Jonathan M

I like that! Today drilled and poked the snake through in more or less the right place for the throttle servo and fashioned its mount, then got the saddles and bearers for the fuselage servos made up. The AAA NiMH pack will go in the space in-between, and the switch/charge-port and Rx thereabouts too... hoping this 'rearward-loading' will help keep the CG from coming out too far forward.

Wow!

That's great Matt, helpful pictures, thanks! I think that, with the bottom sheeting providing extra stiffness, 1/16" sheer webs should be fine. I suspect the centre-section will be strong enough as designed (once one has worked one's way past all the inking mistakes on the drawing!), plus the two 1/8" u/c planks. The u/c will be attached using saddle-clamps both forward and aft. I'm planning on a pair of 1/8" riblets in the very centre to hold a 1/4" dowel, and use a M5 nylon bolt (or two?) at the rear.

That looks really great Matt. I reckon it flies a treat on that setup. Out of interest, what changes from the drawing did you make to the wing? I'm planning on sheeting the bottom forward of the spars rather than just the top, and adding 1/16" sheer webs, both of which should make up for using (hard) balsa spars instead of replacing them with spruce.

That's much the same cost as here - in fact cheaper than the best-known brand but the same cost as the stuff I now use. Nothing wrong with only 5% nitro - I fly with that all year round.

I wouldn't bother replying to this thread - it looks like an AI generated user or the suchlike!!

Thanks for the encouragement chaps! @Learner To be honest if the tank ever needs withdrawing (it's the sealed SLEC type so I hope not!) then it wouldn't be a major op to whip the engine off, clear the silicone sealant and sever the pipes if need be, then pushing a screwdriver through the oval should be easy enough. The curved 1/64" ply decking needs to go on shortly so I can develop the mid-part of the fuselage and cockpit, but I'll leave gluing in the bottom of the tank-bay until the very last moment before covering. Talking of the SLEC tank, the moulded join between the opaque white and blue front stands quite proud all round, which is why I set about the tank with a Permagrit file to smooth the burrs, and slightly chamfer the ends and corners as well, for snag-free movement. I realised, as I settled into bed with a novel late last night, that I'd forgotten to prepare the snake outer for the throttle, but it'll still be easy enough to drill through...!

As it makes sense to work from the firewall (captive T-nuts for mount already in place) aft, the next item to wrap my head around was the fuel tank. Found just the right thickness of slightly squishy plastic foam to hold the tank between the sides, snug but not too tight so that it can be withdrawn aft (if needed for future maintenance) through the square hole in F2, and at the correct height for the carb etc. Then cobbled together top and bottom foam mattresses, each with support slats for a cosy night's sleep, while another piece of the same foam forms a forward-stop a quarter inch from the firewall. I was originally going to suss some sort of pull-string to withdraw the tank aft (e.g. a zip-tie with the tail as the piece to pull on) but then realised that the whole caboodle can in fact be pushed out with a screwdriver from forward! The bottom of the tank bay will be filled with a piece of 1/8" grain sideways glued in on the bearers already in place; Dereck's drawing shows a removable bottom, but my system avoids this and keeps the whole bay insulated against fuel residue ingress; and the oval hole will be filled with semi-permanent silicone sealant once the fuel tubing is all finally in place.

Time to go 3D... I first tried setting up the fuselage sides and formers in my SLEC jig, but didn't feel confident using this... too many variables all at once! So (following Dereck's breezy instructions from 1991) decided to break the exercise down into two parts, laid one side down flat and epoxied the formers in place, then epoxied the other side down on top... checking that everything remained properly square and perfectly parallel right down to the tail end!

'With' or 'without' WOT for long period?

No, those words relate to my recent thread on running in an OS46AX (hence reference to smaller prop size) but I believe the principle would be exactly the same...

Absolutely. Always sort out the mechanicals first. Just exercised a set of four new HS85MGs for my current build and one HS81 for its throttle on the servo tester, an hour each. Starting with a fully charged 4.8v 2000mAh NiMh, put back 700mAh on subsequent recharge, so each servo on no load drew 140mAh each hour.

Sounds like budget servos in PNF models aren't always fit for purpose! The only ones of that sort that I have are those in my Riot which are fine. All my other servos fitted to self-build or self-fitout models are quality Hitec, MKS and KST types, even the more affordable Bluebird and EMAX servos seem rock solid.

Thanks for the replies. So it's all just to do with testing for potential failure, and the term 'burning-in' (which suggests some sort of wearing-in like new leather shoes or running-in a new IC engine) is a misnomer. PS: my Tx (old-school Taranis on OpenTX) also has function, but frankly it is SO much easier to just use my separate servo-tester on the new set of servos before installing in the model.

So it's just for testing really, no other purpose?

What's the rationale behind doing this, how long to do this for (time or number of cycles)... and does anyone actually do so?

Exactly the same in model thermal soaring. Various aren't permitted in comps, so it's best to leave them switched off and forget relying on them altogether, especially as one quickly learns to read what air the model is in by observing changes in its behaviour. Entering an area of strongly rising air the model visibly speeds up and becomes more lively, often also observed initially by the tail rising first (the fulcrum being the wing), with the opposite effects occurring when exiting an area of rising air: tail drops first, model becomes relatively more sluggish, etc. Model thermal soarers exploit their variable camber to much more efficiently achieve the same speed differences: the wing is reflexed slightly (ailerons and flaps up say 1-2mm) to speed through and away from sink, or given extra camber (say 3-5mm down on flaps, or similar on ailerons if no flaps present) to slow right down in thermals, in both cases with suitable elevator compensation. This is where modern radios really come into their own. Agree. But, once geometry is understood and even applied (e.g. offset control horns), electronics allow fine-grain adjustment. Plus, referencing thermal soarers as above, not only can the wing be cambered or reflexed, but the amount of differential increased for Thermal mode (3:1) compared to Cruise Mode's (2:1) and to Speed Mode's already slightly reflexed wing there is none at all (1:1).

Cheers Geoff... it's just pure fear that concentrates the mind! Yes the maths is straightforward, but with a quick visual sketch I'm less likely to make a mistake! 😉

Yes, and half of it is nonsense as well!

John and Mark, all well-substantiated points, and I defer to your much vaster experience and theoretical underpinning. I don't claim to know the answer (and maybe I'm on a sticky wicket!?) but just exploring possible reasons why slightly flatter turns seem to produce a more workable ascent in weaker thermals compared to more normal angles of bank. I'm obviously not advocating a completely flat skid (which is hopelessly inefficient), more considering the difference between, say, a modest 30° bank angle and a tighter 45° one. Agree, but the point I was making is this: if one removed the factor altogether of the object in the air being a glider producing lift from it's wing to mostly overcome its mass, and also removed questions of drag, efficiency and the whole balance of mechanical components as per John's diagram etc, and considered instead a flat plane, say a lightweight plank, able to hold itself horizontally stable parallel to the ground. In a column of vertically rising air this plank will ascend twice as fast (or descend at half the rate) compared to an identical plank that is self-held at an angle of 45° to the ground - because the first plank is presenting it's entire horizontal surface to the vertical rise of air, whereas the second one's horizontal component is halved. Clearly the above scenario is a bit extreme and entirely artificial, but perhaps in model scale thermal soaring (airspeed, model weight) there's a zone where, in some conditions, the loss of aerodynamic efficiency and therefore poorer glide ratio is at least made up for - and sometimes exceeded - by presenting the wing at a flatter and therefore more optimal angle to the volume of rising air? ???

The last bit of today's fun (while waiting for fuselage sides to dry) was deciding the amount of right-thrust and working out the engine-mount offset. There is none shown on the plan and I suspect Dereck was in the same camp when designing as the late Peter Miller - who never used any right or down as far as I know - but I prefer to build it in. So, how much then? My Gangster 63 Lite has 2° right and no down (the original heavy version specified 3° right and 1° down; I tried 1° right but it wasn't enough so I increased it and had to fiddle the cowl so it didn't look crank!), while the Smooth Operator currently being built by David Hayward has 2.5° right and 1° down. The Amelia seems to have 0-0 wing incidence and I'm not troubled by adding any down-thrust at this stage, which can be tweaked later if needed with washers without changing the appearance of the nose as such. But I've decided on 2° right from the outset, and a quick sketch on the plan translated that to a 2.5mm engine-mount offset. Finally, had a dig through my collection of foam cadged from various forms of packaging.... think I'll be able to cobble something together for Mr Blue.

And got basic fuselage sides glued up (with rock-hard 3/16" longerons), the second one directly over the first so all my mistakes are identically replicated! Tomorrow I'll bond the 1/32" doublers on with contact adhesive... then all the fun of fitting formers and turning it into a 3D object.

Today successfully tackled a job I was nervous about before: hacksawing the delicate canopy frame former, then sanding back to the line mostly with the Dremel (Henry vacuum nozzle to suck the dust), finishing off with Permagrit sticks.

Mark, I'm no expert or theoretician but think it might simply be down to this: the upward force of a thermal on a surface (the wing) that is perpendicular (flat) to that force, plus the 100% vertical component of lift generated by the unbanked wing (however compromised) are together probably more significant than the sink associated with a draggy inefficient skidding around.

Just to illustrate my post above, although this isn't an aileron model at all (rudder and elevator only with polyhedral wing), it is nonetheless very responsive and demonstrates clearly and close-up the tradeoff between bank angle and thermal turning efficiency.