The AFB (Amazing FleaBike)

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
15,448
Location
Memphis, TN
Joints are a pain on a straight spar, adding a crook in it makes it a monster. Straight spar, even if the platform around it curves. Unless you want two concurrent spars? A bent spar wants to twist with lift. That’s extra twist from normal flying twist. Swept/ tapered platform is twisting the spar too, with varying lift at each station. It adds up quick. There is a reason rectangle wings are used, so you don’t have to figure all the extra out.
 

rotax618

Well-Known Member
Joined
Oct 31, 2005
Messages
1,351
Location
Evans Head Australia
If you are to use the traditional Flea 3 point wing mount with 3 hinge points then your curved swept spar makes that very complex for alignmen.
 

Victor Bravo

Well-Known Member
HBA Supporter
Joined
Jul 30, 2014
Messages
9,583
Location
KWHP, Los Angeles CA, USA
Duncan if you tie in the tip of the (swept or curved) main spar directly to the tip of the (straight) rear spar out near the wingtip, you can get away with it a lot easier.

The attachment to the rear spar keeps prevents the tip of the main spar from moving or twisting, which keeps it from losing its ability to carry the load much better than if it were just hanging there in space. You will need a good structural rib between the main spar and the rear spar at each "break" or joints, in addition to the other "regular" ribs.

There are several highly experienced engineers here who can quantify and qualify and calculate and extrapolate all this out to eight decimal places, and I'm not one of them. I'm just giving you the general principle on a model airplane builder level :)
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
Duncan if you tie in the tip of the (swept or curved) main spar directly to the tip of the (straight) rear spar out near the wingtip, you can get away with it a lot easier.
So you are suggesting TWO spars? And the reason for this is to anchor the front and rear portions of the wing to counter twist?

The attachment to the rear spar keeps prevents the tip of the main spar from moving or twisting, which keeps it from losing its ability to carry the load much better than if it were just hanging there in space.
In my configuration, I have two heavy duty steel hinges with removable pins (red circles) very firmly anchoring panel A to panel B both in torsion and to counter vertical lift. The green circles do the same job between panels C and D. How would this be any less effective than having an extra spar in this position?

1623448349634.png
Don't get me wrong - I'm not married to the single spar concept. If two spars are demonstrably better, I'll change the design. It's not too late. If I use two spars, are you suggesting the rear spar is straight?
1623448930591.png
Or are you suggesting the following arrangement?
1623450353694.png
 
Last edited:

WonderousMountain

Well-Known Member
Joined
Apr 10, 2010
Messages
2,408
Location
Clatsop, Or
Am I too late to join the party?
My question was about panel length.
The Chord er usual, is not to scale.
If you will look back, to my post 479;
IMG_20210611_183047.jpg
As to the Spar, just run A-B straight.
That way you have the most critical
connection in the best position over all.

Sincerely, ~CK LuPii
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
LuPii,
Why do you think that adjacent panels will benefit from having their spars aligned in a straight line? In other words, WHY does panel B's spar need to be in line with that of panel A? Simply giving advice without reasons doesn't actually help me.

Duncan
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
If I use two spars (like in the second alternative diagram), this wing is going to be as strong as a brick s*h*i*t house, but I checked the weights, and almost no difference. Maybe this is the way I'll go.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
Day 3:
Well, the epoxy has cured, so it was time to put the two sides in the cradles. First job was to fit the firewall, aligning it as accurately as possible. It fits perfectly between the extended skins, so it was just a matter of aligning it vertically. A quick drill through the provided holes into the airframe, and the application of a screw to keep the firewall in place. On ONE side only.

Then on to the tailstock. Repeat. Finally, align and screw in the other side.

Next step is to fit the cowl base, seat, seat-back and turtledeck. Fitting these pulls the fuselage "square"

1623467239456.png
In the above photo, you can see the fuselage resting in the fuselage cradles. All external "bulkheads" are fitted (no epoxy, just dry fitting at this stage). It's amazing how these bulkheads pull the fuselage STRAIGHT.

Here's a close-up view of the fuselage seat and two-part control panel. Notice first that the seat is split into two. The front section is fixed with 6mm screws into captured nuts, and the rear bit (the "bum" bit") is hinged to allow easy access to the storage area under the seat.

The upper control panel will accommodate a full-size iPad running OzRunways. The lower panel will accommodate a mini-iPad running my digital engine monitor system.
1623467460106.png

There was one SNAFU, however. I got the width of the turtledeck wrong by 4mm. BUGGER. But I have lots of 3mm gaboon ply left over, so with a bit of careful measuring, I was able to re-CAD this part, and I'll cut it tomorrow.
1623467827066.png

So, that's how far I've come in 3 days.

Build time: 30 min

I don't know if anyone will ever want to build one of these things, but if they do, it's going to be absolute plain sailing.

Tomorrow I'll fit the aft (internal) bulkheads, and figure out in what sequence to bond them in place. Then I'll do it.

Duncan
 

WonderousMountain

Well-Known Member
Joined
Apr 10, 2010
Messages
2,408
Location
Clatsop, Or
With an abrupt angle spar joint, lift of the outer panels is converted to torsion,
Same as with coefficient of moment. Except, all at once. Your I beam would
not react well in torsion. The moment of inertia in tortion goes agains the wide axis, not tall. Your spar becomes two caps & a web deep 19x2+6 44-45 mm.
The amount of torsion converted is based on the angle made with the next
panel, that is it's spar. So if you place A-B straight, the amount converted is
none, and coefficient of moment is merely carried to the next section. So while some of my suggestion is aesthetic that part is not. My cranked planeform is never for C or I beam, always a closed structural. A continuos curve, would not create a point load, spreading out the force. You can do it your original way, I and the others just want you to consider what you are getting into. It does look Organic (°∆°). Find the best compromise for you.
 

Victor Bravo

Well-Known Member
HBA Supporter
Joined
Jul 30, 2014
Messages
9,583
Location
KWHP, Los Angeles CA, USA
If I use two spars, are you suggesting the rear spar is straight?
View attachment 111633
I am suggesting that the two-spar arrangement in your drawing copied here (above) will be more stable than the previous one with the red circles at the joints. The rib sections between the spars at A5/B1 , C1/B5 , C2/D1 , and D2 will need to be a stresses for the twisting loads between the spars as well as the "normal" aerodynamic loads.

The other drawing with two straight spars, and the beautiful oblique view of the completed wing with those straight spars, seems to be more complex, because the main spar upper and lower caps will not be parallel, and all the ribs will be a little wonky to build.

As I said, the art and science of quantifying the loads on all these parts is ten levels above my education, and you will need one of our genius level engineers to help with that.

The scheme I suggested, and your rendering of which is copied in this post, has worked successfully for decades in sub-scale model gliders. Before any one scoffs at this as a useful comparison, understand that 3 or 4 pound model gliders with a 9 foot span wing built this way routinely withstand an electric winch launch where the equivalent of 10G lifting load is put on these wings as the models are launched. If the wing structure was not fairly stable in torsion, this type of routine repeated launch would be breaking these gliders instantly... whether in twist or bending or both.
 

rotax618

Well-Known Member
Joined
Oct 31, 2005
Messages
1,351
Location
Evans Head Australia
Unless you have a CG problem that requires the centre of lift moved rearward and there is no other possible solution, there is no good case for a swept wing in an ultralight aircraft, either structurally or aerodynamically. You are opening a can of worms, apart from the structural problems, swept wings require more washout to delay the stall, that goes for a tapered wing where the RE at the tip is much lower than the root.
Consider the people who would like to build your design, a complex wing on a simple aeroplane is not what would encourage them to take on the project
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
Consider the people who would like to build your design, a complex wing on a simple aeroplane is not what would encourage them to take on the project
Hi Rotax618,
If they had to build a wing like this from scratch, I absolutely agree with you. I certainly wouldn't be keen. But when you consider any prospective builder gets:
  1. Every rib already cut
  2. A full set of rib jigs precision cut out of 10mm MDF to ensure everything lines up, like the screenshot below
  3. And then consider the sexiness of the wing.
Quite a selling point I think.

1623491595295.png
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
I am suggesting that the two-spar arrangement in your drawing copied here (above) will be more stable than the previous one with the red circles at the joints.
Good. That is my preference also. It is difficult to come to a decision regarding this, and I lean on the collective wisdom of this group for guidance. I say "guidance", not a slavish "yes-sir" attitude. But adding a second spar to my original design (and moving the main spar forward to the 20% chord position) allows me to mount the wing pivot point on the front side of the spar, which is right where it needs to be, and it can be fixed to the spar itself. Quite neat...

The scheme I suggested, and your rendering of which is copied in this post, has worked successfully for decades in sub-scale model gliders. Before any one scoffs at this as a useful comparison, understand that 3 or 4 pound model gliders with a 9 foot span wing built this way routinely withstand an electric winch launch where the equivalent of 10G lifting load is put on these wings as the models are launched. If the wing structure was not fairly stable in torsion, this type of routine repeated launch would be breaking these gliders instantly... whether in twist or bending or both.
I'm pleased to hear this. I've decreased the ribs from 30mm foam to 7mm plywood to minimise the identified weakness of having such wide gaps between the spar webs. Then by adding a second spar, I'm increasing the load bearing ability of the wing still further. Bring on the static load test!

Duncan
 

Vigilant1

Well-Known Member
Lifetime Supporter
Joined
Jan 24, 2011
Messages
6,800
Location
US
Bring on the static load test!
With the wing design you are contemplating, for the static load test it would seem particularly important to put the sandbags, etc in positions (chordwise and spanwise) that accurately reflect the way the wing will be loaded under high Gs. Just stacking the sandbags over the center of the chord never reflects the conditions of flight, but given your concept of spar loading, testing the wing in the most "realistically pessimistic" manner would be prudent.
 

Sockmonkey

Well-Known Member
Joined
Apr 24, 2014
Messages
2,274
Location
Flint, Mi, USA
Just my two cents, but you can get sexy with a sweet-looking engine cowling. As something non-structural, you can make it as swoopy as you like and give it a cool paint job.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,951
Location
Rocky Mountains
however. I got the width of the turtledeck wrong by 4mm. BUGGER. But I have lots of 3mm gaboon ply left over, so with a bit of careful measuring, I was able to re-CAD this part,
Am I seeing a bit of tunnel vision here?
CAD/CAM, especially non-parametric CAD, often lets us make our mistakes faster.
For a simple part such as this a tape measure, a pencil and a hand saw can get the part made - to fit - faster than a redo in CAD/CAM. Then once fit is assured the real world numbers can be transfered to CAD for future reproduction.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
a tape measure, a pencil and a hand saw can get the part made - to fit - faster than a redo in CAD/CAM. Then once fit is assured the real world numbers can be transfered to CAD for future reproduction.
Exactly what I did, except for the hand saw bit. Measured, saw why I'd made the mistake, corrected it in CAD, and ready to cut new (repeatable) parts tomorrow. No big deal. Except being annoyed that I made the mistake in the first place.

Not tunnel vision - this plane is made from CNC cut parts. If the parts prove to be wrong, I need to correct them and re-cut.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,655
Location
Brisbane, Australia
With the wing design you are contemplating, for the static load test it would seem particularly important to put the sandbags, etc in positions (chordwise and spanwise) that accurately reflect the way the wing will be loaded under high Gs.
That's the plan.

As far as weights are concerned: 4 G's = 300k x 4 (1200kg). Rear wing = 40% of this (480kg) so each wing half needs to bear 240Kg I'm really not expecting any surprises... 6G was my initial goal, but Fleas seldom/never see these sorts of G forces. But even at 6 G's, each wing half needs to bear 360kg
 
Last edited:
Top