This distortion of the camber line has niggled at me for some time, so during the writing of this article I called up the section in Compufoil, my new computer airfoil-plotting programme, and printed off HQ 3.5/12 thickened to 17%. (This programme allows such a thickening whilst maintaining the camber, or at least, that's my assumption.) Then I scanned the section at its original thickness and thickened it as previously described in the scanner. On a section of approximately 8" in length I could see no visible difference when one section was laid over the other, although at a larger scale you may have been able to detect something. Given the inexact nature of wooden construction, I seriously doubt that there is any measurable difference, and, given that the plan was for the section to thin to 12% at the gull join anyway, I really don't think there's too much to worry about here.

The article in the last issue was written quite some time ago and languished in the vaults for many months, since then I have discovered Compufoil. Now, for those with a PC and the desire to design models, a programme like compufoil is an absolute must when you consider the benefits to be found therein. Look at the current and traditional alternative, the Sandwich Method for making wing ribs, whereby a pile of balsa blanks are squashed between two templates and carved and sanded into submission in a great cloud of dust. Imagine being able to design your wing on your monitor, to establish all the important parameters such as the size and positioning of the spars, the thickness of the sheeting material, the positioning of the aileron drag spar and so on. Needless to say you can instantly call up your section of choice from a large library of sections, you can modify it, change the thickness, the camber, or both, plus a pile of other thing too numerous to mention. Now for the good bit...you can then print out all the ribs for you wing onto A4 sticky label paper, cut them out and stick paper to balsa, why, you can cut this little lot out with so little dust, the wife will hardly notice that you're not polishing the French windows.

For me there's only one catch, the programme is wing-specific, and it seems to me that there enormous scope here for a fuselage plotting programme, after all, the fuz is far more complicated than the fairly geometrically simple wings. Then of course, maybe there already is such a programme, in which case why doesn't somebody whisper in my shell like? (An e-mail on this subject to Compufoil-Meister Eric Sanders resulted in him sending me a file for a mathematically described circle. It seems that this can be manipulated within the programme to enable the lofting of ellipses as formers, so we'll have to see.)

Don't think that the process is entirely without hassle mind you...anyone who has used any computer programme will know what I mean. You will need to become familiar with this or any such application, and there will undoubtedly be frustrations along the way. Clubmate John Kerley bent my arm and persuaded me to plot the ribs for his Slingsby Prefect a little while back, and it was not all plane sailing. The problems involved a lack of knowledge of and practice with the programme, but it was interesting to say the least to see the possibilities that were thrown up. John decided to forgo the usual cap strips and have the LE sheeting cut into the front of the ribs. This we were able to achieve after a few false starts, and after having seen the built-up wings in the flesh I am impressed all over again with the programme's possibilities.

CONSTRUCTION BASICS

So, you've drawn up the basic shape of your glider, you've decided on the size of the model, and, hopefully, you have some idea of the projected all up weight. Starting with the most interesting bit, the fuselage (and you only have to build one of 'em) how do you get from a two-dimensional drawing to a three-dimensional object.

SLAB- SIDED CONSTRUCTION

The author with his Slingsby T-46

There's a lot to be said for choosing a subject with a flat-sided fuselage, especially if you're new to the own-design process. (We're talking here about such models as the Slingsby T31, the Tutor, the Grunau baby and such like.) The mission-profile here is a relatively straightforward one...simply build the two side frames flat on the plan, and then fix them to a couple of formers around the centre-section area, draw in the nose and tail with suitable formers to get the top-view curves, and add turtle-decking and stringers/sheeting as necessary. Where, you might ask, is the catch? Well, the one thing that most usually strays from the straight and narrow is the top-view alignment of the two sides, not to put too fine a point on it, it's tragically easy for your fuselage to resemble the shape of a banana if you're not too careful. There are ways to avoid this scenario however, and you don't need to be a brain surgeon to apply them. One way is to start by drawing a straight line on your building board. Draw centre lines vertically through all of your formers. Now, when in the first stages of glueing the two sides together you can simply lay the whole structure over the line and ensure that the centres of the formers correctly follow it. This construction process has been in use since Aeromodelling began, and although relatively simple, gives the builder the same sense of deep satisfaction that comes with the more complicated forms.

STEEL FRAME STRUCTURES

A full size Schleicher Ka-7 steel fuselage

In between the early wooden gliders and their modern glass counterparts, there is a whole range of sailplanes whose fuselages are made from welded metal tubing. Some of these are slab sided, but many are of a more complex shape that does not lend itself very well to the previous construction method. Some examples are: K18, K8, Bergefalke 2 etc. Although I have seen some beautifully crafted models actually made from metal tubing, most of these types of models are rendered using a wooden framework to represent the full-size. So, how does one assemble this disparate arrangement of stringers and spars? The answer is to use a jig, and although its construction may seem to add a daunting extra burden to the building process, there isn't actually that much extra work.

The principle is a simple one: from a wooden base you glue in place supports that mirror the bottoms of the formers in your fuselage. The formers sit in these supports and are held vertically with an arrangement of cross-supports which themselves are fixed to two horizontal rails, which run fore and aft each side of the fuselage. The drawing up of the supports is easy; you simply extend your former centre line down to the baseline when you are drawing up the formers, and draw the supports at the same time. Once you have the formers set up in the jig, you start to add the longerons, and it's quite fascinating to see a complex structure take shape before your very eyes. The fuz stays in the jig to the point at which it becomes safely rigid, then you can remove it to gain easier access.

Once again, keeping the fuselage straight is an issue; a line drawn through the centre of the jig base, lines drawn vertically through the formers, and, to be absolutely sure a thread of cotton or electrical wire run from the front to the rear of the jig on top of the fuselage will ensure that things do not become derailed and add to the stress that is such a feature of modern living.

One further point involves those longerons that show through the fabric surfaces. If the full-size used round-section tube it follows that the part of the tube on which the fabric rests will show its shape to the external gaze. So, it would be best where possible for the wooden longerons to have a consistently smooth round surface too, and you can achieve this by using half-round hardwood dowel from the mouldings section in your local DIY store, and glueing it in the appropriate places over your square longerons. At the same time you will also need to remember to scallop out your formers in the spaces between the longerons to prevent them showing through the fabric.

COMPOUND CURVATURES

There is no doubt that the best looking, and hardest to achieve in fuselage shapes, are those that have the smooth, rounded contours that are associated with designers' efforts at streamlining. Most of the wooden gliders from the '50's are built like this, gliders like the Oly 2B, the later models in the Slingsby range and the K6, to name but a few. In reality, a lot of it was faked...ply was wrapped flat around a series of fuselage frames, the final shape of which cunningly suggested a smooth curve... well, why should we be surprised, it's extremely difficult to get plywood to bend in more than one dimension. Attempts were made to get a true compound curvature by steaming the wood in a press,

The pod-and-boom Bowlus Baby Albatross being a case in point, which was sold very successfully as a kit with its pod halves pre-manufactured in this fashion.

Traditionally, to build this type of fuselage for a model, it is built in two halves. Imagine, if you will, sawing a completed fuselage in half through a line drawn vertically through its length down the middle. If you were to lay a series of keels flat on the board, glue in the half formers and a few stringers until the half shell was reasonably rigid, then remove it from the board and repeat the process for the other side then you have in a nutshell the method by which I personally prefer to build such a fuselage. There are, however, many variations. Some people prefer to sheet the entire half shell whilst still on the board, then build the other side separately and join the two sides afterwards. There is also the issue of strength; if you sheet the fuselage in a series of flat panels as per the full-size, the only longitudinal strength available to the fuselage lies in the number of stringers that are designed in, and I have seen models of this nature consistently crack a joint after a heavy landing. My method is to sheet the rear of the fuselage in ply that runs as far as possible in one length from the wing LE position to the tail. The front is sheeted with planks of 1/16" ply, and it takes a pretty heavy landing to stress that lot!

STRENGTH VERSUS WEIGHT

This is the designer's true test of skill, to find the best balance between a structure that can take all the slings and arrows of outrageous gravity, and yet still be light enough to lift off the ground without the need for a huge leather belt around the waist to hold the vital organs in place. Let's start with that most crucial of design decisions the wing joiner. I'm not going to talk about round steel joiners here, as I have very little experience of them, (although they work very well in the foam-winged models I have flown) but instead of the almost universally accepted flat steel bar that has served us so well. Say you expect your

The strutted T-46 after finishing

model to weigh in the region of twelve pounds or so, then you should be able to get away with using only one. I'll immediately qualify this by pointing out that if you intend to whizz your model about at high speed in strong winds, then you might need to reconsider. The furthest I have stretched this was with my seventeen pound Bocian which had only one joiner and wasn't too bad in this respect. On the other hand, if I fly my thirteen-pound Bergefalke in a strong wing, the wings do tend to flap a little bit. All of my 3.5 scale model which are twenty pounds or over are fitted with two joiners, the only exception being the Condor which has a three-piece wing, each outer panel having its own single joiner.

If your model is going to have a strutted wing, then it makes sense to have the struts working for a living rather than just bits of heavy decoration: this allows you to save weight at the centre section where the wings need only to be braced fore and aft. But hang on though, what about rigging? A true-to-scale system can be awfully fiddly to rig so how about a compromise? On my T21 I utilised a short steel wing joiner with bolt-on struts. This allows me to slide the wings in place where they will sit unsupported whilst it is then relatively east to bolt on the struts. The struts themselves can be made from a piano wire core with modified electrical connectors soldered at each end to cope with the loads under tension, and with spruce and balsa fairings to make up the streamlined shape. (In fact, on the T21 Gelutong sheet and carbon fibre rovings were glassed together to make the strut strong enough to cope with any compression loads that might result from an embarrassingly heavy landing.)

The joiners have to apportion the flight loads to the spars, so what should they be made of? Most of my earlier 1/4 scale models had 1/4" (6mm) spruce spars and seemed entirely adequate. These days I prefer to be a bit more elegant and use tapered spars for my larger machines, usually 1/2" by 1/4" at the root, tapering to 1/4 sq. out at the tip. Either way, when you choose your wood, eye it up carefully to ensure that it has no knots or splits that might later on prove to be a weak point. It is important whatever the size of the model to lock the spars together with strips of webbing to achieve torsional as well as lateral strength, and for me, 1/32 ply is the ideal material. How do you taper spars, I hear you ask? First of all you need a good straight edge as long as the spar itself. Then you need to affix the spar to your building board against the straight-edge to ensure that it is perfectly straight. Now mark out the reduction in thickness at one end and with yet another straight edge, draw a line from one end to the other. The difficult bit is now over...all that's left is to clamp the wood into a Workmate or something similar and attack it with a razor plane followed by a long sanding block with some 40 grit attached. Ten to fifteen minutes per spar should see the job done.

Sometimes the question is asked about wood selection, balsa wood that is. This is a matter that goes back to the earliest days of Aeromodelling when free-flight models had to be light to fly and the quest for lightweight structures was uppermost in everyone's mind. Conventional wisdom tells us, for instance, that 'C' (quarter) grain balsa is best for ribs, the reason being that because the grain runs in more than one direction it is better able to resist splitting along the grain. After the construction of a long run of large scale wooden sailplanes, my experience is that it is better to select wood for strength than for lightness, although it is a matter of common sense to keep the structures at the rear end as light as possible. A built up quarter scale wing covered in hard balsa sheet for the 'D' section is still pretty damn light for it's size, and this sort of balsa selection stops your pudgy digits from going through the sheeting when you lift the model up hurriedly to avoid that angry bull. When it comes to filling in gaps in structures with 1/2" balsa sheet, such as at the wingtips, or tailplane centre-section, you can afford to err on the side of lightness, and this of course applies to any part of the structure where strength is not the deciding factor.

When trying to set the balance between strength and weight, there is one universal rule that will help to guide you. The smaller the model, the more you have to move the balance towards weight saving, the larger the model the less it matters. Of, course, the larger model hits the ground with more of a thump, so you can see the conundrum..

The author's Minimoa	in her elements

WING RETENTION SYSTEMS

There are basically two ways you can choose to keep the wings fixed to the fuselage; rigid attachment, and flexible attachment.

Rigid attachment, as it implies, involves the use of bolts and screws, or maybe metal pins as the situation demands. The advantage of such a system is that the wings are fixed with undeniable reliability, the downside being that with the onset of the rather sudden forces associated with an unplanned impact with the ground, the bolts and screws and pins turn out to be rather stronger than the structure to which they are attached. One way around this is to use 'weak bolts' i.e. plastic bolts or partially sawn through metal ones, the obvious problem being the job of calculating the flight forces that they must overcome, and the catastrophic arrival forces to which they must surrender.

The most reliable form of flexible attachment involves the use that dear old modeller's friend, the humble rubber band. Two or three hooks on the roots of the wing, suitably aligned holes in the fuselage and you have a system that some people believe Icarus used to retain his crude wood -and -feather wings when he flew too close to the sun. (If cyano had been available then, he made still be alive today...well, you know what I mean) The big advantage of any flexible system is that it will absorb a lot of the forces involved in a gravity -based mishap, thus saving damage to the main structure. The downside being that in the sudden acceleration of say for instance, a winch launch, the wings may come out part-way under the stress, making this system unsuitable for plug and socket connections for the wing servos.

Some people, noticeably those that operate glass ships, have come with some ingenious systems utilising strong steel springs within the fuselage to combine the best of both worlds. The important thing to remember is the springs should allow the wings to be tensioned against each other, rather than fixed to the fuselage sides, as this could lead under extreme circumstances to the fuselage being ripped asunder.

Strutted glider pose a different set of problems: although the struts are simple to fabricate strongly enough to take the normal flight loads, once again in a heavy landing they can bow or snap under the sudden compression loads. Also, in normal i.e. not inverted flight, the forces operate to push the wings into each other, whilst in that heavy landing the forces are doing their best to separate them. A there usually isn't much room in the centre section to use rubber bands, a different method must apply, and on my T21 that other reliable standby, wing-tape came to the rescue. Tape provides quite a reasonable compromise between rigidity and flexibility, making it very suitable for plug/socket connection systems, which make for such quick and hassle-free rigging. Here the problem is that this tape is only available in black or white, eminently suitable for colour schemes where the wing roots are white (I've not seen a black sailplane yet!) but you're stuffed if any other colour is involved. By the way, if you plan on taping your wing to your built-up wing root fairing, you'd better ensure that the two fairings are interconnected in some way, as experience has shown that tape is strong enough to rip off the entire fairing. It's not a foolproof system, of course, what is? The other day I threw the Condor off a hill in a non-existent breeze. The resultant lack of lift caused to her hit the ground with a thump and then bounce out into the void where flying speed was quickly attained. Unfortunately, the thump had caused the tape holding the port outer wing panel to split, and the connectors became disconnected. This left the machine with one aileron inoperable and an airbrake semi-deployed, leading to a tense and untimely trip to the bottom.