Quantcast

Are tube spars an insult to the engineering community?

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

jedi

Well-Known Member
Joined
Aug 8, 2009
Messages
2,224
Location
Sahuarita Arizona, Renton Washington, USA
The title of this thread (Are tube spars an insult to the engineering community?) is taken from a post in the "Prandtl lift distribution for conventional configurations" thread?

This inference has been mentioned several times in various areas. I think the subject needs more discussion, in particular in relation to a D spar.

I expect there are several unspecified assumptions among those making the negative statements about tube spars. I have the following questions.

What are those assumptions?

Straight constant wall tube versus tapered or stepped spar or variable wall thickness.

Assuming a cantilever Hershey bar wing what is the % weight penalty for a tube versus I beam?

How is this modified by a strutted wing?

Is the D spar designed as an I beam with a leading edge D for additional torsional stiffness?

If the D spar is constrained to the same material for the entire D is it still superior to the tube spar?

Does the tube versus I beam spar assume or require both a forward and aft I beam spar versus a single tube spar or Ultralight like leading edge and trailing edge tube spars?

How does a non Hershey Bar plan form change the results of the comparison? Taper, elliptical, sweep, etc. ?

How does airfoil pitching moment and / or twist affect the trade study?

All comments welcome.
 
Last edited:

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
4,131
Location
Warren, VT USA
Manufacture of individual parts. Time to fabricate, cost, complexity of wing assembly to build. Tubes are viable. It is a compromise that structures nerds hate. But it has been proven for decades to work adequately within the context of the entire wing design.

Tube truss fuselages people don't hate for some reason. Everything is a compromise.

Tube spars don't work as easily for tapered wings but can. The Bede prototype used a rolled sheet tube for the spar. That get's rid of the problem of sourcing DOM aluminum extrusions completely and was pretty easy to fabricate in the overall scheme of things. A composite tube wrapped over a mandrel is a pretty easy thing to fabricate as well and can have a ply drop schedule to tailor bending/buckling loads.

Gossamer Albatross, Condor... Paul MacReady not dumb.

Devil is in the details.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
14,013
Location
Port Townsend WA
How does airfoil pitching moment and / or twist affect the trade study?
A simple D shaped tube with thick caps would be a more efficient beam spar than a round tube. But a single spar is generally placed at 25% chord. So how do you place the single spar at the leading edge without excess pitch down twist? It could be done if designed correctly.
 

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
770
Location
Lt
A composite tube wrapped over a mandrel is a pretty easy thing to fabricate as well and can have a ply drop schedule to tailor bending/buckling loads.
Yeah, but done on hot wired foam plug is not much more complicated, but gives your another level of experience..
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,369
Location
Rocky Mountains
Tube truss fuselages - some - people don't hate for some reason. Everything is a compromise.
Changed quote

I don't like either. I'm one of those structures nerds. But I see more reason to use a tube for a wing than for a fuselage.
A tube can be an elegant engineering solution - depending on the project parameters. I've just never been interested in a project where the parameters lead to a tube spar or fuselage.
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
5,365
Location
US
Short answer: Engineers in the real world make compromises all time to allow for lower costs, easier fabrication, etc. If a design meets the requirements and the compromises are the best fit for the overall objectives of the project, then there's nothing wrong with a less-than-theoretically-ideal part.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
4,131
Location
Warren, VT USA
It is about one structural member doing both bending AND torsion. The Bede 4 wing is about the perfect compromise. If it had held fuel as planned it would have been fully genius. 12" bays single mold with joggle to bond to the next section with a flange for mounting to the tube and a rib molded in each one. One rib bay mold (female) so outer skin requires little or no finishing just sanding out the ooze from the joints or taping smooth. The rear spar flap/ail gap and shear web molded into each bay. Rigging and building is trivial. Two saw horses to hold the tube. Level a string to make sure the TE stays level to the wing. Prop in some washout just using adhesive clearances or just rig ailerons slightly high or put twist in the aileron. Wing structure built in hours not weeks or months. Simple parts. Ailerons and flaps have a tube in tube drive but are the same skin, simple radius bend leading edge, rivetted trailing edge, a few foam bonded ribs or metal riveted if preferred.

Could be the simplest wing ever conceived for a fast 4 place plane. Never, ever had a structural failure. Huge amounts of HP hung on these things. Easy to add or subtract span. Longer central joiner tube to move the buckling farther outboard, add on tips were just tube fit in main spar and more rib bays. Sort of a wing erector set. I wouldn't build a wing that way heading for composites but for what it was considering designed in the 60s I think it was pretty clever.

Compromise, you bet.

Sold my project in the end more due to circumstances but not the design. Always appreciated the overall platform as a good experimenters platform.
 

pictsidhe

Banned
Joined
Jul 15, 2014
Messages
8,812
Location
North Carolina
They are very simple to 'build' compared to built up spar. They can also take bending, torsion and drag/antidrag loads. An all in one spar that is clickety-click for most if it's construction. They are well suited to fabric designs which can't resist any torsion or drag/antidrag on their own. A rigid skinned wing can do that, so a tube is getting less efficiently used, there.
 

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
6,287
Location
Australian
Lots of Gruman Yankees say no.

Bede design of course, and nothing wrong with the other Bedes either.

The Grunman Yankee specs are as follows to give people a starting point:


The AA1 series spar outside diameters are 6.625” for the carry through and 6.75” for the wing.

In metric, 168.3mm for the carry though, with a 5mm wall thickness, and 171.3 mm for the wing panel, with wall thickness 2.6mm.

On the AA1, AA1B and AA1C both carry through and wing were 2024 T3.

On the AA1A, the carry through was 2024T3, and the wing was 6061T6.

The ends of the carry through and wing spars where the wing slides over the carry through are sized (expanded) to provide a precise fit, instead of relying on the tube manufacturing tolerances.

The clearance allowed is 020” or 0.5mm, and if it is more than this, shims have to be used. The overlap is 290mm long.
 

Victor Bravo

Well-Known Member
HBA Supporter
Joined
Jul 30, 2014
Messages
8,060
Location
KWHP, Los Angeles CA, USA
The Kolb ultralights and LSA's use tube spars very well. One of the clever things is that they moved the tube spar rearward, further aft than the maximum thickness point of the airfoil. This balanced the torsional loads, so that the nose-down pitching moment from the airfoil balanced the nose-up loads on the leading edge from having the spar further back... so there was no net torsion load on the spar tube. The two opposing twisting loads just canceled each other out.
 

Norman

Well-Known Member
HBA Supporter
Joined
Nov 28, 2003
Messages
3,047
Location
Grand Junction, Colorado
Tubes make great posts. A hollow circular tube is a very good shape for a compression member and pretty good in tension so a fine cross section for truss elements. Circular tubes are not as good for beams though because 90% of the material is not in the right place so they end up being heavier than a good beam cross section such as an I-beam, C-channel, or even a square box-beam in the same application. This is because these other cross sections place most of the material in the flanges and have thin, strait, vertical webbing. If you look at the cross section of an I-beam you'll see that most of the material is in the flanges at the maximum distance from the center (the neutral axis). The distance of a "fiber" (literally fibers if it's wood, figuratively for metals) of the material from the neutral axis is what determines that fiber's contribution to the strength and stiffness of a beam so you want most of the fibers in the flanges. The webbing only sets the distance between flanges and resists sheer forces, the web can be surprisingly thin. A thick web increases weight by steeling material from the flanges which weakens the beam so you have to add more material to the flanges to get the strength back. Also when the beam fails the webbing bends.

In a round tube there basically isn't any flange and most of the material is in what could be considered the webbing but it's thick and curved. Remember what I said about bent webbing? A beam with a tube cross section is pre-failed.

I wouldn't call using round tubes for beams in severely weight restricted applications "insulting" but it's not very smart to waste material that could be used elsewhere.
 
Last edited:

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
770
Location
Lt
The Bede 4 wing is about the perfect compromise. If it had held fuel as planned it would have been fully genius. 12" bays single mold with joggle to bond to the next section with a flange for mounting to the tube and a rib molded in each one. One rib bay mold (female) so outer skin requires little or no finishing just sanding out the ooze from the joints or taping smooth.
Could we think of something similar, but with C beam ? Plus as far as i know - many have added additional ribs to each bays. And it goes well with Jim Marske monarch - he placed full ribs each 18" + false ribs beetween. False ribs could be done with Boku pe strips. And then bays shells be made in longer pieces - just to be ok to reach by hand.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,833
Location
Memphis, TN
Elegant engineering. It all depends on the solution you are try to engineer.

Bede had one thing in his mind, ease of building. An AA1 wing panel is a spar tube, 5 ribs, 6 skin panels, cove, and a wing tip. Yes you still have flight controls, identical flap and aileron skins and end caps. 14 pieces in the main wing panel. That is pretty elegant. Pretty smart. The but is it’s not pretty. Quite industrial in fact. Not a show off. 1969 bonded built airplane is quite forward thinking, though.

BD-4 almost to far ahead. If the original wing buckets were made today and designed to have 2/3s fewer needed for the same span round off the top of the fuselage and windshield, people would go Gaga.

Traditional aluminum, is elegant as it steps loads for need; less waisted weight penalty. There is elegantly placed parts, sized. Might even look a little Deco. Harder to build, but not so hard, but it takes time. More thought has to go into it designing it to get the payoff. Same with building. Elegance except in a pretty shape is muddled in necessity.

Composites. The exact opposite of the tube spar. Elegance in being smart enough to take a bolt of fabric, some glue, and an occasional styrofoam coffee cup or a million, and make an airplane in just about any shape you want. That’s freakin amazing. The problem is you better be smart. It’s elegant. Done right, it’s a sculpture, done wrong and it looks like a bad Transporter accident. People building with it find it easy or super hard. You got a half gallon of glue mixed, a whole bunch of fabric cut out. No pee breaks, no phone calls, no food. You go until done. Well organized and schooled get through. Oh the messes made that don’t.

Wood. It’s elegant like the tube. It’s a chunk of wood. It’s a good chance you are sitting at a wood table in a wood chair. Relatable. Quality aside, you can go into your front yard cut the tree down and make an airplane. Is a Tailwind wing elegant. Spar and drag spar, some ribs, and two sheets of 4x8 ply for skins. That’s it. Talent to build is subjective, but it’s almost as simple as the wing you tried to build as a kid in the garage in a week. The talent is subjective, but because it’s relatable, it gets a boost over working with aluminum. An Emarude or a Jodel. The thought of making the wing the main structure and adding the fuselage is completely opposite of what others do of wings added to a fuselage. Different solution, is that not elegant? Not the most space efficient, so not all is perfect. Mini Max. As easy and relatable to build as anything possible. Is that not elegant?

They are all elegant. They may not be your elegant. That’s good.
 

ScaleBirdsScott

Well-Known Member
Joined
Feb 10, 2015
Messages
1,274
Location
Uncasville, CT
As definitely not a member of the engineering community, I feel like if one had to get custom extrusions made up anyway, I'd just go with a highly engineered, rectangular tube section, maybe even a Double-I section. Then you can make a one-in-done box spar, made to the specific desired dimensions, with extra material and skin riveting flanges where needed (if needed). Of course the web sections would be the challenge as a big extrusion like that you're going to run into issues with thinning the web too much. So go with whatever is minimally possible while being reliable, then have a CNC that goes in and pockets out the excess material in that web, with likely some full-on lightening holes happening by the 2/3 mark or sooner. This assumes the rest of the design is going to be manufactured on some kind of equipment as well. Even if not, building a specialty CNC machine that just cuts spars is actually not hard to do. Just make a very narrow, very long version of the common CNC router table. That'll remove weight from spars all day. And if one was smart in the design that table could also cut a lot of ribs from strips of sheet when it isn't on spar duty. Just an idea.

But if there's off the shelf tube available, then tube is proven. My father's Grumman Cheetah is still going strong on its tube spar, and then you don't have to go get custom extrusions, and being tube probably no point going and trying to relieve it of material too much so it just needs to be cut to length.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
4,131
Location
Warren, VT USA
Tubes make great posts. A hollow circular tube is a very good shape for a compression member and pretty good in tension so a fine cross section for truss elements. Circular tubes are not as good for beams though because 90% of the material is not in the right place so they end up being heavier than a good beam cross section such as an I-beam, C-channel, or even a square box-beam in the same application. This is because these other cross sections place most of the material in the flanges and have thin, strait, vertical webbing. If you look at the cross section of an I-beam you'll see that most of the material is in the flanges at the maximum distance from the center (the neutral axis). The distance of a "fiber" (literally fibers if it's wood, figuratively for metals) of the material from the neutral axis is what determines that fiber's contribution to the strength and stiffness of a beam so you want most of the fibers in the flanges. The webbing only sets the distance between flanges and resists sheer forces, the web can be surprisingly thin. A thick web increases weight by steeling material from the flanges which weakens the beam so you have to add more material to the flanges to get the strength back. Also when the beam fails the webbing bends.

In a round tube there basically isn't any flange and most of the material is in what could be considered the webbing but it's thick and curved. Remember what I said about bent webbing? A beam with a tube cross section is pre-failed.

I wouldn't call using tubes for beams in severely weight restricted applications "insulting" but it's not very smart to waste material that could be used elsewhere.
Classic engineering review. First it makes assumptions that that spar has companions and in some way the torsion is not really carried by the spar. If that is the case then yes. But when you only have one spar, one structural member and no rear spar then all the bending AND torsion goes through that one shape. Then that tube is pretty efficient. You wouldn't use an I beam for a tail boom... A consistent wall thickness is the compromise. A tube with a central web would be a great compromise.

That is the Bede spar. That tube is the only structural member that isn't skin or ribs. And it was easy to make back then and not that expensive and nesting tubes got rid of all the carry through parts. Slide tubes together. It was and it a good concept and a good compromise.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
4,131
Location
Warren, VT USA
Apples and oranges. D tube could be better at the cost of higher parts count, harder parts to make, harder joints to get right, etc... Not the point. Optimization can go toward any goal. If weight is the goal then complexity might suffer or parts count or both. Skill of the assembler is another one and required tooling, fixturing.
 

Norman

Well-Known Member
HBA Supporter
Joined
Nov 28, 2003
Messages
3,047
Location
Grand Junction, Colorado
Classic engineering review. First it makes assumptions that that spar has companions and in some way the torsion is not really carried by the spar. If that is the case then yes. But when you only have one spar, one structural member and no rear spar then all the bending AND torsion goes through that one shape. Then that tube is pretty efficient. You wouldn't use an I beam for a tail boom... A consistent wall thickness is the compromise. A tube with a central web would be a great compromise.

That is the Bede spar. That tube is the only structural member that isn't skin or ribs. And it was easy to make back then and not that expensive and nesting tubes got rid of all the carry through parts. Slide tubes together. It was and it a good concept and a good compromise.
I agree that it worked and it was a good compromise for simplicity but nothing else. A built up spar is not a horribly complex thing. Two sticks and some 1mm plywood makes a very good and light beam. yes it won't resist torsion at all but that same 1mm ply also makes a good leading edge skin and if that sheeting is glued to the two sticks you now have a D-tube that can resist torsion better than a round tube spar because torsional rigidity is proportional to the cross sectional area enclosed by the tube and a D-tube has a much larger cross section than a round tube. Of course there's some loss of structural efficiency due to the out of roundness of the D but it's not that much. You could also constrain the spar against twisting by bracing it against the trailing edge or a compression spar. Also the ribs break up the D-tube into short segments which massively increases the torsional stiffness so the wall thickness can be pretty small.



All I'm really saying is that a vertical spar with a constraining system such as a D-tube will always be lighter than a round tube spar of the same material. If you know that I'm wrong about that pleas explain.AreaOfTubeSpar.jpg
 
Last edited:

stanislavz

Well-Known Member
HBA Supporter
Joined
Aug 21, 2016
Messages
770
Location
Lt
Apples and oranges. D tube could be better at the cost of higher parts count, harder parts to make, harder joints to get right, etc... Not the point. Optimization can go toward any goal. If weight is the goal then complexity might suffer or parts count or both. Skill of the assembler is another one and required tooling, fixturing.
Different question - you are building extrusion mandrel. Price is same for tube or D spar. Same weight. which one to choose ?
 
Top