maniago
Well Known Member
I put ~190 hrs on a Swift that had Bendix injection (Lyc IO-320, then IO-360 engines) that had an STC'd aux tank. Standard practice on cross countries was to TO-main, switch to aux, run the aux dry, and switch back to main. You always knew within a few minutes when the aux would run dry, so you just monitored the fuel flow/pressure gauge & switched at the 1st drop in pressure & hit the boost pump. Yes, occasionally you could get distracted and the engine would cough/sputter before switching. The engine driven pump has no problem recovering; it'll just recover faster if the boost pump pushes fuel up there.
If the AFP pump is a gerotor or roller vane style pump (and it's almost certainly one or the other), the [AFP] statement makes no sense as written. Gerotor and roller-vane pumps are positive displacement pumps; they will self-prime at any level of lift you're likely to see in the planes we fly. Self-priming by definition means that there will be a bubble of air on the output side of the pump after the pump self primes; it means that the pump is pulling air through itself to get the gas up there. That air has to go somewhere; it doesn't just disappear. So....when the boost pump self-primes, there will always be a (temporary) bubble of air on the output side of the boost pump.
If AFP used a turbine style pump in that assembly, they made a grievous error in pump selection.
EDIT: Just realized what they might be talking about. Positive displacement pumps must have a regulator, to avoid damaging downstream stuff or stalling the pump. The regulator on that pump (and the other brands that are configured similarly) bypasses excess fuel *back to the intake of the pump*. I can see how the pump could fail to pump (gas), once there's air in the system around the pump. It's a lot easier to move air than fuel, so the pump *system* can vapor lock *itself*, if air ever gets in the line.
Something to think about, while flying over that forest, if you think there's ever any risk of unporting your fuel pickup while operating on the boost pump....
Would not be an issue, if the regulator bypass went back to the tank, as is done with the auto style injection systems that are gaining popularity on this forum.
So this is all good stuff. The operative question then is if the AFP has this problem, but your Swift pump (and presumably any stock RSA system e-pump) does not:
1)how do those pumps differ from the AFP
2)what make and model are those pumps
3)why would anyone choose to use anything but those pumps?
I havent decided on a boost pump, so this is an important issue for me, and until this thread, I hadnt even considered it an issue. Thanks for any help.
------
Edit: Ok I get it. Weldon pumps are Piper stockers, 8120G or 8150s. I now get that the AFP is meant for EFI - guess I've been ignoring that, despite it being obvious. I now also get why the Andair pumps are the obvious go-to pumps for the RSA style inj for EAB. I'll leave this post here; maybe it'll help someone else in the future.
Last edited: