The inevitable fuel spill disaster

[DCASonex]

Photos and drawings of my modifications to the A series Sonex fuel filler box can be found on the Sonex Foundation website, under fuel vent and fill box about 2/3rds of the way down here: https://sonexfoundation.com/the-amsler-files.

David A.

[BobDz]

Thanks for that link David!
Excellent info!

[13brv3]

Back from the dead. It only took 2 weeks and $1300 to correct a momentary lapse in attention... I considered some sort of box around the fill cap, but I don't want anything blocking my view permanently, so I put it back like it was. I'm working on an improved version of the fuel fill funnel to keep fumes and any overflow out of the cockpit.

Along the way I added a fuel return to the tank to hopefully cure some really odd intermittent low fuel pressure readings. Rotax "requires" a restricted fuel return to eliminate any vapor lock, and I do think that was the cause of some of my low pressure readings.

I also took the opportunity to install the Peter Anson canopy latch. It required some modification because I put rivets in the way of the block. I'd recommend getting one of these latches when you're originally building the canopy frame, because it's about 1000 times better than the stock arrangement on a Onex. That's his fuel cap too, BTW.

Naturally, the weather is forecast to be crappy for the next week. And because I can never say it enough, I HATE fitting canopies!

[WesRagle]

Hi Guys,

Catching up on reading.

A few years ago we did a fuel, windshield test. We took a piece of windshield and duncked it in fuel. No effect. Then bend the piece and put some stress on the surface and drip a few drops on the surface and you get cracks. I presume if our windshields were composed of flat plates we wouldn't have a problem. All the fuel is doing is relieving the stress of the bending.

And here is my experiment along the same lines.

https://sonexbuilders.net/viewtopic.php?f=58&t=5813&hilit=stressed#p43915

The (partial) solution I am using. A 15 gal hand crank fuel transfer tank. I insert the spout and hold it in place while I crank with the other hand being careful not to over fill.

I agree that this is a horrible place for the fuel fill. I understand wanting to make the build easier, and to reduce parts count, but .....



Wes

[sonex892.]

Fueling even the A model sonex is not ideal. I mostly use mogas and would sometimes spill fuel when pouring from the 20 litre container.

Here is how I now syphon fill. The larger tube extends to the bottom of the container. Blowing into small tube pressurises the container to start the syphon. Its not real quick but I have never spilt fuel when using it.

[Skippydiesel]

Random thought;

Why not have a section (larger than the usually small hatch) that hinges back towards the canopy, staying almost vertical. The hinge line could be made splash proof and the large "door" would protect the canopy from almost every possible fuel spill/splash/spatter scenario.

Down side would be the risk of opening in flight but I am sure some fail safe latching mechanism could be applied.

Note: My Sonex centre tank, is filled vie either wing tank, so no fuel close to canopy.

[13brv3]

I seriously considered making a metal box that enclosed the fuel cap, but in the end I just couldn't accept having something like that permanently blocking part of my forward view. The plans call for an aluminum cover for the fuel hole in the windscreen, but I like the clear polycarbonate cover for obvious reasons. I stole that idea from someone, maybe Wes?

I wish I knew how much fuel spilled, but I can't see how it was more than a few ounces. The problem is that the curved glareshield channels the water to the left and right, where it runs along the side toward the back of the windscreen on a taildragger. A little liquid made it all the way back to the rear of the windscreen on the left side, and at least a few drips ran off the rail. I never saw liquid fuel on the right side where the canopy hinge was, but some must have made it there, and wicked under the hinge. I don't think brief fuel exposure would hurt the Lexan or Acrylic in an area where there are no edges or drilled holes.

For the first year or so, I used these pumps- https://www.amazon.com/gp/product/B010GNM7H4 They're a little slow, but they work well, and have an auto shutoff. I'd never trust that, but it was nice as a backup, and I did test it to make sure it worked. Later, I wanted something faster, and with a good filter, so I put together a fuel cart with a 12V pump, powered by a Dewalt 20V pack via 12V adapter. This works really well, but there's no auto shutoff. I did try an auto shutoff nozzle, but it just won't prime and pump fuel uphill to the tank. If you point it into the tank opening to get it started, then it will work, but that was (or seemed like) too much effort.

I'm closing in on a final fuel funnel solution, maybe finished later this week.

[Jester504]

For my -A I'm going to make a plastic sheet with a hold in the middle that is slightly smaller than my fuel cap. It'll cover everything around for 2-3 feet around and live in the cargo compartment. I might also permanently attach a funnel to that hole so there is literally nowhere for fuel to go except in the tank or on the ground.

[Skippydiesel]

13brv3,

Nice neat refueling set up.

On the matter of pump priming - from the photo it looks very like your pump is a centrifugal type. Centrifugal do not easily self prim when mounted above the liquid source. You have two choices;
[list=]For best scavenging (removing all fuel) mount the pump below the tank. Mounting at the same level will also work but if pump is stopped/shut off when the fuel level is low it may not self prime, unless you fit a non return valve.
Fit a posative displacement pump. I use a cheap copy of a high volume Holly - its brilliant! - self primes and will easily lift fuel to the tanks on high wing. Down side is that you can not use a tap/shut off nozzle on the delivery end without fitting a pressure relief bypass circuit.[/list]

[13brv3]

Priming might not have been the right word for it. The pump came with a standard nozzle that doesn't have any auto-shutoff feature. It works perfectly with that, and fuel flow seems instant. The picture actually has the auto-shutoff nozzle that I bought for it, but then abandoned. The way the auto-shutoff nozzles work is really clever, but seems to require some liquid pressure to allow the auto-shutoff to disengage. I'm going to revisit this and try to make it work.