Fuel Flow Discussion, Moved from Float Mounting

[Battson]

What is this risk you speak of? Are you talking about having less fuel flow?

I am talking about the risk of running out of fuel in the tank you're using.

The reason the "BOTH" position was introduced, was pilots falling out of the sky with fuel left in the other tank.

Some of the most experienced pilots I know have run out of fuel in one tank, and forgotten to check fuel quantities during the limited time they had to respond. Then been forced into landing somewhere undesirable. Some ran out of fuel on short finals and fell short of the runway, just seconds to take action... Most modern POH address this by requiring the use of either fullest tank or both tanks prior to any critical flight phase. But people are not good with checklists and procedures. The safest option is leaving the selector on both at all times when in the air.

People are innately forgetful and prone to tunnel vision. If the aircraft fuel system design sets the pilot up to fail, sets a trap so to speak, then it's only a matter of time.

[schu]

There is plenty of risk with not having a both position on your fuel selector.

What is this risk you speak of? Are you talking about having less fuel flow?

Seems to me you are either risking your tanks staying balanced and having similar head pressure on the both position or, if I'm picking up what you are laying down, less fuel flow by removing one tank.

If it were me, I'd rather risk less fuel flow on one tank. Why?

1. The system must be designed to have enough fuel flow on either tank. If you can't operate at full power with minimal fuel in one tank, then you probably need a placard to declare minimal fuel volumes and the requirement for both tanks on take off. In my mind that is a hack. If the fuel system can't flow enough fuel on one tank, then it needs to be fixed, not placarded.

2. We have heard from two different bearhawk builders that have had engine stoppage by using 'BOTH'.

3. Other builders have reported enough fuel using a single tank. (Whee obviously tested fuel flow with one tank given his lacking 'BOTH')

Here is a post:

https://bearhawkforums.com/forum/bearhawk-general-discussion-news/1326-precertification-fuel-flow-testing-results?p=1340#post1340

Results:
Brake Specific Fuel Consumption rate of .55 lbs/hr was used to determine BSFC for the 195 hp I0-375 engine installed in aircraft.
195 hp x .55 =107 lbs/hr
107/6(lbs/gal)=17.8 gal/hr max flow rate.
17.8 x 150% = 26.7 gal/hr
To measure in oz/min:
First we need gals/min= 26.7/60=.45 gal/min
Second we need ozs/min= .45x128 (oz in a gal)=57.6 oz/min at 150% flow rate.
Therefore: 26.7 gal/hr = 57.6 oz/min

Gravity feed test:
Right tank = 88 oz/min = 41.25 gal/hr
Left tank = 72 oz/min = 33.75 gal/hr
Both = 108 oz/min = 50.63 gal/hr

High pressure fuel pump test:
Both = 136 oz/min = 63.6 gal/hr

It shows that both tanks by themselves flow enough to feed an o-540 + 50%.

What I find interesting is the variation between tanks in this test. That leads me to believe that fuel line routing greatly affects fuel flow, which could account for the seemly large differences between airplanes and fuel flow numbers. This is strong evidence that using fleet statistics to determine if there is a problem is dangerous. Bearhawks are all on-off airplanes when compared to a cessna.

My $.02 on this subject is pretty simple. If people have noted engine stoppage when selecting 'BOTH', we see unbalanced fuel flow in many airplanes (which indicates different head pressures), and the FAA suggests cross venting to ensure that the top AND bottom of the tanks act as one, then I'll be putting in a cross vent. I've already taken the time to retro in a skylight, float fittings, 4-pt harnesses, floor anchors, extended baggage, and seaplane doors, what's another 2 hours to cross vent.

schu

[Bcone1381]

Hummm....intersting.... And if we changed angle of attack, would the pressure on top of the wing change and submit a change of pressure inside of the tanks, and then would the change in the two unbalanced tanks change at different rates which might cause an unbalance? The fuller tank would change quicker....the fuller tank's pressure would go down quicker.

[svyolo]

With Bendix FI or a carb, I wouldn't use a header tank either. With EFI, I wouldn't use them if I used 1/2" fuel lines. If I remember right Whee made the same choice with TCM FI.

[zkelley2]

I also see no benefit of a header tank.

[Battson]

I am not an expert, so I am happy to be corrected....

Personally I see little benefit in a header tank with a Bendix injection system or carb system in a Bearhawk. There is plenty of risk with not having a both position on your fuel selector. There is plenty of risk with having a tank of fuel near the exhaust or in the cabin.

[Nev]

I have decided to use left and right and fit a two-gallon header tank, I have 3/8 fuel lines and 1/2 back from the header tank to the mains. IO540 C4B5. electrical and mechanical fuelpumps.

It makes intuitive sense to select the fuller fuel tank when balancing fuel, which works nearly all the time. But when a system malfunctions things often stop making sense, and in this case it may be that the fullest tank has plenty of fuel, but it’s not accessible.

I wonder if a header tank will actually mask this issue rather than solve it, by providing an additional 2 gallons before the issue makes itself known.

For example, with BOTH selected, if one tank is failing to feed properly due to tank vent blockage, pressure differential, fuel line vapor lock etc, the other tank will continue feeding the header. If the selector is later switched to the fuller tank in an effort balance, the fuel feed problems from that tank will then become known shortly after changing tanks, or 2 gallons later with a header tank. It may also create a delay in getting fuel to the engine to restart, once the unobstructed tank is selected.

[Archer39J]

Oh and to add, an additional vent on each tank would help mitigate this possibility, but I'm not going to be installing one. I'll be going with a check valve to prevent fuel stoppage if the vented caps get plugged, ice being my concern, McMaster has some I'm looking into. The check valve won't open at the pressures we've been discussing.

[Gerhard Rieger]

I have decided to use left and right and fit a two-gallon header tank, I have 3/8 fuel lines and 1/2 back from the header tank to the mains. IO540 C4B5. electrical and mechanical fuelpumps.

[JimParker256]

Rollie, I think if you put the cross-over vent at the highest point of the tank, and as far outboard as you can get it, you could avoid the "slope-transfer" effect. The fuel would naturally gravitate toward the inboard side of the uphill tank, thus away from the uphill vent. On the other hand, using a vent on the inboard side would have the opposite effect.
The Commander 114 that I used to have was a fuel-transferring demon... If you didn't place the fuel valve in the OFF position, it would transfer fuel until the tanks found a "level" point. Unfortunately, the nice, squishy trailing-link landing gear (which made for fantastic landings) allowed the now-heavier wing to sag even further as the fuel transferred. If left unattended on a slope with the fuel valve anywhere but OFF, it would transfer almost the entire 35 gallons in the uphill wing to the lower wing, which would then run out through the overflow vent. Many new Commander owners came back to their birds to find their newly-topped-off aircraft sitting in a puddle of 35 gallons of AvGas. An expensive mistake you only make once!

[Archer39J]

It's looking to me like in a property constructed gravity-fed Bob system a cross vent is unnecessary because the worst-case pressure differential isn't enough to be an issue. For engine stoppage at least, you may think it's a problem it's possible to dump fuel like this. Assuming a 100kt ram air pressure of about 0.23 psi, that means at most the pressure differential between the two tanks would be on the order of just under 0.5psi.

The fuel pressure head of a minimum fuel tank is on the order of 1.2psi (aft tank outlet to selector valve). So while fuel will be pushed from the tank with positive pressure to the one with the vent issue (and you'll dump fuel overboard if your tanks are full) the air won't reach the selector valve, and this is worst case when your tanks are effectively empty. Now in systems that don't have that pressure head this can absolutely cause a problem, but for us it doesn't look like this is too much of an issue. Even less so with more fuel and slower speeds.

This is of course for a gravity-fed system, if your moving massive quantities of fuel and your pump is sucking then I wish you well with your analysis.

[Rollie]

I think I might be starting to grasp this concept. I really want to understand it because I routinely fly down to about 10 gallons on board. And I always select both tanks.

After reading everything here, this is where I'm at: At first I thought that if there is enough pressure diff for one tank to suck the fuel out of the other and basically override the fuel pump and gravity, wouldn't it be enough to stop the fuel flow out of that tank even if the selector valve was on that tank instead of both? Maybe not. With the vent partially clogged on say the left tank, as the fuel level decreases, the air pressure in the left tank will decrease, maybe not to the point that it would stop the fuel from draining out of the tank but when you add in the right tank with the vent clear and oriented forward, the airflow might be partially pressurizing that tank. So now we have a differential. Fluids take the path of least resistance. (the one thing I understand about fluid dynamics) Usually that path is down from our tanks and into the engine, but in a rare case of the one tank being pressurized enough to be pushing fuel out and the other tank not having any significant pressure, the path of least resistance could become going down from the tank with the properly functioning tank, through the selector valve and up into the other tank. Once the fuel is flowing that direction, it would probably keep going that way and bypass the engine.

That is what I got from this discussion. I might be way off, I'm actually looking for clarification to see if I got it, I'm definitely not trying to say this is what happened.

Now I am also curious if check valves in the fuel lines just upstream from the selector valve would have the same result as a cross over vent line. It would prevent fuel from transferring so it seems like it would only be able to go to the engine where it belongs. I am not planning to add a cross over vent line because I park on slopes frequently requiring me to select L or R tank to prevent fuel from cross flowing and overflowing the downhill tank. I'm pretty sure if I had a cross vent the fuel would transfer through that in that situation.

[Nev]

My understanding at this stage is that the cross tank venting is added for two reasons:

1. Provide alternative venting in the case of a tank vent blockage.
2. Aid dispersion in the case of a fuel line vapor lock.

If the tank vents are operational, it could perhaps be a vapor lock issue that is only presenting itself very occasionally in the right conditions.

If certain geographical areas are providing the right conditions of high ground temp, low temp at altitude, low pressure (individual fuel system intricacies) etc then it may only manifest itself in specific countries/states, and during certain times of the year.

Just another 2c worth in what is a very interesting discussion.

[Ed.Meyer]

I was thinking the same thing. You said it very succinctly.

One additional thought that came to me is this: if there is significant difference in air volume in the tanks, like when one is near empty and the other is not, and there is altitude change then the tank with more air will have to vent more, either air coming in or going out, depending on increased or decreased altitude. This could create some imbalance until venting is complete. I have no idea how much this would or could be.

[JJohnston]

One thing that hasn't been adequately explained is how, without a malfunctioning tank vent, such a pressure differential arises. Technically, if both tank vents are working, the tanks ARE cross-vented - via the atmosphere - even if that doesn't satisfy the reg.