Some thoughts on the production of smaller gliders...

[FrankS]

I beams are better than other similar weight sections for withstanding bending loads, it's all to do with the Second Moment of Area, https://mechguru.com/machine-design/sec ... -sections/

Where lightweight structures are required, the web will often be lightened by having "holes" cut out of it, so if you wanted to take your weight savings to extremes you could do the same.

[chris williams]

That's very interesting Frank, and gives me more confidence in my guesswork!

Now, the next step is to smooth off the bottom of the wing and apply the lower sheeting. (I'm abbreviating the process for simplicity). An un-sheeted wing has the capability to wander off the original planform, once off the board, so it's a good idea to clamp a straight-edge to the aileron spar before applying the sheeting. Once the PVA has dried, the wing will then be locked in the correct planform shape. Now for the pivotal moment: Adding the top sheeting will lock the 'D' section into its final shape, so the wing is rested on a minimum of four balsa jigging supports and weighed down until the glue has dried. Interestingly, on the Petrel wing both the top and bottom sheeting were added in one piece, the top sheeting needing a little wetting up to conform to the compound curve at the gull-join... This left the wing without any joins at the gull interface at all...

[Peter Balcombe]

Perhaps I will use the split spar/ I beam approach on my next larger built-up wing design then

[chris williams]

A quick look at servo mounting... There is no hatch as such, I use sticky-backed vinyl sprayed a similar colour to the wing. In the interests of efficiency, which increases in importance the smaller the model becomes, the ailerons are top hinged and the lower gap sealed with Draughting Mylar, which, to my surprise you can still buy...

[chris williams]

Just out of interest, the Petrel wing, all servo'ed up and filmed, weighs just over 18ozs

[chris williams]

Buoyed by the success of the smaller models so far, I wondered what could be done at a larger scale. 6mm sq spruce is a son-of-a-female-dog to persuade to bend in any significant way, but 5mm sq spruce seems a fair compromise between that and the 3mm sq stuff. I decided it would be interesting to see what a lightweight 1/4 scale Minimoa would be like to fly. This is a real test of a one-piece gull wing, as the spar has not only to bend to accommodate the gull, but have a significant sweepback, too...

[Nigel Argall]

Just to add to the debate (and as always, forgive me if this is ‘old hat’) but there is more than one way to bend wood. Tradition and Cliff Charlesworth’s book dictates that we steam wood but that is not how instrument makers (well – guitar makers anyway) do it. Their method is to soak the wood so it is really wet and then apply heat. My understanding with steaming is that you soften the wood so it is pliable. With the instrument method, you apply a lot of heat to one side of the wet wood so it dries and shrinks – bending the wood – I enclose a pic of an acoustic guitar I made in about 1983 – never did it again as it was such a faff.
I also enclose some pix of a bit of ¼ x 1/8 spruce I bent today. I soaked it for an hour then put it over a paint stripper. This is not a great idea s you are mixing electricity and water but also because the metal on the paint stripper is really too thin. A guitar maker would use something like a piece of steel scaffolding pipe with a blowtorch up the end – you really want a lot of hot metal so the wet wood doesn’t immediately cool it.
Needless to say there is a knack – you have to slide and rock the wood on the hot surface to get a bend – being an organic product it is also impossible to get a perfect curve but probably good enough. There is also risk of scorching the wood but that is not too critical for us as you can always sand it off. Is a problem if you make a guitar with burn marks... Good new is that it is pretty quick - just a couple of minutes to effect the bend.
What I don’t know is whether this reduces the overall strength of the wood – I assume it must (but so must steaming?) but again, probably not critical.
In theory (!) you can bend thick wood but it does need to be soaked – for guitar sides I left them in the bath overnight.
The skid on my Grunau Baby was done this way – it is a solid piece of ½ by 1/8 inch spruce. Once done, you might like to leave the wood clamped just to make sure it doesn’t move but fundamentally, it will stay bent, not spring back. I kind of over-did the example here just to show what is possible.

[chris williams]

You can also do some pretty heroic (and quick!) bending of 0.8mm ply by wetting it and attacking it with a domestic iron.

[Jilles]

I just steamed the spars of my Short Bros Nimbus gull wing spars, they are 1/2"x1/4' I agree with Chris that they are hard to bend with that cross section but with perseverance it will happen. The four items I had to bend took me 2 hours total.
In case it was not possible to bend Plan B was to laminate two layers of 1/2"x1.8" together in the correct shape to form a 1/2"x1/4 cross section. This would probably not require steaming at all.
If you find a 1/4" square spruce section hard to bend why not using a 1/2"x1/8". this has the same cross section area ( same strength), is easy to bend and gives a stronger spar as well in the end. For the minimoa you could taper past the gull radius the 1/2" width to say 1/4" width at the tip.

In my experience on larger scale gliders one should not go to light, too light will not make them fly better. As the saying goes the weight of a glider is its engine, so less weight less engine power !!!. If you want a minimum sink rate yes built them light but they will not go anywhere fast.
My designs are in general a bit heavier than average But I have not had a complaint to any of them on that subject.
It is more important to finish a model that has the minimum amount of drag possible. The glide angle of a glider is the ratio between flying weight and drag. so the smoother the surface the better the glide angle. Increasing the weight will not change the glide angle it just achieves the same glide angle at a higher speed.

[chris williams]

Due to the more severe angle of the gull join and the added complication of the sweepback, the D box sheeting was applied in three parts. The gull join sheeting spans three ribs and the compound curves are quite easily taken up by the balsa sheet. This works so well, I have used the same system on the new Flamingo