Seeking expert opinions on my design approach

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
116
Location
Bellingen, NSW. Australia.
Hi, I'm new to posting on this site though I've spent many late night reading through the composite threads. I'd like to let you know my experience as it is relevant to a question I wish to raise. I am a multihull yacht designer and builder so I have plenty of experience with most areas of composites including carbon masts and spars. I want to design and build a composite aircraft and would like to hear from some of the more expert members of this forum whether my approach will work designing an aircraft.
When I design a new model of multihull yacht I draw it up in 3d CAD using Rhino software. I use my experience to determine core, laminate and reinforcement thickness and some simple engineering to calculated rig loads to determine laminates for carbon chain plates etc. When I have the entire boat modeled I send it off to a composite engineer who then does an finite element analysis and who then comes back to me with advise on where I can save weight or gain needed strength or stiffness.
This is basically the approach I would like to take to design my plane. I will draw it up using cores and layups that I think will work from my experience, from calculations and from information gleaned from existing designs. I will send it off to my composite engineer friend who can run it through an FEA.
I am particularly taken by the Millennium Master, Tarragon, Blakshape sort of plane so I'm thinking of buying a set of Asso x plans as a good design basis for designing a composite version.
What do you guys think?
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,460
Location
Marion, Ohio
I'm not a composite expert by any means (although I do have a Master's in Engineering), but it sounds like your method would work, at least as far as the physical design is concerned.

You still have to do the aerodynamic design. Will it fly? How stable is it? What specs are you going to build it to (LSA, GA, USA ultralight, Euro Ultralight, etc)? Dynamic and static loads?

I see a lot of people post pretty pictures on here, saying they are "designing" a plane, but little or no mention of the above mentioned factors.

[Edit] I just saw that you are in Oz, which would have other different requirements. I'm sure you are aware by now that there is a large contingent of your "mates" here on HBA that can give you advice. One comment I have seen several times is on the cost (shipping) and availability of materials in Australia, which can affect your decisions.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,376
Location
North Carolina
You'll find that with the weight requirements, large areas of small aircraft are designed to handling strength, not aero strength.
So, from your nick, I had hopes of the world's first pushme-pullyou aircraft!
 

BJC

Well-Known Member
HBA Supporter
Joined
Oct 7, 2013
Messages
10,801
Location
97FL, Florida, USA
You will need to develop the loads assuming an initial weight and initial performance parameters, then begin an iterative process to optomize weight, structure and performance.


BJC
 

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
5,990
Location
Australian
This answer is in question form, is there a 1.5 strength factor for composite compliance for many Governing bodies?
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,111
Location
CT, USA
You will need to do an aerodynamic analysis first, to determine the loads that you will need for the structural analysis. You will need to analyze several different loading conditions for different regimes of flight.

Dana
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
12,963
Location
Port Townsend WA

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
116
Location
Bellingen, NSW. Australia.
Thanks for the answers so far. I will be designing the aircraft in the Experimental category here in Oz which means I won't be too restricted to an all up weight as a lot of the sporty ultralights; although I do want to keep it light and efficient.

My question is less about the aircraft design itself but more about designing it in composites. Designing the aircraft looks pretty straight forward but it concerns me when I read Bilsky saying one must have a clear understanding of matrix algebra and calculus and read massive tomes on composites before designing a wing spar. It took me over half a century to realise I don't need to know everything so my approach is to get the design as close as possible using my own knowledge then hand it to a composite engineer for analysis. I will then build the plane and do load limit testing
.
I am wondering why I have not seen any mention of FEA in this forum. I realise FEA is only a tool and like all tools is only as good as the operator, however, while it is used extensively in marine composite design I haven't heard it mentioned here.
 

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
5,990
Location
Australian
I don't need to know everything so my approach is to get the design as close as possible using my own knowledge then hand it to a composite engineer for analysis.

I will then build the plane and do load limit testing
That's the one, that's the system that works for me, other than I work metal not that gooey smelly plastic stuff ;)



I am wondering why I have not seen any mention of FEA in this forum. .
I use it.

Love the Bellingen River, may end up living in Yamba yet (ex- Mornington and Brissy here).
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,738
Location
Rotterdam, Netherlands
My question is less about the aircraft design itself but more about designing it in composites. Designing the aircraft looks pretty straight forward but it concerns me when I read Bilsky saying one must have a clear understanding of matrix algebra and calculus and read massive tomes on composites before designing a wing spar. It took me over half a century to realise I don't need to know everything so my approach is to get the design as close as possible using my own knowledge then hand it to a composite engineer for analysis. I will then build the plane and do load limit testing
.
I am wondering why I have not seen any mention of FEA in this forum. I realise FEA is only a tool and like all tools is only as good as the operator, however, while it is used extensively in marine composite design I haven't heard it mentioned here.
The big difference between sailing/offshore and aviation is that in aviation we have to optimize it a lot more to get to reasonable weights. A boat, designed the way aircraft are would probably be less than half the weight they normally are. And a hell of a lot more expensive.

That mostly shows up in that you have to calculate loads on an aircraft way more accurate. There's a lot of load cases and most of them can be the critical one. Plenty of non-linear ones. Calculating the loads on the airframe is the majority of the work, not calculating the stresses.

Another major difference is because aircraft are to light but relatively large, almost every single structure will fail in buckling. That's considerably more complicated to do FEA on than relatively thick-walled ship structures.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,460
Location
Marion, Ohio
My question is less about the aircraft design itself but more about designing it in composites. Designing the aircraft looks pretty straight forward but it concerns me when I read Bilsky saying one must have a clear understanding of matrix algebra and calculus and read massive tomes on composites before designing a wing spar......

I am wondering why I have not seen any mention of FEA in this forum.....
But as Dana said, you have to do the aerodynamic analysis first, before you know the loads to design the composites to.

As far as FEA, there are some on here that use it, I think. You don't really need it to design an airplane, in metal or wood that is. Where it comes in is for optimizing a composite design. Although in my opinion, I would think that there could be a "simplified" composite design method. It may not be "optimum" (what ever you want that to mean), so long as it is "good enough" (that is, satisfies the load requirements with adequate safety factor). In fact, I remember coming across a NASA paper on that very topic, but don't have a link to it right now. All I recall is that they came up with a simplified design process that used unipolar (or was it bipolar) fabric in basic units. Then stack the units together to meet the requirements. Better not say any more, as I don't remember much.
 

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
116
Location
Bellingen, NSW. Australia.
The big difference between sailing/offshore and aviation is that in aviation we have to optimize it a lot more to get to reasonable weights. A boat, designed the way aircraft are would probably be less than half the weight they normally are. And a hell of a lot more expensive.

That mostly shows up in that you have to calculate loads on an aircraft way more accurate. There's a lot of load cases and most of them can be the critical one. Plenty of non-linear ones. Calculating the loads on the airframe is the majority of the work, not calculating the stresses.

Another major difference is because aircraft are to light but relatively large, almost every single structure will fail in buckling. That's considerably more complicated to do FEA on than relatively thick-walled ship structures.

I can see that, and the Americas cup catamarans are a fine example of how much money can be spent when working on the bleeding edge of composite technology. I've been told each foil is taking months to build at a cost of hundred of thousands of dollars. I'd imagine the engineering going into those alone would make aircraft design look very basic. My designs were very simple in comparison though our free standing carbon masts were an interesting challenge as they had to bend exactly to the leading edge of the sail shape.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,506
Location
Saline Michigan
Remember this first, last and continuously - WEIGHT IS THE ENEMY.

Your basic scheme can work. It may require a lot of iteration to get close to optimal design. Once you have your configuration figured out, get all of the shear/bending/torsion loads mapped out for wing and tail, then wing/tail reaction into the fuselage. Your FEA guy will need it all.

For aircraft primary structures, you really do not need FEA. You might use FEA on details, but for the fuselage, wing, tail, and control surfaces, you can size and confirm you are OK on failure criteria based upon simpler analysis, which allows short cycle iteration and close to optimum weights. For things like wing and tail connection to the fuselage, control surface connections, control systems, engine mounts and landing gear, FEA can be useful, but hand methods, automated via Excel work just fine.

You do need to have an adequate ability to check all of the laminates in your primary structures. Wings and tails carry huge bending, shear and quite a bit of torsion, and still have to be light.

So, if you do not want to bother with a pass through Jones or Tsai and Hahn and the time to work up some tools for iterating your design, your FEA had better be able to include specifying all of your lamina and providing failure criteria checks on each of said lamina. You may be asking a LOT of your analyst...

Some simplifications will present themselves:

The external skin has to be adequately sturdy to handling to survive build and finish and paint, then the slings and arrows of use, including strikes on the occasional starling and bumblebee and people who want to use your bird as a writing desk. In fiberglass that is generally about 21 oz/yd, and graphite about 12 oz/yd. In fiberglass we usually use 2 plies 7 oz UNI at +/- 45 degrees and then a third ply running the long way. In graphite, having square weave at +/- 45 is common. Use these in your first run through the structures, and you may find you are done with everything except spars/ bending structures. This is common in airplanes of the 160-200 mph cruise range, and will usually be overbuilt for aeroloads. The inside skin of the sandwich can generally be somewhat thinner except where it is subject to people and baggage. In vacuum bagged or vacuum infused structures, glass panels (both sides of foam cores) will be below 0.7 lb/ft^2, graphite below 0.4 lb/ft^2, this may sound low compared to your history;

Sandwich cores are generally 1/4 to 1/2", with some folks going down to 1/8" in places. I have seen thicker cores in main wing spar shear webs, but not anywhere else. Phenolic plate and solid lamination is common around hardpoints with bonded in metal for carrying contact with pins or bolts, and spreading loads into the surrounding parts;

Tapes for joining panels will be BID cloth on the bias of similar thickness to the panels being joined, with a minimum of 2 ply 8.8 oz/yd, 1" on each side of the join, wider as you add plies;

You may already know this from boat work, but bond lines and hard points will be large fraction of your weight in your primary structure. Design with bond line weight and hard point weight included. Optimizing your big panels but with a lot of joints and hard points sub-optimizes airplane structures. Anytime you can reduce part count and thus adhesive volume and number/size of hard points reduces weight. Think fewer bigger parts to keep overall weight down.

If you do intend to iterate your design through FEA, make sure that your FEA guy knows that you are likely to be changing the thickness of structures - use of tables for dimensioning of the models will be useful, particularly in spar caps, shear webs or in skinning of wings, tails, and control surfaces where spars are not included.

Billski
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
12,963
Location
Port Townsend WA
I am particularly taken by the Millennium Master, Tarragon, Blakshape sort of plane so I'm thinking of buying a set of Asso x plans as a good design basis for designing a composite version.
What do you guys think?
Redesigning the wood Asso X to composite would be some work. Looks like the Millenium Master is the composite version.
https://en.m.wikipedia.org/wiki/Millennium_Master

"Weight is the enemy"
I would consider a one piece composite D-cell wing with fabric aft of spar.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,738
Location
Rotterdam, Netherlands
I can see that, and the Americas cup catamarans are a fine example of how much money can be spent when working on the bleeding edge of composite technology. I've been told each foil is taking months to build at a cost of hundred of thousands of dollars. I'd imagine the engineering going into those alone would make aircraft design look very basic. My designs were very simple in comparison though our free standing carbon masts were an interesting challenge as they had to bend exactly to the leading edge of the sail shape.
And that is maybe the best way to explain the difference.

That hydrofoil is tweaked for optimal hydrodynamics and to be strong enough to cope with 3G or so. But structurally, while optimal it's not very complex; just a thick-walled composite structure.

For an aircraft wing however, we're looking at a couple dozen different load cases. Every single one can be critical. A much larger proportion of the engineering is put into load analysis compared to anything related to sailing. This post (http://www.homebuiltairplanes.com/forums/showthread.php?t=28221&p=397842&viewfull=1#post397842) and the posts before that discuss composite airframe structures. As you can see, the stress analysis isn't that complex, but it's the shear number of load cases that makes it considerably more involved.
 

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
5,990
Location
Australian
I would consider a one piece composite D-cell wing with fabric aft of spar.
Have seen a fiberglass wing have appropriate triangles, 4 leaving an 'X' cut from the upper and lower skins in between the ribs, then fabric applied. Saved a few pounds apparently.
 

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
116
Location
Bellingen, NSW. Australia.
For an aircraft wing however, we're looking at a couple dozen different load cases. Every single one can be critical. A much larger proportion of the engineering is put into load analysis compared to anything related to sailing.
We'll have to agree to differ on that one though I get the point you are making. Check out the Americas cup boats these days I think you would be interested in some of the engineering challenges. Boats need to be engineered around two liquids, one a thousand times denser than the other. The AC50s fly on foils in one medium while being driven by a 24 metre (74') wing mast in the other medium. They have a budget of over $6,000,000 each with a 15 man design team mostly borrowed from aerospace and Formula one. I think a fair proportion of the engineering would be put into load analysis. Of course these boats and all the round the world multis are extreme cases and you are right about boats vs planes in general.

Mark
 

Markproa

Well-Known Member
Joined
Feb 9, 2017
Messages
116
Location
Bellingen, NSW. Australia.
So, if you do not want to bother with a pass through Jones or Tsai and Hahn and the time to work up some tools for iterating your design, your FEA had better be able to include specifying all of your lamina and providing failure criteria checks on each of said lamina. You may be asking a LOT of your analyst..
I've just ordered Jones.

Mark
 
2
Top