What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Rigid Fuel Lines- What is too short?

Freemasm

Well Known Member
Sincere question looking for legitimate answer. What constitutes "too short" for rigid fuel lines? I'm aware of the variables (relative rigidity of structure between the two connecting points. Linear versus planar versus compound curve, etc) and consequences (related strain = stress in line and fittings.)

Background: I mocked up my wing/flop tube connection to the fuel selector. Too long, too many bends, and too many bulkheads to pass through to make in one piece. Of course, I didn't think about until after I'd made the line; another waste of effort and potential material. Story of my life. The line is ~6.4 inches and almost a single plane curve; pic attached. I thought about forming a steam loop in it (still might) but I'm comfortable with the relative rigidity of the structure, direction of forces/deflection and resultant strain in the line. Anyone is welcome to convince me otherwise.

I didn't find anything in my ancient copy of AC65-9A. I like to follow codes, standards, etc, when able. Any references that can be passed along? It would be appreciated.
 

Attachments

  • FLop tube connection.jpg
    FLop tube connection.jpg
    497.9 KB · Views: 436
If Vans put it into the plans I would be comfortable using it. But ease of assembly could yield a break somewhere in there.

I made a tubing straightener after practicing making some difficult pieces, I straightened the practice piece and used it for other practice. Then made a perfect virgin bent piece to impress inspectors. :eek:
 
It's a Rocket. No specific fuel system layout to draw from. Plus, I'm attempting to have the fuel system configured for the lower VP of Mogas in case I'm ever forced in that direction. Hence, trying to keep the pump suction losses as low as possible; no el's or bulkhead el's, smooth bends, etc.
 
The rockets I’ve tinkered on had the 3/8 feed line run back along the tank to a spot to penetrate the skin aft of that bulkhead, than straight across to the fuel valve- one piece. Also note where the tank vent line will go, usually straight thru the skin, than turn forward thru that bulkhead (about where you have your 3/8 bulkhead fitting is now) than run straight up along the front inboard edge of the vertical bulkhead to the longer on, than forward to the firewall, than turn down to the floor & out.
Also I note the hinge on the cover plate on that space behind the bulkhead. I hope you will have enough screws holding it soundly down as that cover should be considered as structural.
 
Do you know that the suction losses are using AN hardware?
Tom

There should be no suction losses using AN hardware. After installing a mini split Air conditioner with flared copper tubing and brass fittings, the system was vacuumed down to 500 microns and held for over an hour.

So I don't understand how a properly installed AN fitting would cause any suction loss unless you are talking about creating turbulence as the fuel passes by.
 
As an aside, have the rivets already been set in the doubler that the male B fitting passes through? The reason that I'm asking is that there appears to be a gap between the parts where the bottom left rivet goes, like they were not held together tight while riveting.
 
The rockets I’ve tinkered on had the 3/8 feed line run back along the tank to a spot to penetrate the skin aft of that bulkhead, than straight across to the fuel valve- one piece. Also note where the tank vent line will go, usually straight thru the skin, than turn forward thru that bulkhead (about where you have your 3/8 bulkhead fitting is now) than run straight up along the front inboard edge of the vertical bulkhead to the longer on, than forward to the firewall, than turn down to the floor & out.
Also I note the hinge on the cover plate on that space behind the bulkhead. I hope you will have enough screws holding it soundly down as that cover should be considered as structural.

Thanks Ralph.

The run you mention was contemplated but I would had to make a (very) compounded curve to get past the fwd attach brackets. As mentioned, trying hard to keep those pump suction losses minimized in case I have to go to Mogas (summer blends) one day. I'll be duel electric pump (and elect. architecture before anyone asks) so no real elevation losses. The parasitic losses of a constant 35 GPM flow is the challenge; hence, my approach regarding limited use of fittings. Won't know until the whole thing is done and tested what my suction margins are. Can plan and calculate all day. The proof will be no phase changes under full flow at ~120F fuel temp.

Rockets do their tank venting differently. I'm still inquiring how it evades physics but they swear by it. A double loop is added in the tank wing root and exits there. Here's a pic from the manual (showing single loop) below.


As far as the cover/floor observation, I don't have a real answer. I too wondered if it were secondary structure. I try not to justify anything here without applying my own justification/rationale/calculations (not always the norm here). The mod is at least a couple of decades old and I'd say "common" in Rocket world and at least a few RV-4s from what I've found in blogs and other. The F1 fuse longs and skins are "upgraded" to thicker gages, wider angles, etc. Were any calcs done? No idea. Has it proved itself as a safe config? Seems to have but it's not for me to say. Sorry I don't have a real answer but will forward what can find out.

I'm quite sure you are fine. Looks good to me.

I've never heard of a minimum length requirement or concern.

The related structures aren't that rigid. There will be some relative movement between the connection points. Strain (which results in/is linear with resultant stress) is deformation/unit length so the resultant stresses are less with a longer "distance"/length of pipe/tube/structure/etc between them. Short and straight is the worst combo. I've looked for an answer but can't find anything in writing. I think my going forward rule of thumb will be ~1/2 of the allowable unsupported length. This is based on the assumption that some high degree of rigidity was assumed/tested when adding a support when exceeding the max allowable and the tube stresses were still in check. I'll ask again, if anyone with appropriate knowledge/experience can weigh in I would appreciate it.

Do you know that the suction losses are using AN hardware?
Tom

Measured? No. A shop air "test" showed a pretty noticeable flow difference between a 90 deg El and a 90 bend. Good guess is the original system requirements weren't thinking about full flow fuel systems when they went into the build manuals. The losses are a function of the square of the flow so it gets big, quick. The resultant differences between new and legacy systems is pretty huge. I chose a Aerolab gascolator versus a relatively cheap/already in-hand in-line filter for packaging purposes alone; greatly simplified (less bends/less losses) the tubing install in the tunnel. See pic 2 below. Will I be right in the end, who knows? I'm trying to hedge my bets/keep the design margins.


There should be no suction losses using AN hardware. After installing a mini split Air conditioner with flared copper tubing and brass fittings, the system was vacuumed down to 500 microns and held for over an hour.

So I don't understand how a properly installed AN fitting would cause any suction loss unless you are talking about creating turbulence as the fuel passes by.

Yes Sir, the frictional losses that relate to static P on the pump suction side of things. Tried to hit on it in the previous replies. We're talking design margins that could be measured in inches WC. Vapor lock concerns me. Proving it out would be one less concern bumping around in my brain.

As an aside, have the rivets already been set in the doubler that the male B fitting passes through? The reason that I'm asking is that there appears to be a gap between the parts where the bottom left rivet goes, like they were not held together tight while riveting.

Guilty AF. Saw that in the photo as well. Hi-def photos beats the cr@p out of my old a$$ eyes. Should have held it with the BH fitting instead of just clecos. That one gets replaced.

Thanks to all. Any help with the original question is appreciated.

Cheers boys.
 

Attachments

  • tank vent.jpg
    tank vent.jpg
    676.9 KB · Views: 287
  • gasco4.jpg
    gasco4.jpg
    846.9 KB · Views: 229
Last edited:
Back
Top