The AFB (Amazing FleaBike)

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
3,689
While these suggestions are well-meaning, the proper way to design a fuselage truss is not by Internet committee. YouTube and other basic resources are full of the principles of truss design--two-force members, elimination of zero-force members, aligning members to intersect at single-point joints, etc. Better to discuss those concepts than redesign Duncan's fuselage for him--especially since none of us has determined any of the loads on it. The depth of the truss beneath the seat may or may not be a problem depending on the loads imposed on it.

Duncan, I look forward to more progress. I like the concept and your persistence.
 

TLAR

Well-Known Member
Joined
Sep 29, 2020
Messages
223
While these suggestions are well-meaning, the proper way to design a fuselage truss is not by Internet committee. YouTube and other basic resources are full of the principles of truss design--two-force members, elimination of zero-force members, aligning members to intersect at single-point joints, etc. Better to discuss those concepts than redesign Duncan's fuselage for him--especially since none of us has determined any of the loads on it. The depth of the truss beneath the seat may or may not be a problem depending on the loads imposed on it.

Duncan, I look forward to more progress. I like the concept and your persistence.

I don’t know if forum design is bad or good. It certainly lets a lot of eyes examine and comment.
In the end the responsibilities still lay at the feet of the designer.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,583
Location
Rocky Mountains
Still bad... the entire bending loads come down to the height between the back of the seat and the lower longeron,
"Bad" is relative. This area is just one of those parts of an airplane that is hard to optimize because we have to accommodate the pilot. Fully triangulating the cockpit entry to eliminate indeterminate structure is really hard without using a door with a latch that can carry the structural loads. That brings problems of it's own.

I agree that plywood isn't the best material to use for structure under the seat. If there is enough to safely carry the loads and the designer has accepted the trade for weight/time/cost then his job is done.

My 'feel" for this part of his structure it that it is going to be good enough for the torsion loads for a 2 axis Flea. Bending loads are at the weakest part of the structure and only good math is going to justify this part of his design.

@ RTFM
Perhaps the most practical and developed iteration of your Strongback of almost a decade ago? Maybe a massive set of laminated bows under the seat? Would make a nice shape to sit on?

edit:
In the end the responsibilities still lay at the feet of the designer.
And why we should use all of the resources available to us, such as a second set of eyes on the project, as is being done here ..... and math using conservative and verifiable material properties.
 
Last edited:

Victor Bravo

Well-Known Member
HBA Supporter
Joined
Jul 30, 2014
Messages
8,622
Location
KWHP, Los Angeles CA, USA
I also am not advocating or suggesting "design by internet committee" as a final answer to anything. But our enormous and unwieldy group design and discussion threads on VP-21,s and Rangers and flying motorcycles have been a delight and a real treasure to many of us. Like everyone else, I am in awe of the amount of creativity and interesting ideas that have been brought to light with these.

As someone with a very limited and un-refined understanding of how structures work, watching these problems get discussed and commented on (by people who do and do not know what they are talking about) is a very educational exercise for me.

OF COURSE this is still not any substitute for a proper engineering analysis. Anyone who thinks it is a substitute immediately gets a loud and clear warning by this entire group for the most part. But I don't want to lose out on my becoming more educated because the trained engineers here have been frightened away.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
14,613
Location
Port Townsend WA
Those plywood routed truss strut ends appears a new invention?
Are traditional plywood gussets glued on also?
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,752
Location
CT, USA
Duncan, several things:

First, the deep cutout for the seat. This puts all of the fuselage bending moment in a very small space, between the two members indicated by the blue arrows. That's a very small truss for the loads it needs to take. Now, I realize you can't do this with the airbike concept, but for decent strength you'd need a member where I put the green line.

1604625722607.png

Since you can't do that, you need a massive beam running fore and aft under the seat. Not a truss, but something more substantial, with a shear web... not unlike a wing spar. I'll get back to this.

Next, your choice of plywood. Its strength is much less than a straight grained solid piece of wood of the same dimensions. Only the outer plies, and perhaps one inner ply, have the grain running the long way. The other plies are crosswise, contributing weight for almost no tensile strength. Plywood's entire reason for being is to make large, flat, stiff panels. It's great for gussets where solid members come together, and for shear webs between the outer members of a built up wing spar... or the beam under your seat.

Consider making all the truss elements from solid lumber, with thin plywood gussets at the joints. With somewhat heavier elements under the seat, and a complete plywood skin both sides and top and bottom forming a shear web in that area, tapering off as the loads are transferred back to the open truss, it might work. But you still need to do a proper load analysis before cutting any more wood.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
14,613
Location
Port Townsend WA
I think some ultralight fuselage truss longerons are made with plywood. Lots of models do use plywood routed out.
Not sure why not route a full sheet like a model instead of gluing separate pieces?
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,416
Location
Brisbane, Australia
Hi guys,
It's both interesting and informative to read your comments, and very nice to think my little design has warranted comment. Thank you.

I cut the main members today, and laid them out loosely on the template. Everything seems to fit really well. I managed to destroy one piece (forgot to add holding tabs, and the piece moved. The router ploughed through 19mm of plywood for about 200mm before I was able to shut it off. Nice to know that my cnc machine can do 19mm depth of cuts without complaining too much!

Here's the first side laid out. Note (1) two missing parts (turtle deck braces) - no idea how I missed this, (2) the new part suggested by VB (3) a redundant part - since I've beefed up the in-built gusset anyway.

Some comments based on your feedback:
  1. The 19mm ply I'm using has five plies running in the surface grain direction, and four at 90 deg
  2. The in-built gusset are a new idea, so no additional gussets are required
  3. The seat area will have a 3mm doubler inside - so - a 3mm skin, the frame, and a 3mm doubler
First side layout.jpg
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,416
Location
Brisbane, Australia
I'm extremely interested to see how much this lot weighs. I'll be bonding these pieces to the underlying 1.5mm plywood template with West System epoxy. Rather than clamping, I'll be:
  1. Placing a 10mm strip of tape at intervals along each piece
  2. Using a roller to spread the epoxy over both template and pieces
  3. Removing the tape, and adding a blob of superglue in the un-epoxied surfaces
  4. Placing weights on the piece to allow the superglue to set, and to act as glue clamps
I've tried this technique on test pieces, and it works great. The pieces are held firmly together by the quick-setting superglue while the epoxy takes its time to set.

Regards,
Duncan
PS
I'll photograph everything and post it here.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,416
Location
Brisbane, Australia
The wing ribs are cut from 20mm Dow foam. Since this is a Flea, I'll be using the Fraser FrasF5 airfoil. Foam cuts REALLY nicely in the router. I've cut six ribs so far (more of a test of the method than anything else). The wing is made of three panels: a central section, and two outer sections. The outer panels fold on hinges (SS yacht hatch hinges) reducing the folded width to only 2.5m (legal max width for trailering in Aus). More importantly, this means the wings can be constructed in 1.25m lengths (the two central panels are simply bonded together).

I have a jig for the 1.25m panels all set up. The spars are also 1.25m long. All very easy to do. I haven't decided yet, but I will probably make the central section spar 2.5m long, and bond in the two inner panels.

Duncan
PS each foam rib weighs 71g
 

Protech Racing

Well-Known Member
Joined
Jul 10, 2020
Messages
411
Might consider extending the lower seating strut to the upper engine mount for a stronger/straighter load path for the engine thrust.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
14,613
Location
Port Townsend WA
Ok, I see how it assembles after reading post 1 again. Looks like a neat method. But I might suggest 3mm ply for the "template" also to be the same thickness as the 3mm outer sheet. Actually, don't really need the template if built on the precut outer ply side. Then apply small gussets to inside.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,416
Location
Brisbane, Australia
Hi,
The airframe pieces are already 19mm thick. That's SOLID stuff. I think the weight/strength ratio would be excessive with double 3mm ply.
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,752
Location
CT, USA
The in-built gusset are a new idea, so no additional gussets are required
That's not quite correct. Even with your "built in gusset", the gluing area isn't as much as a gusset glued on the side. But more importantly, gussets on the side load the glue joints in shear, which is what you want for a good glue joint. Your method loads the joints in tension, which is not good.
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,583
Location
Rocky Mountains
Other than being cut from 3/4" ply, not solid wood, how is this method any different than a traditional stick, filler and ply joint?
One glue line being replaced (blue lines) with uncut wood looks to me to be the only other variation. One less glue surface = one less poor glue joint to make.
gussets.jpg

As for the ply being used in compression and tension just think of it as bunch of little sheets of wood being used in compression along the grain line with the cross pieces being very heavy sandwich material preventing buckling of the vertical grain. It is heavy and material inefficient compared to other methods but if it gives him the strength he need it does seem to have some advantages when it comes time to assemble the parts.
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
14,613
Location
Port Townsend WA
It's like the corner "blocking" is "built in" on the cross pieces ends. Cool idea I have not seen before. But still needs adequate gussets on both sides of each end, in my opinion.
It isn't a "built in gusset". It is a "built in blocking".
 

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
3,689
Depending on the orientation of two members, the alternating grain direction of the plies may preclude long-grain to long-grain interface at the glue joint, so in that sense, it's not necessarily providing additional strength as compared to traditional blocking. However, the 'built-in blocking' is providing additional face gluing area for the skin, acting as a large gusset. I agree with BB that an inner plywood gusset would be a good idea to ensure that the joint is loaded symmetrically.

Some testing seems warranted to understand how the axial stresses in the member are transferred to the outer ply.
 
Top