Fuel Starvation With A Full Tank

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
10,469
Location
97FL, Florida, USA
Why are airplanes designed to allow a pilot to starve his engine when he has another tank that is full???

My design has two wing tanks thet feed into a common sump. No selector valve.
I’ll bet that even your design has a fuel shutoff valve.


BJC
 

narfi

Well-Known Member
Joined
Feb 23, 2016
Messages
795
Location
Alaska
Easier control of how much fuel you have. (weight)
Controlling CG. (either a heavy wing in case of wing tanks, or fore/aft in case of multiple belly tanks)
Preventing cross feeding into a downhill tank when parked or with a flat tire.
Saving enough fuel to land safely in case one tank springs a leak.

Like most things, there are compromises which you have to choose for your own design.
However, I would be cautious to think your own way is superior to all others if you haven't tried to understand them and their advantages as well as drawbacks.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
12,234
Location
Memphis, TN
Low wing can cross drain to low wing if bank too long. If you have some one way valves you trust, you can try that. Just hope one does not lock out. Also many low wing planes will use the tanks to trim roll. One tank gets light, feed on the other as it trims the other way; then back again till destination. Lots of ways to skin the cat. Pilots doing their job should be just as important as design.
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
12,800
Location
Port Townsend WA
C-150 cross feeds on a side slope. Just live with it.
I would go with no fuel valves at all, if possible. My car has none.
 

Marc Zeitlin

Well-Known Member
Joined
Dec 11, 2015
Messages
609
Location
Tehachapi, CA
Why are airplanes designed to allow a pilot to starve his engine when he has another tank that is full???
Depends on the plane. On high wing planes, where there's substantial head pressure between the tanks and the engine fuel inlet a "both" setting is reasonable - even with slightly different fuel levels in the tanks, the head pressure will ensure the fuel always flows to the engine.

But on low wing planes, where the fuel tank can easily be below the engine fuel inlet, if you have a "both" setting and due to slips or variations in vent pressure into the tanks, one tank runs dry, you then lose the ability to "suck" fuel from the tank that still has fuel, as air is easier to suck than fuel. This has happened to more than one EAB airplane with low or mid-tanks when the builder thinks they know better than the designer (or the designer doesn't know what they're doing).

So it all depends on the configuration of the aircraft, with respect to engine/tank position and vent design. Some airplanes can get away without a "L" - "R" valve - others can't. Fuel systems can get very complicated...
 

Rockiedog2

Well-Known Member
Joined
Dec 11, 2012
Messages
2,318
The Zenith 701 plans fuel system has 2 wing tanks feeding into a common sump like yours. ON OFF valve only. I built mine IAW the plans and it was supposed to be gravity feed but mine wouldn't reliably feed a 912S at full throttle with the nose up in a steep climb. So I assumed it wasn't gravity feed period and installed a backup electric pump. It had most of the problems mentioned by TFF and narfi. The tanks wouldn't feed evenly and since it wasn't reliable gravity flow whenever one of the tanks got low I had to land for fuel with maybe an hour or more unusable fuel on board. And yeah all the crossflow and trim problems etc. So I designed an OFF L R system that would gravity flow and kept the engine driven pump as well as the backup electric pump. End of problems. Total control. Van's has L R OFF as well...he's got it figured out. The Zenith system could have easily been designed as a good fuel system...the only apparent reason it wasn't was gross oversimplification.
I liked TFF said the pilot doing his job is as important as system design. Personally, I would build a system that I could control instead on one that does what it wants to.
 

GESchwarz

Well-Known Member
Joined
Oct 23, 2007
Messages
1,179
Location
Ventura County, California, USofA.
But on low wing planes, where the fuel tank can easily be below the engine fuel inlet, if you have a "both" setting and due to slips or variations in vent pressure into the tanks, one tank runs dry, you then lose the ability to "suck" fuel from the tank that still has fuel, as air is easier to suck than fuel.
My design is a low wing. Both wing tanks gravity feed through their own filter and check valve to the centerline sump tank, which is lower than the two main wing tanks. All three tanks are vented. An electric boost pump draws fuel from the sump and sends the fuel to the mechanical pump. This is how the T-34 Mentor works. The check valve is a simple free-swinging flap door that allows fuel to flow into the sump tank, but not out of the tank. So it is impossible for the sump to run dry before both tanks run dry. Each wing tank is a wet-wing type, between the spars, occupying 4 rib bays. The inboard most internal tank rib features the same sort of free-swinging gate check valve as there is in the sump tank. This ensures that the inboard most bay always has a high fill level. It is from this bay that the fuel exits to the common, centerline sump tank. The sump tank will always be full.

Does anyone see an issue with my design?
 

Marc Zeitlin

Well-Known Member
Joined
Dec 11, 2015
Messages
609
Location
Tehachapi, CA
Does anyone see an issue with my design?
Sounds basically OK, but you might want to evaluate what happens with the flapper valves, depending upon how they work, when in a prolonged slip. The other thing I might caution, although the design SEEMS immune to this issue, is that differential vent pressures can do weird things, and possibly override the pressures that the flappers are meant to deal with. I'd recommend manifolding the vents together so that all three tanks see exactly the same vent pressure, whatever that might be. This way, the tanks won't be TRYING to be at different levels, even if you're protecting against it, which it sounds like you're trying to do.

Try to think about ALL possible aircraft orientations - nose high, nose low, slips to either sidewhile nose high and nose low, full tanks on one side and empty on the other (and everything in between), etc. Where are the tanks in relation to the engine, where's the boost pump, etc., etc. If you've done all that and can find no situation in which the center sump (you don't say how large) can EVER run dry when there's fuel in either wing tank, even with prolonged non-coordinated flight, then you should be good to go.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,372
Location
North Carolina
One way valves all need some pressure. How accurately can you match yours? Any slight imbalance and you are sucking air. Inertia effects can also do this and a few inches of fuel is very little pressure. My point is that it's very easy to get an imbalance and rather tricky to avoid one. A slight slip will also mess with the system.
If one tank has more fuel, that side of the plane is going to want to be lower. With joined tanks you'll have a very wide and thin tank that will need to be very level to be drained dry.
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
8,958
Location
CT, USA
My design is a low wing. Both wing tanks gravity feed through their own filter and check valve to the centerline sump tank, which is lower than the two main wing tanks. All three tanks are vented. An electric boost pump draws fuel from the sump and sends the fuel to the mechanical pump. This is how the T-34 Mentor works. The check valve is a simple free-swinging flap door that allows fuel to flow into the sump tank, but not out of the tank. So it is impossible for the sump to run dry before both tanks run dry. Each wing tank is a wet-wing type, between the spars, occupying 4 rib bays. The inboard most internal tank rib features the same sort of free-swinging gate check valve as there is in the sump tank. This ensures that the inboard most bay always has a high fill level. It is from this bay that the fuel exits to the common, centerline sump tank. The sump tank will always be full.

Does anyone see an issue with my design?
If you're gravity feeding the sump tank from the wing tanks and all three tanks are vented, what's to stop the sump tank from overflowing?
 

Toobuilder

Well-Known Member
Log Member
Joined
Jan 19, 2010
Messages
4,600
Location
Mojave, Ca
...Does anyone see an issue with my design?
Yes, a few things.

The main tanks will have to be a considerable height above the sump tank to allow enough head pressure to make this a viable gravity feed system. As a low wing, I doubt you have this.

A check valve will need some spring pressure to allow a seal. A free swinging door will allow leakage over time, which will defeats your purpose. A flapper door works in a highly dynamic environment like the oil sump on a road race car where the sump cavity is replenished every few seconds, but is not a long term level control solution like an airplane parked on a sloping ramp or uncoordinated flight.

Do not underestimate the difficulty in achieving equal vent pressure. Even high wing Cessnas with high head pressure working for them will empty tanks at different rates when "Both" is selected.

Your system might be the magic one that actually works, but I'd still install a "L BOTH R OFF" selector just in case it doesnt.
 

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
3,042
Location
Thunder Bay
Thanks for that explanation Toobuilder, I had always wondered why very few light low-wings have a “BOTH” selection while nearly all high wings do.
 

Lucrum

Well-Known Member
Log Member
Joined
Jun 10, 2008
Messages
956
Location
Canton, GA
Why are airplanes designed to allow a pilot to starve his engine when he has another tank that is full???

My design has two wing tanks thet feed into a common sump. No selector valve.
I'm going with a single fuselage tank. But it's for simplicity more than being worried about forgetting to switch tanks.
Maybe a better question might be, why are airplanes designed to allow a pilot to crash them when nothing is mechanically wrong with the plane.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,372
Location
North Carolina
I'm going for twin tanks. One per engine. One tank slightly bigger than the other so that fuel exhaustion is staggered a bit. Engine vacuum operated fuel shutoff valves are a maybe.
 

GESchwarz

Well-Known Member
Joined
Oct 23, 2007
Messages
1,179
Location
Ventura County, California, USofA.
These one-way flapper valves are a simple flat plate of aluminum on a piano hinge. When closed, it lays across a port on a flat surface. In the main tanks, that flat surface is the web of the wing rib that doubles a tank baffle. In the sump, it is a plate of aluminum with a 1/4” hole in which the AN inlet hose fitting mates.

This valve opens easily and will allow flow by gravity at even the slightest head pressure. Certainly, they will seep fuel in the opposite direction, but the difference in flow between the two directions is somewhere around 100x, so the concern that these valve can allow significant backflow is unfounded.

As I mentioned already, the main tanks are baffled by three internal wing ribs. The inboard-most rib features one of these check valves. That ensures that than inboard-most bay is always nearly full. The port of that valve is 1” in diameter. Any slow rocking of the wings would quickly top off that inboard bay. That bay is where the port is that feeds the sump.

I cannot imagine a scenario in which the sump would ever not be full.

I will certainly tie all three vents together. Overflow from the vent is accomplished the way Vans does it, with a high loop to just under the canopy rail, then out the bottom of the fuselage.

The main tanks are self-leveling because the pair of valves in the sump naturally allow greater flow from the side with the higher head pressure, regardless of how slight the difference in pressure.

I greatly appreciate your generous responses.
 

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
10,469
Location
97FL, Florida, USA
These one-way flapper valves are a simple flat plate of aluminum on a piano hinge. When closed, it lays across a port on a flat surface. In the main tanks, that flat surface is the web of the wing rib that doubles a tank baffle. In the sump, it is a plate of aluminum with a 1/4” hole in which the AN inlet hose fitting mates.

This valve opens easily and will allow flow by gravity at even the slightest head pressure. Certainly, they will seep fuel in the opposite direction, but the difference in flow between the two directions is somewhere around 100x, so the concern that these valve can allow significant backflow is unfounded.

As I mentioned already, the main tanks are baffled by three internal wing ribs. The inboard-most rib features one of these check valves. That ensures that than inboard-most bay is always nearly full. The port of that valve is 1” in diameter. Any slow rocking of the wings would quickly top off that inboard bay. That bay is where the port is that feeds the sump.

I cannot imagine a scenario in which the sump would ever not be full.

I will certainly tie all three vents together. Overflow from the vent is accomplished the way Vans does it, with a high loop to just under the canopy rail, then out the bottom of the fuselage.

The main tanks are self-leveling because the pair of valves in the sump naturally allow greater flow from the side with the higher head pressure, regardless of how slight the difference in pressure.
Good luck with all of that.


BJC
 

Turd Ferguson

Well-Known Member
Joined
Mar 13, 2008
Messages
4,946
Location
Upper midwest in a house
The main tanks are self-leveling because the pair of valves in the sump naturally allow greater flow from the side with the higher head pressure, regardless of how slight the difference in pressure.
I'll post a schematic if I remember when I get home but a Cessna Caravan has a similar system. 2 wing tanks flow to a common tank called a reservoir tank (similar flapper valves and all) where it is then fed to the engine. Fuel flow is augmented from the reservoir tank to the engine because of demand but that part is irrelevant. If the plane is parked on uneven ground, fuel will flow to the low wing (through the reservoir tank). Fortunately, Cessna installed an On/Off valve for each wing tank so that can be prevented if recognized ahead of time. If the plane is flown uncoordinated, the wing tanks will become imbalanced. Too much imbalance and you run into an autopilot limitation. Every pilot I have flown a Caravan with blames the airplane for fuel imbalance in flight, lol. So they use the fuel On/Off selectors for a purpose which they were not designed, restrict flow to the reservoir tank to balance the fuel. That could lead to problems in a critical fuel situation which brings us back to the question of why airplanes are designed so the the pilot can run out of fuel with fuel in another tank?
 
2
Group Builder
Top