Returnless FI?

[schu]

I'm not misrepresenting what is being discussed.

Svyolo writes:

We have the dual benefits of a gravity fed fuel pump inlet, and double the operating pressure of Bendix, after the pump. I can't think of a reason not to do it.

This is a comparison of the benefits of the bendix, with the "after the pump" modifier, with the conclusion that there is no reason not to, but that "after the pump stuff" makes them completely different systems.

At any rate, I think I've "helped" enough... back to building my own airplane.

[Archer39J]

Context matters. If you saw that as a statement the systems are identical so we don't need to worry then you're not paying attention. A common theme in this thread...

[Battson]

I don't see the high wing thing as a material advantage, yes it's better than suction but only the pressure difference is not material compared to the massive pressure your proposed recirc-SDS system needs to operate - I understand they need about 40 to 50 psi?

Some relevant facts:
1. Bendix FI runs continuously at just 3 or 4 psi.
2. Gravity feed in a Bearhawk will not run a Bendix, not even close, thus gravity head post losses is much less than 3 psi.

[Archer39J]

Yeah it’s only material on the pump inlet side with the concern of “heated” recirc fuel being subjected to negative relative pressures. ”Heated” because with proper layout heat from the engine isn’t allowed back, hence the focus on how much the pump will heat fuel itself, which does need to be quantified. SDS specs 45-50 psi on an adjustable regulator. Minimum static head pressure in a 4-Place was like 1.15 psi back of the napkin.

[svyolo]

I don't see the high wing thing as a material advantage, yes it's better than suction but only the pressure difference is not material compared to the massive pressure your proposed recirc-SDS system needs to operate - I understand they need about 40 to 50 psi?

Some relevant facts:
1. Bendix FI runs continuously at just 3 or 4 psi.
2. Gravity feed in a Bearhawk will not run a Bendix, not even close, thus gravity head post losses is much less than 3 psi.

I believe EFII's and AFP's boost pumps put out 30 psi, non adjustable (no MAP reference, ATMO only) I thought they both claimed that is about 5 psi greater the the engine pump. Does the servo drop the pressure to 3 or 4? I was trying to figure out why the divider was so complicated. I never found an explanation, but it looks like the valve with the spring and diagram might also act to smooth out pressure variations. Just a guess.

The 1 psi of gravity pressure won't run the system, but as long as it can gravity feed faster than the pump is feeding, it should keep a slight positive pressure at the inlet. That is much better than several PSI of suction at the face of the pump. Much less chance of vapors coming out of suspension. I didn't realize why SDS and EFI don't recommend gascolators. I believe it is because they have a small pressure drop, and create some turbulence internally, both of which can cause vapors to form. But that is on a suction time system, like an RV. With a gravity fed system, it shouldn't be an issue, or at least much less of one. If I don't use my 2 liter header tank(with a drain at the low point), I will use the kit gascolator.

Gravity isn't much pressure, but it is a lot better than a bunch of suction.

90's fuel injection it was common to run the system at 30-35 psi. That is what EFII uses on their system now, as they use a fuel rail with a full return to the mains or header tank. SDS went away from that and use a small manifold that sits where the Bendix divider sits, on top of the engine. It is a manifold only, no moving parts. Fuel is returned from there, so from that manifold to the injectors, is return-less. So the place where the fuel is hottest on shutdown, will be at the inlet to the injectors. Because of this, SDS recommends their system be run at 45-50 psi. They said the extra pressure is because they already run return-less to the injectors. Helps with hot starts. A lot of newer cars run even higher pressure for the same reason.



As far as testing, the easiest test and worse case, is just turn a fuel pump on with the engine off. 100% of the fuel will be returned, continuously. Keep monitoring the temp of the pump and fuel. At cruise rpm and fuel flow, 50-80% of the fuel is still returned, so some of the same fuel will go through the pump 2-5 times. To me this is a cause for concern, but AFP rates their boost pump, which does the same exact thing, for continuous operation. My pumps will be mounted on an aluminum plate, maybe that will help to act as a heat sink/radiator. We will see.

[svyolo]

I still have a Bob rebuilt carb that I bought, pickled and wrapped in plastic. I bought it in case I couldn't figure out a fuel system layout for the FI that I liked. It has to work the the stock BH feed.

[Archer39J]

After looking at the gascolator I absolutely wouldn't use one with a full return. Pressure loss increases exponentially with flow, so 45GPH through that would be a non starter to me. For the reasons you state SDS says, if you have to, put a gascolator on the pressure side, still not ideal imo

[svyolo]

I read about half of AFP's Intallation Manual from their website. Under Documents. The first half of it is much less about their system, but aircraft fuel systems, Bendix FI, and the different ways to run and plumb FI. Things that work, and things to avoid. It is a wealth of knowledge. There is even a decent section on plumbing your air intake, and pitfalls to avoid there. It was definitely worth a look for me.

[Battson]

...
1 psi of gravity pressure won't run the system, but as long as it can gravity feed faster than the pump is feeding, it should keep a slight positive pressure at the inlet. That is much better than several PSI of suction at the face of the pump
....
Gravity isn't much pressure, but it is a lot better than a bunch of suction.

If we assume a metre of head generates +0.15psi in the fuel system, then one might assume -300mm of head in a low wing would relate to -0.05psi [suction] assuming a linear relationship and all other things being equal.

That was my thinking when I said the difference is probably immaterial compared to 45psi required by the system.

Yes it does help, however it might be an extra ~0.5%. A lot better might be overstating it.

[Battson]

But the both systems have the same dynamic pressure loss through 3/8 lines, so the different between dynamic and static is common to both assuming both systems are the same up to the point of fuel supply... In reality, the more complex piping in the Bearhawk could add to the pressure loss, or not if you chose a low-wing with complex plumbing. You would need to test it to know for sure. But it's surely an insignificant difference anyway, so it's a moot point I guess.

[svyolo]

Yeah I think you are right. The BH lines are a little longer than the RV. But the BH has an extra T over the RV's, but I am guessing at the RV fuel system. I couldn't find an equation, but found a really detailed on line calculator. I had to estimate the losses from a gascolator and fuel valve and I estimated them as having the same loss as T's. I am not sure how accurate that is, but for comparison purposes it doesn't matter because the RV and BH both have them.

Feeding 25 gph through 3/8" versatile (.305" ID) was around 2 psi loss. Bumping up the flow to 45 gph (SDS with full return) it jumps to over 5 psi. 1/2" lines drop that down to way under 1 psi, but that assumes that the gascolator and fuel valve are a size up as well.

[svyolo]

That calculator had a lot of assumptions. Based on the BH fuel flow tests I kind of think the real measured loss is a bit under 1 psi. Not a huge amount if you are running Avgas all the time. but I never really thought about the vapor pressures of fuel and how that related to atmospheric, and pumping. I have been out of this side of aviating for a long time, and didn't do much of it, when I did it.
Avgas is almost twice the price of Mogas here. The vapor margins for running Mogas are thin. Flying high or operating from high airports in the summer you might not have any margin, or worse. I guess if you are that close, you shouldn't run Mogas.