Any engineers here??

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

Jul 27, 2004
Gulfport, MS
I have a design problem that I need help with. The airplane I am working on is a 2 place, high performance composite amphibian that looks like a cross between a sea ray and a lancair. For the most part everything has been thoroughly thought out and researched. The one thing that eludes me is how thick to make the fuse. Can anyone give me a reasonable method to come to this conclusion?? The cozy looks like it is about 1 3/8" thick (it probibly has a 1" core). Some aircraft appear to only have a 1/4" thick solid lay-up. (8-10 layers maybe??) Now, these are observations made from online photos, not in person. Being a composites tech for Northrop Grumman, I have the nessicary experience to build such an aircraft. The problem is that I am not an engineer, and lack the education to come to a reasonable conclusion on certain design aspects. They hand me blueprints, and say "here, build this". Behind those blueprints is an engineer that has already figured what is needed.

I have, however, designed model aircraft for 20 years. (mind you I am in my late 20's) The latest of my designs are giant-scale TOC birds that are all composite. Although they did not go to production, the flight tests were phenominal. Am I wrong in believing that I can use the same methods used on these 30-40% scale airplanes to design a full-scale aircraft?? Since I do not hold the sheepskin of an aeronautical engineer, I will read every book I can get my hands on about aircraft design/physics/aerodynamics.

I can design the wing, and test it for strength, as well as design the fuse. But how do I compute the required strength of the fuse?? What do I need to figure how thick it should be?? I need to know things like how many layers, core thickness, core type, stich, woven?? If someone could enlighten me on this, I would really appreciate it.

Thank you,


Super Moderator
Staff member
Log Member
Oct 18, 2003
Saline Michigan
This question is more complicated than you know...

Most of us are building land planes, and the fuselage has to stand several things:

Shear Load and bending moments from wing pitching moment, tail forces, and internal masses at max g times our factor of safety (2.0);

Aero loads from moving air outside the fuselage vs stationary air inside (in my Cherokee, this is really obvious when looking at the upper skins of the wing - You can see lines of rivets and the skin bulges out between them);

Landing loads, which are smaller than the flight loads, but are enough different to warrant a check;

Local loads around inspection ports, baggage or equipment doors, and the cockpit;

Handling by flightline personnel, fly-in morons (someone will want to use it as a desk), passengers, etc;

And, if you are concerned about meeting the spirit of the crashworthiness rules, the cockpit has to be good for a one time loading of 19 g's.

Now, the monkey-see, monkey-do approach can work pretty well for most of this. Most everything out there has something like this for fuselage skins:

Foam or honeycomb core;
Outside skin of three plies 7 oz UNI, one the long way of the fuselage, one each at +45 and -45 degrees;
Inside skin of two plies 7 oz UNI at + and - 45 degrees.

Now comes the tricky part. The core has to exist to resist local buckling and the pressure difference between the moving air outside and the stationary air inside. The greater the curvature (smaller the local radius) and the thicker the core, the farther apart the frames and longerons can be...

What's that, frames and longerons? If you have a large panel with membrane loads and you have it supported at its edges, it might have to be really thick, or it might need some frames... On metal birds, the spacings are pretty small and we have a bunch of lines of rivets holding skins to structures. On composite fuselages, frames at fuselage stations tend to be pretty limited, firewall forms one, instrument panel, main and drag spars form partials, the seatback bulkhead, baggage bay, and we usually put more under the forward and aft edges of either the horizontal or vertical tail.

We also tend to put in a couple of longerons. Yeah, think about it, we reinforce the edge of the canopy opening, and usually have a tunnel on the floor that houses control runs and wiring, etc.

Anyway, these things serve to break the fuselage skin into several plates or shells. If your frame and longeron positions are fixed, and you know your air loads, there are published functions for computing deflection and stress in the plates. Make the cores thick enough and the standard composite layers will work, and you will have the lightest structure possible with your materials, or make the cores thin and add composite laminations, but they get heavy in a hurry.

Stresses in plates is covered in Roark's Formulas for Stress and Strain, but for composites, you will have to back into the bending stiffness of your skins... and I recommend Tsai and Hahn's book on Mechanics of Composites to figure out that stuff. Avoid Martin Hollmann and Marshall for structural calcs... They have just enough of it right to be dangerous.

Now, when you get to seaplanes, the parts that will be in contact with the water have to stand its being a hydroplane, and everything has to stand everything that a land plane has to, plus deal with water landing loads... Which takes me back to Monkey-see, Monkey-do as an engineering practice. There is considerable experience out there on hull design, but I do not have any of it, and you will have to find that on your own... There are several outfits with kits at the major fly-ins that you can look at, and then there are other folk's plans, and the folks who build hydroplanes, speedboats, etc.



Mar 2, 2003
Western Washington
I think that if we boil down all the issues that need to be covered in the design process of an airplane you may come to realize that there is a rather substantial amount of work in all that the effort encompases to do this right. Billski indicates the process of "monkey see - monkey do". This can work well for you in a general sense however in looking at any composite structure it is rather difficult to establish the laminate schedule of said structure without actually cutting a chunk out and analyzing the layers. It tends to be rare for the manufacturer to tell you what they used and of course the sales reps that are at most shows will most likely not have a clue.

Each structure is unique and thus it may not necessarily be a good idea to try to copy someone else's work since their design requirements and assumptions may be different from yours. This becomes especially critical in the design process of amphibious structures as they need to withstand loads that have nothing to do with flight. This includes everything from handling loads as Billski alluded to above, to water loads, as well as locallized impact loads that your structure must withstand when your hull contacts a submerged immovable object.

All this then has to be coupled with the hydronamic design, which encompases everything from attitude and trim issues for take-off and landing (wing incidence and hull design must be closely related for proper take-off performance), to planing and spray behavior, to flight perfomrance (you don't want the hull to create any more drag than necessary). All this then also has to deal with buyoancy requirements, as well as those that address flotation and post-damage safety considerations.

The experience required to do all this in a coordinated manner requires the designer to have a background in not only the structural end of things but also in hydrodynamics, as well as aerodynamics. It is possible to do this in a self-taught manner however, as you may or may not have already found out, it often happens that when folks try to do this themsleves without the proper background and expereince, they often find that the more they read, the more they come to understand what they don't know.

Regarding scale models, this has been covered in more detail in other areas of this discussion board so you may want to use the search feature of this board to read those threads. In short though, the answer to your question is no. Scaling issues are such that the behavior of a model is not rally a good indicator of what the full scale bird will behave like. The model will only provide you with a general idea of behavior but the answers are qualatative in nature, not quantatative. This is true in aerodynamics as well as in structures. Yes, some organizations do use scale models for initial flight tests but the results tend to be rather limited and the organizations doing those test do have the necessary background and tools to enable then to make some form of correlation of the data. For the average person though, even one with an engineering background, it is unlikely that the information gained will be uselful in anything but a very "general idea" sense.

If you are serious about doing this right, get help.
Jul 27, 2004
Gulfport, MS
Thank you all for your speedy response. I will seek assistance with this project. I intend to educate myself as much as possible. This project, even with it's complexity, does not shake my willingness to succeed in it. The knowledge is out there, it is just a matter of obtaining it. (ok, understanding, and properly exicuting it is in there somewhere too.) There have been successfull ametur designers before me that have done it, and more will come after. I will draw from the experience of the group, and be sure to know what I am getting into before I start slinging resin!!:ban:

Thanks again,