General Design Questions for Project

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

birdus

Well-Known Member
Joined
Apr 3, 2012
Messages
185
Location
Tacoma, WA
I'm working on designing a 70% F4U Corsair made of carbon fiber. I asked a question in another thread regarding foam vs. honeycomb. The conversation drifted into more general design, so I'm moving that conversation here. Following is where the thread was heading.

But first, some basic ideas about the plane:

- all carbon fiber and foam
- Graphlite rods in the spar, but maybe steel or aluminum joining the outer spar sections to a central spar section
- no fabric control surfaces
- no folding wings
- no oil cooler inlets in the wings
- real, functioning F4U-like landing gear
- engine will be a Vedeneyev M14P

No, what I said was that your speed and panel size and facings combine to determine what core thickness you will need. Please do not oversize things. The panels I refer to are the sizes of the skin between other supports, like longerons, spars, and ribs. You have homework to do on how big the panels will be. You will have a rib at each end of the wing, defining fuel bays and equipment bays, maybe one or two as fuel tank baffles, and probably have one at each end of each bend. Main and drag spars will establish other edges. If you arbitrarily use 1" foam, and 3/8" would have done, you will have excess weight and you will have to extend your fuel tanks further out the the wings too.


Billski



I definitely want to take advantage of what carbon fiber offers and not oversize things.


I was thinking of making the fuselage in two halves, each of which would be foam sandwich, then joining the two halves with additional layers of carbon fiber at the seam. I don't know what sorts of longerons would be necessary or what they would look like with this kind of construction. Would they even be necessary?


For the wings, I was thinking of making four panels per wing: upper/lower for the inboard wing section, and upper/lower for the outer wing section (inner and outer delineated by the low point where the landing gear mount).


As far as fuel tank(s), I was thinking of a single tank in the fuselage for simplicity. I realize I would need a stronger spar for that. Wing tanks would increase the complexity a fair amount. Am I better off with a single tank in the fuselage and greater simplicity or a tank in each wing and greater complexity?

Thanks,
Jay
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,502
Location
NJ
As a designer by trade, (not aircraft, machines, mechanical product) I would say these decisions are for you to make. Design is personal, and all design is compromises. I personally do not like to design by committee. The devil is in the details and every thing you have mentioned is possible if properly executed. You need to decide what you want to do, make preliminary drawings/sketches do a preliminary stress and then do it again with changes until it is optimized to your satisfaction. Most here are not thinking very deeply about your project so it is doubtful you will get much help toward your goal.
 

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
11,043
Location
97FL, Florida, USA
I would think that you would set some target performance parameters. Weight empty (you mentioned 2200 pounds gross, IIRC), stall speed, cruise speed, designed load factors, stall speed, minimum RoC, etc.

A “real, functional F4U-like landing gear” will add weight, complexity and cost.

Composite and M14 make me think or the Radial Rocket as a good example (except for the bent wing) of what you are thinking of.


BJC
 

birdus

Well-Known Member
Joined
Apr 3, 2012
Messages
185
Location
Tacoma, WA
As a designer by trade, (not aircraft, machines, mechanical product) I would say these decisions are for you to make. Design is personal, and all design is compromises. I personally do not like to design by committee. The devil is in the details and every thing you have mentioned is possible if properly executed. You need to decide what you want to do, make preliminary drawings/sketches do a preliminary stress and then do it again with changes until it is optimized to your satisfaction. Most here are not thinking very deeply about your project so it is doubtful you will get much help toward your goal.
Thanks for the encouraging words. I appreciate it!

Jay
 

birdus

Well-Known Member
Joined
Apr 3, 2012
Messages
185
Location
Tacoma, WA
I would think that you would set some target performance parameters. Weight empty (you mentioned 2200 pounds gross, IIRC), stall speed, cruise speed, designed load factors, stall speed, minimum RoC, etc.

A “real, functional F4U-like landing gear” will add weight, complexity and cost.

Composite and M14 make me think or the Radial Rocket as a good example (except for the bent wing) of what you are thinking of.

BJC
Yeah, I mentioned the Radial Rocket in the other thread. I think I'll be lighter, but I imagine some similarities. I've also been filling out some spreadsheets created by engineers (available online) trying to flesh out some of these numbers. Guess I just need to keep at it.

I know the landing gear will be a challenge, but I think that would be the icing on the cake. Some of the components, I think I can make myself. Maybe I can get oleos from another GA plane. We'll see.

Thanks,
Jay
 

birdus

Well-Known Member
Joined
Apr 3, 2012
Messages
185
Location
Tacoma, WA
Have you seen http://corsair82.com/ ? Seems like a phone call to that builder could help influence your design.
I have! I've been through his site multiple times. I'm trying to glean what I can from his experience. It's loads of fun watching other people's builds. Mine will be quite a bit smaller than his, but I know there are still things I can pick up from him. And of course, he stopped quite short of completion.

Thanks,
Jay
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,684
Location
Saline Michigan
Jay,

One piece wing is lightest. Next best is two piece with main spars overlapping each other through the fuselage, pinned to each other and to the fuselage. The more joints, the more weight and fuss making everything work. Four piece? Make you crazy.

If I were doing this project, I would do a two piece wing and bend the spar. Bending Graphlite rods puts stresses into the rods, using up some of the load carrying capacity. The math is straightforward but can be fussy, but the strengths of carbon cloth laminates can also be iffy. I would probably use the smallest diameter rod offered and then check stresses in the rods at scale radii. If they are too high, then maybe you can use an even larger radii at the bottom bend without compromising the look too much...

Billski
 

birdus

Well-Known Member
Joined
Apr 3, 2012
Messages
185
Location
Tacoma, WA
Jay,

One piece wing is lightest. Next best is two piece with main spars overlapping each other through the fuselage, pinned to each other and to the fuselage. The more joints, the more weight and fuss making everything work. Four piece? Make you crazy.

If I were doing this project, I would do a two piece wing and bend the spar. Bending Graphlite rods puts stresses into the rods, using up some of the load carrying capacity. The math is straightforward but can be fussy, but the strengths of carbon cloth laminates can also be iffy. I would probably use the smallest diameter rod offered and then check stresses in the rods at scale radii. If they are too high, then maybe you can use an even larger radii at the bottom bend without compromising the look too much...

Billski
Hey, Bill.

I just posted a response to your comment about this in the other thread (since this specific issue came up there earlier). Bottom line, the radii are larger than I had thought. Jim agrees with you—that a one-piece spar is the way to go if possible. Based on some numbers he gave me, it looks like the initial stress will be well under 12%. So, I'm going to shoot for that. I'll just need to throw some more rods in there to make up for the initial stress (compared to starting with a straight wing). I suspect I may need thicker foam and/or more layers of wrap, too. The point being, some combination of those things will be necessary to make up for that "pre-stress." (essentially what you told me) I feel encouraged about going with a one-piece spar.

Thanks,
Jay
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,898
Location
Warren, VT USA
The pulltruded rod strategy might not work. But you can get carbon tow bundles in any size you want or you can use carbon UD fabric/tape and make it work with enough FOS.

The gear while tricky is a fun design project. The existing gear is a well documented design so you can get it done simpler than the original. The rotating element is just a simple rotating single axis gimbal at the top end.

One piece wing will work well. You can do hotwire molds for all the skins and do the spar as one or two pieces depending on space and then have a doubler or joiner in the middle buried in the fuselage where you have to space for it. CG for that small of a design will be tricky and a single tank might be a problem. One thing you can do is say mold tanks into the bottom skins of the inner wing panels with internal ribs so that they are structural. The ribs can be anti slosh baffles. You are only limited by your imagination, flight loads, and what you can figure out how to build tools for. That's the great part about composites, you just work from the outside in and start solving problems.

You haven't mentioned flaps. That is your real nemesis on a project like that. Take a look at the F4U flaps. Tricky. And they had folding wings for carrier ops and they all stayed hooked up in both folded and mission ready positions.
 
Last edited:

Aesquire

Well-Known Member
Joined
Jul 28, 2014
Messages
2,402
Location
Rochester, NY, USA
Consider the original.

The center section of the wing is a very complex structure that is part fuselage, and carried the wing, landing gear, and engine loads. ( plus complex ducting for ram air to the supercharger & oil cooler )

The landing gear and wing fold hinges/connection are all in that center section.

The prototype had fuel in the wings, but wanting both better range and armament, changed on the production version to guns in the outer wings and a simple big fuel tank in front of the cockpit, right on the CG. This also moved the cockpit back and hurt visibility, a problem for a fighter and carrier aircraft.

It seems to me that it would be "simpler" to design the wing spars straight in the outer wings panels, to plug into the center section as removable like most sailplanes, overlapping spare and pin joints. Or hinges like the original, but then you have hanger height issues.

And design the center section including landing gear mounts as one piece, curved rods and all, so the plane can sit on the gear with the wings removed. This also simplifies building, as you can be on the gear earlier in the build.

Building the wing as two separate curved parts that join at the center seems like the hard way for multiple reasons. That two piece wing works ok on a FW-190 replica, if not optimum, because the wing panels and spare are straight & under the fuselage, but then again, you're not up on the landing gear into much later in the build.

I could be all wrong with my reasoning. The original was a series of cascading compromises, and took a long time to develop. It also had a very long production run and service life.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,684
Location
Saline Michigan
Birdus,

I had skipped the whole issue of fuselage construction, longerons, etc in your initial post, but want to address them briefly.

The fuselage, like each wing, is a beam. Big weight at the front, wing lift a little ways aft, other weights scattered along it, and the tail download aft. Sum vertical forces to zero by adjusting wing lift, sum moments to zero by adjusting tail downforce, and you know the shear and moment diagram for the fuselage. Unlike the wing, which is a slender beam, fuselages tend to be pretty deep and max bending moments tend to be smaller. The result is most fuselages do not end up with pieces that look like beams, but you still can analyze them that way. In welded truss fuselages, truss theory works fine, but in monocoque fuselages, it does get more complicated... They are usually a big thin wall tube, which means buckling and crippling can find you before raw strength does. In sheet metal, you end up with several longerons and a bunch of frames/bulkheads and stiffeners all designed to keep the skin from going into elastic collapse. Chapter 8 Of Peery and Azar talk about this as do the other airframe design texts.

Since you are building in composites, sandwich skin makes the skin panels way more stiff in bending, and can eliminate the need for most of the frames and almost all of the longerons. The firewall will still be needed, so we keep that. Most of us still go ahead with a frame or bulkhead at the instrument panel, at the seat back, at the tail and one in between seatback and tail. After that, the elliptical shape of the fuselage cross section tends to be self supporting and may take care of itself. We must cut out the tube for a canopy - when you try to figure out how to make up for the lack of the upper third of the tube there, you beef up the cockpit edges, viola, longerons from in front of the cockpit through to somewhere behind the cockpit. If a fuselage fuel tank is used like in the original F4U's and you make it removable (a good idea), that area will likely have the upper skin detachable too, or might even be the top of the fuel tank, so maybe you will have longerons from the firewall back through the aft end of the cockpit cut out.

For my side-by-side arrangement, I ended up with a wide relatively uncurved floor, and put in a tunnel from firewall to the aft baggage bay bulkhead, and that broke the flat floor into several small enough panels that stresses and deflections were acceptably small. The edge of the tub is reinforced heavily from the firewall past the cockpit opening and tapers to nothing going aft of the cockpit. So, I have two longerons 4" apart on the center of the floor and two more along the cockpit sole. Frames only where I mentioned before.

Beginning to define the way you build it... Many of us build a boat from firewall to tailpost, fill it with wing carrythrough, other structures and systems, tail, controls, plumbing and wiring, and everything else, and then add the turtle deck, forward deck, and canopy. Eventually, the whole thing sits on its gear and looks like an airplane.

I would advise against trying to build it in left and right halves for a couple reasons. First, you will basically have to build and install all of the stuff that goes inside and attach it to one side of the fuselage or the other, and then figure out how to get inside and attach that stuff to the other sidewall at assembly. Now some certified birds are/were built this way in factories with precision jigs for assembly, etc. But I would advise against it for a one-off done by a homebuilder. Much more straightforward and easy to visualize and assemble in a boat.

None of this is to say that you would not build the fuselage tub in pieces and put them together, just that you may want to leave the top open and able to stand freely by itself for most of the construction.

Billski
 

Aesquire

Well-Known Member
Joined
Jul 28, 2014
Messages
2,402
Location
Rochester, NY, USA
In pictures of the original, you often see tape in a big square on top of the fuselage, in front of the cockpit. That's tape to seal the seam around the top of the drop in fuel tank.

Sure looks like a nice place for an access panel to get to the instruments.

The center section is a big triple "U'.... Actually "VUV" in shallow shape as seen from in front.

The fuselage is an open boat with bulkheads at the firewall, front & rear of cockpit, and the fuselage behind the cockpit was longeron & bulkheads, with the skin spot welded on, made from stamped metal sheets. The internals were installed through the open framework before skinning.

Changing to composite construction, the open tub design as suggested above, makes a lot of sense.

If you choose to make the rear fuselage a one piece elliptical tube, just be sure you can reach/crawl inside to put in the few things you Have to put back there. Like control runs & brackets. ( if you're large, like me, the open tub, removable turtle deck makes more sense, even if you need to mold in a few longerons to make up the loss in elegant tube structure. )
 
Top