. With your design,
what happens if you have run one tank down low, want to switch to the
other tank, and the valve handle on the tank that is feeding breaks
so you can't close that valve? Won't the engine quit once that tank
is empty and the engine starts sucking air? How is this better than
a valve handle failure with the normal design?
Kevin Horton RV-8 (finishing kit)
Ottawa, Canada [b]
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
Bill Denton
Joined: 10 Jan 2006 Posts: 97 Location: Chicago, IL USA
|
Posted: Wed Sep 13, 2006 5:17 am Post subject: Fuel Pump Switch(es) |
|
|
Just an observation...
While you're sitting there with your head down troubleshooting valves and breaking the handles off and flipping switches until they fall out of the panel, you may well end up sticking your airplane in the ground.
I know that it is sometimes necessary to have multiple tanks, valves, and pumps to manage complex fuel systems. But some of what I've been reading on this thread sounds like it's coming from graduates of the Rube Goldberg School of Engineering...
--
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
gcomfo(at)tc3net.com Guest
|
Posted: Wed Sep 13, 2006 5:17 am Post subject: Fuel Pump Switch(es) |
|
|
[quote]
--
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
N6030X(at)DaveMorris.com Guest
|
Posted: Wed Sep 13, 2006 5:47 am Post subject: Fuel Pump Switch(es) |
|
|
I have a little different point of view that I'd like to just throw out there.
In my experimental Dragonfly, I was gripped for a long time by a
NASA-like desire to have multiple redundant systems everywhere. I'm
a software guy, so I love complicated, microprocessor-controlled
everythingies. But when I started working through the cockpit
workload part of dealing with multiple independent pumps and switches
and bypasses for everything and backup this and that, I realized that
there was a single point of failure that was getting more and more
overloaded with every new complexity: the pilot (me).
Look at the NTSB reports, and you'll see there are many more failures
of the human to throw the switch than of the switch breaking. Fuel
Starvation due to forgetting how much fuel is in the tank, not
knowing how much fuel is in the tank, forgetting to switch tanks,
taking off without adequate fuel, etc. etc. I don't remember seeing
too many "the fuel selector handle broke off between tanks". The
oldest airplanes still flying anywhere (the survivors of all our
human failures) have the simplest systems. Take a look at how they did things.
My advice, which could be worth exactly as much as you're paying for
it is this: build it light, build it high quality, and build it
simple enough that you can intuitively grasp it without schematic
diagrams and lengthy troubleshooting charts while handling an
inflight emergency. If you have requirements such as "don't ever
turn off this tank before turning on that tank", I think you're
starting to set yourself up for failure. Think Gravity Feed from a
header tank wherever possible. Instead of buying two redundant
valves and plumbing, buy one high quality valve and replace it every
2 years just for the helluvit. What's that gonna cost you? A
hundred bucks? How much weight will you save, and how much cockpit workload?
When you're doing your failure mode analysis, include a list of all
of the things the pilot has an opportunity to do wrong. We try to do
this in software development, and it's a useful exercise before you
get carried away anticipating equipment failures that are just never
going to occur. When designing software, we no longer worry about
whether there will be a bit error on the memory board, or whether the
disk drive will crash a track. But we spend a lot of time worrying
about whether the user will forget to enter a value or will enter the
wrong value. If the user (pilot) CAN do anything wrong, he
WILL. Try applying that to your aircraft design and see if you come
up with some interesting scenarios you had not thought of before.
Dave Morris
1960 Mooney M20A
Manual gear retract
3 fuel tanks - single valve
No fuel system failures in 46 years (knock on wood wing)
At 07:51 AM 9/13/2006, you wrote:
Quote: | Yes, there are still a number of failure modes that could create a
problem, even with two independent fuel valves. In the case
mentioned below, the low tank valve breaks and won't shut off. In
that case the (presumably fuller) tank could be turned on, feeding
fuel from both tanks and the plane landed at the earliest
convenience. The fuller tank would probably back-feed into the
lower tank, extending the range compared to just feeding off the
lowest tank, depending on the level of fuel in the fuller tank. In
any event one has to assume there is only the fuel left in the
lowest tank. If the valve for the "new" tank failed to open (broke
in the closed position) the response is simple - land before the low
tank goes empty. Not a perfect condition, but there are still
options available that might not be available with a single
valve. If a single valve broke in the "low" tank position fuel in
the fuller tank is completely unavailable. If it sticks and breaks
between tanks (what I think is a more likely failure) neither tank
is available and the engine quits right away. The normal operating
procedure with two valves is to open the valve for the new tank,
confirming that it actually moved, and then shutting off the valve
from the old tank. Never shut the old one off before turning the
new one on. Also, I know that there have been some postings
advocating, or at least acknowledging, running a tank dry as a
routine operating mode. I suppose if there were more than two tanks
this isn't so bad, but with only two tanks doing that gives up one
of the redundancies in the system - two independent sources of
fuel. I personally don't like the idea of running a tank dry. My
own standard is to never run either tank lower than what is required
to get to the nearest airport. FWIW.
Gary Casey
Lancair ES
On Sep 12, 2006, at 11:55 PM, AeroElectric-List Digest Server wrote:
>
>You are worried about fuel valve handles breaking. With your design,
>
>what happens if you have run one tank down low, want to switch to the
>
>other tank, and the valve handle on the tank that is feeding breaks
>
>so you can't close that valve? Won't the engine quit once that tank
>
>is empty and the engine starts sucking air? How is this better than
>
>a valve handle failure with the normal design?
>Kevin Horton RV-8 (finishing kit)
>
>Ottawa, Canada
|
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
oldbob(at)BeechOwners.com Guest
|
Posted: Wed Sep 13, 2006 6:09 am Post subject: Fuel Pump Switch(es) |
|
|
Good Morning Dave,
I recognize that this message is really a "Me Too"
message and that such messages are discouraged by the
list administrator, but I can't help myself!
You have mashed that nail very squarely.
I am convinced that any system failure is much likely
to be one induced by "yours truly" rather than by any
failure of the mechanism of my flying machine.
KISS
Do Not Archive.
Happy Skies,
Old Bob
AKA
Bob Siegfried
Stearman N3977A
Downers Grove Illinois
LL22
--- Dave N6030X <N6030X(at)DaveMorris.com> wrote:
Quote: |
<N6030X(at)DaveMorris.com>
I have a little different point of view that I'd
like to just throw out there.
In my experimental Dragonfly, I was gripped for a
long time by a
NASA-like desire to have multiple redundant systems
everywhere. I'm
a software guy, so I love complicated,
microprocessor-controlled
everythingies. But when I started working through
the cockpit
workload part of dealing with multiple independent
pumps and switches
and bypasses for everything and backup this and
that, I realized that
there was a single point of failure that was getting
more and more
overloaded with every new complexity: the pilot
(me).
Look at the NTSB reports, and you'll see there are
many more failures
of the human to throw the switch than of the switch
breaking. Fuel
Starvation due to forgetting how much fuel is in the
tank, not
knowing how much fuel is in the tank, forgetting to
switch tanks,
taking off without adequate fuel, etc. etc. I don't
remember seeing
too many "the fuel selector handle broke off between
tanks". The
oldest airplanes still flying anywhere (the
survivors of all our
human failures) have the simplest systems. Take a
look at how they did things.
My advice, which could be worth exactly as much as
you're paying for
it is this: build it light, build it high quality,
and build it
simple enough that you can intuitively grasp it
without schematic
diagrams and lengthy troubleshooting charts while
handling an
inflight emergency. If you have requirements such
as "don't ever
turn off this tank before turning on that tank", I
think you're
starting to set yourself up for failure. Think
Gravity Feed from a
header tank wherever possible. Instead of buying
two redundant
valves and plumbing, buy one high quality valve and
replace it every
2 years just for the helluvit. What's that gonna
cost you? A
hundred bucks? How much weight will you save, and
how much cockpit workload?
When you're doing your failure mode analysis,
include a list of all
of the things the pilot has an opportunity to do
wrong. We try to do
this in software development, and it's a useful
exercise before you
get carried away anticipating equipment failures
that are just never
going to occur. When designing software, we no
longer worry about
whether there will be a bit error on the memory
board, or whether the
disk drive will crash a track. But we spend a lot
of time worrying
about whether the user will forget to enter a value
or will enter the
wrong value. If the user (pilot) CAN do anything
wrong, he
WILL. Try applying that to your aircraft design and
see if you come
up with some interesting scenarios you had not
thought of before.
Dave Morris
1960 Mooney M20A
Manual gear retract
3 fuel tanks - single valve
No fuel system failures in 46 years (knock on wood
wing)
At 07:51 AM 9/13/2006, you wrote:
>Yes, there are still a number of failure modes that
could create a
>problem, even with two independent fuel valves. In
the case
>mentioned below, the low tank valve breaks and
won't shut off. In
>that case the (presumably fuller) tank could be
turned on, feeding
>fuel from both tanks and the plane landed at the
earliest
>convenience. The fuller tank would probably
back-feed into the
>lower tank, extending the range compared to just
feeding off the
>lowest tank, depending on the level of fuel in the
fuller tank. In
>any event one has to assume there is only the fuel
left in the
>lowest tank. If the valve for the "new" tank
failed to open (broke
>in the closed position) the response is simple -
land before the low
>tank goes empty. Not a perfect condition, but
there are still
>options available that might not be available with
a single
>valve. If a single valve broke in the "low" tank
position fuel in
>the fuller tank is completely unavailable. If it
sticks and breaks
>between tanks (what I think is a more likely
failure) neither tank
>is available and the engine quits right away. The
normal operating
>procedure with two valves is to open the valve for
the new tank,
>confirming that it actually moved, and then
shutting off the valve
>from the old tank. Never shut the old one off
before turning the
>new one on. Also, I know that there have been some
postings
>advocating, or at least acknowledging, running a
tank dry as a
>routine operating mode. I suppose if there were
more than two tanks
>this isn't so bad, but with only two tanks doing
that gives up one
>of the redundancies in the system - two independent
sources of
>fuel. I personally don't like the idea of running
a tank dry. My
>own standard is to never run either tank lower than
what is required
>to get to the nearest airport. FWIW.
>
>Gary Casey
>Lancair ES
>On Sep 12, 2006, at 11:55 PM, AeroElectric-List
Digest Server wrote:
>
>
>>
>>You are worried about fuel valve handles breaking.
With your design,
>>
>>what happens if you have run one tank down low,
want to switch to the
>>
>>other tank, and the valve handle on the tank that
is feeding breaks
>>
>>so you can't close that valve? Won't the engine
quit once that tank
>>
>>is empty and the engine starts sucking air? How
is this better than
>>
>>a valve handle failure with the normal design?
>>
>>
>>Kevin Horton RV-8 (finishing kit)
>>
>>Ottawa, Canada
>
>
browse
Subscriptions page,
FAQ,
http://www.matronics.com/Navigator?AeroElectric-List
Web Forums!
=== message truncated ===
|
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
frank.hinde(at)hp.com Guest
|
Posted: Wed Sep 13, 2006 6:31 am Post subject: Fuel Pump Switch(es) |
|
|
Exactly!
And now back to the one pump in each wingroot solution....left tank
..switch left pump on...Right tank switch right pump on... Both tanks
(TO and landing).. Both switches on. Pretty simple I think.
The other advantages are...
No selector valves to switch.
Highly resistant to Vapour lock...Biggest driver to use auto fuel
SIMPLE
No single point of failure anywhere...OK the fuel line, servo and
battery is (assuming you don't have an isolated twin battery setup)
Plug a fuel filter...Who cares?
Downsides
Uses more electrical power.
Have to design the "what happens if the alternator craps out" failure
mode...I have dual alternaors...8 amp back up is just big enough for a
radio, tansponder and one fuel pump
Extra 3 feet of pressurised fuel line in the cockpit.
Frank
RV7a 4 hours
--
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
khorton01(at)rogers.com Guest
|
Posted: Wed Sep 13, 2006 4:02 pm Post subject: Fuel Pump Switch(es) |
|
|
Fuel system problems are a major cause of engine stoppage in homebuilt aircraft. Any fuel system that deviates from the norm, in either design or installation details, opens the door to unexpected problems. You have made a bunch of assumptions on how the system will perform following various types of failures. It would be wise to validate those assumptions via actual tests, either on the ground, or in the air over a nice long runway.
Kevin Horton
On 13 Sep 2006, at 08:51, Gary Casey wrote:
[quote]Yes, there are still a number of failure modes that could create a problem, even with two independent fuel valves. In the case mentioned below, the low tank valve breaks and won't shut off. In that case the (presumably fuller) tank could be turned on, feeding fuel from both tanks and the plane landed at the earliest convenience. The fuller tank would probably back-feed into the lower tank, extending the range compared to just feeding off the lowest tank, depending on the level of fuel in the fuller tank. In any event one has to assume there is only the fuel left in the lowest tank. If the valve for the "new" tank failed to open (broke in the closed position) the response is simple - land before the low tank goes empty. Not a perfect condition, but there are still options available that might not be available with a single valve. If a single valve broke in the "low" tank position fuel in the fuller tank is completely unavailable. If it sticks and breaks between tanks (what I think is a more likely failure) neither tank is available and the engine quits right away. The normal operating procedure with two valves is to open the valve for the new tank, confirming that it actually moved, and then shutting off the valve from the old tank. Never shut the old one off before turning the new one on. Also, I know that there have been some postings advocating, or at least acknowledging, running a tank dry as a routine operating mode. I suppose if there were more than two tanks this isn't so bad, but with only two tanks doing that gives up one of the redundancies in the system - two independent sources of fuel. I personally don't like the idea of running a tank dry. My own standard is to never run either tank lower than what is required to get to the nearest airport. FWIW.
Gary Casey
Lancair ES
On Sep 12, 2006, at 11:55 PM, AeroElectric-List Digest Server wrote:
Quote: |
You are worried about fuel valve handles breaking. With your design,
what happens if you have run one tank down low, want to switch to the
other tank, and the valve handle on the tank that is feeding breaks
so you can't close that valve? Won't the engine quit once that tank
is empty and the engine starts sucking air? How is this better than
a valve handle failure with the normal design?
Kevin Horton RV-8 (finishing kit)
Ottawa, Canada
|
[b]
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
steveadams
Joined: 10 Jan 2006 Posts: 191
|
Posted: Thu Sep 14, 2006 4:32 am Post subject: Re: Fuel Pump Switch(es) |
|
|
The most common causes of fuel exhaustion when there is fuel on board is pilot error. Switching to the wrong tank or not fully turning a mechanical switch to a completely on position. Failure of a properly maintained mechanical fuel valve is extremely rare. Vapor lock in a well designed system running AV gas is pretty rare in flight. While having seperate wing root pumps on seperate electrical systems may seem like complete redundancy, in reality it only assures partial functionality of the fuel system by giving you 2 independant fuel systems with less capacity. Say you're tooling along enjoying a flight, and forget to switch tanks. I know it will never happen to us because we are such conscientious pilots, but let's just imagine it could happen. Maybe you run one tank dry, or you realize you forgot and switch tanks with minimum fuel in one tank. You switch on the other pump and nothing happens. You still have a perfectly good pump on one side, with redundant power supplies, and no risk of vapor lock, but that side has no fuel. Your redundancy has disappeared and all your additional fail safe modifications are useless. Adding header tanks, cross feeds etc only adds to the complexity and introduces additional failure modes. Keep it simple, stick to proven designs and methods, and keep it maintained properly. In my non-engineer opinion, you'll have a more reliable system.
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
frank.hinde(at)hp.com Guest
|
Posted: Thu Sep 14, 2006 6:18 am Post subject: Fuel Pump Switch(es) |
|
|
Ahh there in lies your assumption....I agree, with AVGAS VL incidences
are very rare...With Mogas they are alarmingly common and a mechanical
fuel pump North side of a hot firewall is in exactly the "hydraulically
incorrect" place.
Yes true, you have to be aware of how much fuel you leave in one tank,
that's the one limitation of the wing root system, but does any of us
really go below half an hour on a tank...Some folks do suck a tank dry
before switching with a standard system, but of course your asking for
it if you did that with my system.
And note there are no cross feeds or any other so called fail safes, its
just one pump for one tank, no cross feeds nothing. I agree with you,
adding complexity is simply adding failure modes. With a simple wingroot
system just don't go below half an hour on each side before switching
pumps.
Works for me
Frank
--
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
oldbob(at)BeechOwners.com Guest
|
Posted: Thu Sep 14, 2006 7:29 am Post subject: Fuel Pump Switch(es) |
|
|
Good Morning Frank,
You state: "And note there are no cross feeds or any
other so called fail safes, its just one pump for one
tank, no cross feeds nothing. I agree with you, adding
complexity is simply adding failure modes. With a
simple wingroot system just don't go below half an
hour on each side before switching pumps.
Works for me
Frank"
And I am sure it will work just fine for any other
pilot that never makes a mistake!
Unfortunately, I have found that I am not one of those
pilots. I make an awful lot of mistakes and I often
forget to switch tanks when I should.
I believe the point that Steve wanted to make is that
it is the human pilot who has the higher failure mode.
Happy Skies,
Old bob
--- "Hinde, Frank George (Corvallis)"
<frank.hinde(at)hp.com> wrote:
[quote]
Frank George (Corvallis)" <frank.hinde(at)hp.com>
Ahh there in lies your assumption....I agree, with
AVGAS VL incidences
are very rare...With Mogas they are alarmingly
common and a mechanical
fuel pump North side of a hot firewall is in exactly
the "hydraulically
incorrect" place.
Yes true, you have to be aware of how much fuel you
leave in one tank,
that's the one limitation of the wing root system,
but does any of us
really go below half an hour on a tank...Some folks
do suck a tank dry
before switching with a standard system, but of
course your asking for
it if you did that with my system.
And note there are no cross feeds or any other so
called fail safes, its
just one pump for one tank, no cross feeds nothing.
I agree with you,
adding complexity is simply adding failure modes.
With a simple wingroot
system just don't go below half an hour on each side
before switching
pumps.
Works for me
Frank
--
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
frank.hinde(at)hp.com Guest
|
Posted: Thu Sep 14, 2006 8:10 am Post subject: Fuel Pump Switch(es) |
|
|
Certainly can't argue that one...I made a mistake myself once...
What I think I was trying to get at though was the 'pump failing just as
you have inadvertantly run out of gas from the other tank'...Is a double
failure...I.e it assumes the pilot forgot to switch tanks when he
should...AND....the pump on the tank with fuel in it has just failed.
That is a VERY unlikely situation...I mean lets say there is a 1 in 1000
chance the dumb*ss pilot (me) will suck a tank dry. Assume also there is
a 1 in 1000 chance a pump will fail on any particular flight.
The risk of both failures occurring concurrently is 1000*1000= 1 in 1
million.....Of course these numbers are completely arbitrary but the
point is clear, double failures are so rare that you almost never plan
for them in both aeronautical or industrial systems.
Of course if the risk so high (like 400 people on a DC10) then there
might be 3 independent hydraulic systems...But for GA its never planned
for as far as I can think of.
So your point about will work fine for any other pilot that never makes
a mistake is not true, if you make a mistake you will not drop out of
the sky...So what our hero sucks a tank dry...then simply switch the
other pump on...I very much doubt you would even have to turn off the
"air pump" to get the engine to run quite happily.
To counter the standard airplane system...Suppose you suck a tank
dry...Are we any more certain that the boost pump will re-prime, i.e
suck the air out of the line before it continues to pump fuel? Or the
handle doesn't suddenly snap off?...It is highly likely it will work as
planned (always has done) but I doubt you could prove it is any more
reliable than the wing root pump system.
Anyway, it really comes down to personal preference. I have run this
system for 410 hours and I like it. It is also done for a specific
purpose, i.e to reduce the risk of vapour lock with MOGAS...Otherwise I
would have just gone with a standard system.
I am not so invested in this solution that I am touting it as the only
way to go. It has its pros and cons, its just that for the specific
purpose (Mogas) its about the best (as an engineer who sometimes designs
pumped systems) I can come up with.
But assuming all is lost because the pilot makes a mistake is a false
assumption.
Young Frank....
--
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
bakerocb
Joined: 15 Jan 2006 Posts: 727 Location: FAIRFAX VA
|
Posted: Fri Sep 15, 2006 5:13 am Post subject: Fuel Pump Switch(es) |
|
|
Responding to an AeroElectric-List message previously posted by: "Hinde,
Frank George (Corvallis)" <frank.hinde(at)hp.com>
Frank wrote: "......skip......Of course if the risk so high (like 400 people
on a DC10) then there
might be 3 independent hydraulic systems.....skip......"
9/15/2006
Hello Frank, This response is not intended to be a criticism of, or
substantive contribution to the switch issue that you were addressing, but
merely a comment on the perfidy or irony that can lurk in aviation designs,
incidents, or accidents.
Incident One: The famous DC-10 Souix City crash of United Flight 232 flown
by Capt. Al Haynes when a disintegrating engine disabled all three hydraulic
systems.
Incident Two: The double, and almost triple, engine failure incident on a
three engined airline transport airplane when a mechanic replaced all three
engine oil sump plugs without first installing an O ring on each plug. The
plane eventually successfully landed with only one engine operating.
OC -- The best investment we will ever make is in gathering knowledge.
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
oldbob(at)BeechOwners.com Guest
|
Posted: Fri Sep 15, 2006 5:40 am Post subject: Fuel Pump Switch(es) |
|
|
Good Morning OC,
And don't forget the DC-10 out of Miami that
encountered a hydraulic problem shortly after takeoff.
This incident happened within the first couple of
months after the DC-10 was placed in service.
The Captain was one of the sharpest tacks in the box,
took immediate action, and got the airplane back on
the ground rapidly.
It was found that the back up hydraulic systems had
all failed due to vibration of the hydraulic lines.
They were "hard" lines. All such lines were
immediately replaced with flexible lines.
Had he had it in the air another couple of minutes,
all control would have been lost.
A major save by an exceptional aviator.
I am sure the Douglas designers had checked the
operation of their design thoroughly on the ground and
in mockups. Unfortunately, the conditions encountered
in the actual failure never showed up in any of the
design flight testing.
There is considerable advantage to sticking with the
tried and true. When departing from that norm, it pays
to test every back up system with a true time failure.
Sometimes, it is extremely difficult to simulate the
actual conditions that will occur following a failure.
Happy Skies,
Old Bob
Stearman N3977A
Downers Grove, IL
LL22
--- bakerocb(at)cox.net wrote:
Quote: |
<bakerocb(at)cox.net>
Responding to an AeroElectric-List message
previously posted by: "Hinde,
Frank George (Corvallis)" <frank.hinde(at)hp.com>
Frank wrote: "......skip......Of course if the risk
so high (like 400 people
on a DC10) then there
might be 3 independent hydraulic
systems.....skip......"
9/15/2006
Hello Frank, This response is not intended to be a
criticism of, or
substantive contribution to the switch issue that
you were addressing, but
merely a comment on the perfidy or irony that can
lurk in aviation designs,
incidents, or accidents.
Incident One: The famous DC-10 Souix City crash of
United Flight 232 flown
by Capt. Al Haynes when a disintegrating engine
disabled all three hydraulic
systems.
Incident Two: The double, and almost triple, engine
failure incident on a
three engined airline transport airplane when a
mechanic replaced all three
engine oil sump plugs without first installing an O
ring on each plug. The
plane eventually successfully landed with only one
engine operating.
OC -- The best investment we will ever make is in
gathering knowledge.
|
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
glcasey(at)adelphia.net Guest
|
Posted: Fri Sep 15, 2006 6:19 am Post subject: Fuel Pump Switch(es) |
|
|
Kevin previously wrote:
"If you don't do the test then, someday you may
stumble across that condition when you hadn't planned it. If the
system works as expected, then everything is OK. But if the system
does not perform, then you may lose the aircraft.
"For example, what is the engine restart procedure after you have shut
one tank down, and were slow opening the other fuel valve, resulting
in engine stoppage? When would you rather try out your planned
restart procedure for the first time? Overhead a long runway at
5,000 ft when you were mentally prepared for problems? Or over rough
country with no fields or runways in sight when a stoppage would be a
complete surprise?
Good points, but what "assumptions" do you have in mind that should
be tested? The one you mention about restarting the engine was
confirmed twice - once on each tank, with predictable results. The
engine started just as it normally does after fuel flow resumes. I
also verified on the ground that if a high vacuum is pulled on the
system nothing leaks or collapses in a way that would block fuel
flow when the fuel was turned back on. Are there others you had in
mind? A similar question might go to those who use conventional fuel
selectors - has anyone checked the width of the "dead band" between
tanks to see what happened if the valve isn't turned all the way? I
doubt that one in a hundred people would check that during the flight
test phase - I don't think I would have. Finally, there was a
discussion about using dual wing root pumps in a plane using an EFI
system. I was going that way for a while and after analyzing the
alternatives came to the same conclusion. And what are the odds that
one will inadvertently run a tank dry and then discover the other
pump doesn't work? The pilot is probably not that reliable and he
might do it one in a thousand or even one in a hundred (I'll admit
doing it once in 1500 hours of flying) times. However, what are the
odds that the pump won't run after having previously run DURING THE
SAME FLIGHT? I suggest it is more like one in 100,000. However,
there is a problem with using automotive roller vane or Gerotor pumps
- they are not very tolerant of contamination. I think a nylon inlet
"sock" just as used in automotive applications is necessary to
protect the pump. The coarse screens typically used in aircraft
aren't adequate for that purpose.
Gary Casey
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
frank.hinde(at)hp.com Guest
|
Posted: Fri Sep 15, 2006 6:25 am Post subject: Fuel Pump Switch(es) |
|
|
Sounds like the single point of failure was the mechanic...Hmm guess
there is way round that one...
Actually the United 232 incident was the one I was thinking of when I
wrote that, in a way the complex 3 hydraulic systems still had a single
point of failure...i.e when the #2 engine blew up!
Mind you in a single engine plane I guee we are preapred for at least
some risk right...
Frank
Responding to an AeroElectric-List message previously posted by: "Hinde,
Frank George (Corvallis)" <frank.hinde(at)hp.com>
Frank wrote: "......skip......Of course if the risk so high (like 400
people on a DC10) then there might be 3 independent hydraulic
systems.....skip......"
9/15/2006
Hello Frank, This response is not intended to be a criticism of, or
substantive contribution to the switch issue that you were addressing,
but merely a comment on the perfidy or irony that can lurk in aviation
designs, incidents, or accidents.
Incident One: The famous DC-10 Souix City crash of United Flight 232
flown by Capt. Al Haynes when a disintegrating engine disabled all three
hydraulic systems.
Incident Two: The double, and almost triple, engine failure incident on
a three engined airline transport airplane when a mechanic replaced all
three engine oil sump plugs without first installing an O ring on each
plug. The plane eventually successfully landed with only one engine
operating.
OC -- The best investment we will ever make is in gathering knowledge.
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
bicyclop(at)pacbell.net Guest
|
Posted: Fri Sep 15, 2006 8:47 am Post subject: Fuel Pump Switch(es) |
|
|
Gary,
As I understand it, the coarse screens at the tank pickups are meant to
stop the big chunks without stopping up with an accumulation of small
stuff. I've always thought the reasoning was that the fuel screens in
the tank get checked rarely if ever, but the finer gascolator and finer
yet carb screens are to be inspected regularly. If finding accumulations
there, it's a real good reason to make the extra effort to check the
tanks.
If there were fine screens in the tank, they could stop up without the
warning afforded by seeing small accretions at more easily inspected
locations. You probably didn't slosh your tanks and that's a good thing,
but the possibility of contamination being introduced into your tanks is
not zero and unless you were to inspect those nylon socks regularly, you
probably wouldn't know until they plugged up.
Pax,
Ed Holyoke
<glcasey(at)adelphia.net>
Quote: | However,
there is a problem with using automotive roller vane or Gerotor pumps
|
- they are not very tolerant of contamination. I think a nylon inlet
"sock" just as used in automotive applications is necessary to
protect the pump. The coarse screens typically used in aircraft
aren't adequate for that purpose.
Gary Casey
| - The Matronics AeroElectric-List Email Forum - | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?AeroElectric-List |
|
|
|
Back to top |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|