Fuel System Questions

[marcusofcotton]

When you most need full power, you'll be climbing. Install the fore-aft tees in such a way that the fuel flows straight through from the rear pickup. There are some pics out there of same, maybe even mention of it in the newsletter.

[whee]

When you most need full power, you'll be climbing. Install the fore-aft tees in such a way that the fuel flows straight through from the rear pickup. There are some pics out there of same, maybe even mention of it in the newsletter.

I had a discussion with Zman regarding tee orientation. I thought the same as you but couldn't proved it using any theory and couldn't reference any reputable articles that came to the same conclusion. I set out to the hardware store for some parts to prove my point. I set up a backwoods lab and proved to my satisfaction that I was wrong. Tee orientation doesn't make any significant difference.

[Mark Goldberg]

Whee - I think you are making a mistake by not using a fuel valve with a BOTH setting. Past experience has proved this is best. Mark

[marcusofcotton]

Build it like you want it. By chance, as I was reading through old Beartracks last night, I came across a Bob Barrows Engineering notice in the fourth quarter of the 2007 newsletter illustrating what we were thinking.

ASHRAE 2013 Handbook Fundamentals page 3.8, table 3, shows K factor fitting loss for straight through tee flow of .5 verses 1.8 for flow through the branch.

Doesn't seem like much difference in work or weight either way.

FWIW

[Zzz]

Curious how many here have experience with high pressure fuel systems that return fuel to the tanks in a constant recirculation? That's what this is. A BOTH setting on a selector valve can't be used because it won't return fuel evenly to the tank it was fed from, and a tank can overfill. The only way around this is a header tank, but that has its problems too.

[patrickh99]

I think this was mentioned previously: Wye fittings have a much lower flow restriction than Tee fittings. Minimize the number of 90 degree fittings as much as possible. Two 45s have better flow than one 90.

[Bcone1381]

Mark, I highly respect your input and I am curious what was experienced that precipitates the warning.

Brooks Cone

[marcusofcotton]

Darn, I don't seem to be able to get the quote function to work with this tablet. I've had to redesign steam and ductwork systems that were not performing properly because they weren't engineered correctly. I also have heard of too many people that have had some inadequate fuel flow troubles in aircraft.

[Battson]

Mark, I highly respect your input and I am curious what was experienced that precipitates the warning.

Brooks Cone

Many planes have crashed due to fuel starvation, while there was still fuel left in the tanks.
I am sure you agree that pilots as a collective are fallible and will continue to run tanks dry if the selector isn't on BOTH habitually. I know several people who've fallen victim to this mistake, fortunately all landed safely or realised their error during the glide.
I guess that is what Mark was referring to.

[Battson]

Curious how many here have experience with high pressure fuel systems that return fuel to the tanks in a constant recirculation? That's what this is. A BOTH setting on a selector valve can't be used because it won't return fuel evenly to the tank it was fed from, and a tank can overfill. The only way around this is a header tank, but that has its problems too.

If fuel is being returned into a tank, one thing I have noticed is the sight gauge goes haywire while the fuel is flowing in. It caused the gauge to drastically under-read the fuel level, so the tank can be quite full but look quite empty on the gauge. This plays right into what Z was saying above about the tank over-filling, fuel could accidentally be dumped overboard. I can't say the reason why the gauge does this, but I guess it's to do with insufficient venting in the tank and air pressure increasing. I experienced this in a Bearhawk with the tank vented only with the Bearhawk fuel cap.

[Zzz]

If fuel is being returned into a tank, one thing I have noticed is the sight gauge goes haywire while the fuel is flowing in. It caused the gauge to drastically under-read the fuel level, so the tank can be quite full but look quite empty on the gauge. This plays right into what Z was saying above about the tank over-filling, fuel could accidentally be dumped overboard. I can't say the reason why the gauge does this, but I guess it's to do with insufficient venting in the tank and air pressure increasing. I experienced this in a Bearhawk with the tank vented only with the Bearhawk fuel cap.

Hmm. Theoretically, if the fuel is being recirculated indefinitely, the only net volume reduction is what is burned in combustion. The tank fuel volume should steadily lower at the same rate as any other engine fuel delivery system.

"While the fuel is flowing in" would be constantly. What type of FI was this happening with? I agree good tank venting is paramount to avoid underpressure but this seems an odd scenario. I wonder if the system was picking up air in the return line.

Not to go too far off topic, but if you plumb vents besides the caps, what sort of external fixture is generally used? Are they meant to utilize ram air?

[whee]

Whee - I think you are making a mistake by not using a fuel valve with a BOTH setting. Past experience has proved this is best. Mark

I appreciate you sharing your thoughts Mark. As Z already stated, "Both" isn't an option with this type of FI without a header tank. Not having "Both" doesn't bother me though, most of my time is in an airplane that didn't have it.

Build it like you want it. By chance, as I was reading through old Beartracks last night, I came across a Bob Barrows Engineering notice in the fourth quarter of the 2007 newsletter illustrating what we were thinking.

ASHRAE 2013 Handbook Fundamentals page 3.8, table 3, shows K factor fitting loss for straight through tee flow of .5 verses 1.8 for flow through the branch.

Doesn't seem like much difference in work or weight either way.

FWIW

Thanks for doing that leg work and posting it.

I want to build it the right way the first time. I'm aware of Bob's engineering notice and I am aware of the difference in K factor which is why I started this conversation with Z. At the flows we will be seeing I don't think it makes any difference and since there are a few Bearhawks flying with the tee's oriented as I have depicted it obviously works. However, I think Bob's change is the best practice though not as clean aesthetically.

If fuel is being returned into a tank, one thing I have noticed is the sight gauge goes haywire while the fuel is flowing in. It caused the gauge to drastically under-read the fuel level, so the tank can be quite full but look quite empty on the gauge. This plays right into what Z was saying above about the tank over-filling, fuel could accidentally be dumped overboard. I can't say the reason why the gauge does this, but I guess it's to do with insufficient venting in the tank and air pressure increasing. I experienced this in a Bearhawk with the tank vented only with the Bearhawk fuel cap.

I'm very curious about this. What port what the return line connected to?

As Z pointed out, this is a high flow fuel injection system and I'm guessing most people here don't have experience with a system like that. According to the data I have found the system will be drawing 25gph from the tanks minimum at all times. That's a lot of fuel, more than most of the carbureted systems typically see so what is best for a carb system may not be the best for this system.

[Mark Goldberg]

I also had mostly flown airplanes with a position for each tank. One Bellanca Viking had 4 different tanks with a position for each. But I believe that a BOTH position selector valve is the safest in the BH. However - it is your plane and you can build it however you wish. I am not trying to convince you - but want to state clearly what Bob & I think is the best/safest way to go. Mark

[whee]

Thanks Mark. I'm not trying to convince anyone of anything either. I'm just looking for factual information about how things have been done in experimental and certified airplanes. It's impossible for me to find fuel system schematics on the internet for every high wing fuel injected airplane out there but I have sure tried. I'm sure your aware that the FAA recommends a cross vent between tanks for gravity feed systems with a "Both" position so that the tanks drain more evenly. It is possible to experience fuel starvation when running on "Both" with fuel still in one tank and no cross vent. I've experienced it...twice; though I'd guess there was something not quite right with that fuel system. If I managed the fuel without using "Both" in that airplane I didn't have any trouble.

[JimParker256]

As far as "safer" to use "Both" position, I can see both sides of that argument. We've all heard stories about pilots who had selected a tank with low fuel, and during maneuvering to land managed to un-port that tank, causing the engine to quit on short final... Not good. The "Both" position would have prevented that situation.

But on the other hand, being able to select L or R can also be a huge advantage. When you depart on a lengthy cross-country flight where you will need to draw fuel from both tanks, it's a good idea to be able to verify that each tank's fuel can, in fact, be used by the engine. The only way to confirm that for sure is to be able to select each tank and run off that tank for sufficient time to know the vent system is working, fuel is flowing properly, etc.

A quick story to illustrate the point: I flew a Grumman Tiger on a cross-country flight during my Commercial training. That plane has only L/R selector - no "Both" position. I started on the R tank, then switched to the L for the runup, takeoff, and first fuel stint. Everything worked fine, so I thought I was good to go... After flying for 30 minutes on the L tank, I switched to the R tank, observed the engine instruments for a couple of minutes as I always do after changing tanks, and everything seemed fine. About 10 minutes later, I heard a weird "CRUNK" sound and started looking around for the source. Nothing seemed wrong, until I looked at the right wing... There was a huge "dent" in the upper surface of the wing, right where the "wet wing" fuel tank would be... I quickly realized that the vent must be blocked on that tank, and the fuel being sucked out by the fuel pump was creating low pressure inside the tank, so it collapsed the upper wing skin... How much longer would my engine have run that way? I have no idea, and had no wish to figure it out by experiment. I switched back to the left tank, realized I could NOT complete the flight on just that one tank's worth of gas, and returned home to get it fixed.

Sure enough, the mechanic found the fuel vent plugged by a mud dauber. Once it was cleared, the tank returned to "normal" and everything was fine. After considerable discussion with my CFI, we realized that I had been saved from a serious issue by the fact that the Tiger had no "Both" position. If I'd been flying in both, I would have drawn all the fuel out of the left tank before ever discovering that the right vent was blocked. As it happens, I would have been directly over the Rockies at that point, enroute from Monterrey, CA to Las Vegas, NE.

After thinking about it, I decided that from then on, isolating each available fuel source early in the flight would be a good way to prevent this situation from ever arising again. Nowadays, in an airplane with L/Both/R switch (as I'm using on my Patrol build), I'll start and run-up on BOTH, and only switch to L or R once I'm well clear of the ground, and would have time to switch back to BOTH if the engine quit... Then, during the first portion of the flight, I will fly for at least 10 minutes on L and an equal amount of time on R to make sure I'm drawing fuel cleanly from each tank. After that, I generally switch tanks every XX gallons (if I have a fuel flow meter) or every XX minutes (if I don't have a FF meter). Not only does that force me to keep track of how much fuel is in each tank throughout the flight, it also helps to ensure that the plane isn't flying with one wing low because I've got 30 gallons more fuel in it than in the opposite tank...

By the way, during the pre-landing checklist, I prefer to switch back to "Both" tanks, so that maneuvering is less likely to unport a tank...

So, my preference is to have a 3-position switch, and use all three positions as appropriate. (But I do understand whee's point about injected systems with a return line not being able to support a BOTH position without a header tank. It is one of the several reasons I am going to stick with carburetion... I like "simple"...)

[Bcone1381]

Beartracks Q4 2008

Bearhawk Operation Notice - Fuel System

Takeoff and landings should be made with fuel in both tanks and the fuel selector valve in the BOTH position. If using LEFT or RIGHT tank selector valve position, selected tank should beat least half full and caution should be used as to NOT have a significant slip condition that could unport the fuel tank selected."