Orcrist | A 2.5m VTPR Glider
Part II: Building, cutting and skinning the tail and wings.
Readers may want to familiarise themselves with the first part of this series before continuing with this follow up article. Also, click any picture in any RCSD story for a higher resolution image. — Ed.
Following on from Part I where I focused mainly on the fuselage, this month’s update is focused on the tail area and the wings.
The Tail
The vertical stabiliser itself is pretty mundane: a plank of hard/stiff balsa cut and butt jointed to align the grain along some of the stress lines, with an infill to make it triangular/normal tail shaped. A vertical spruce post makes up the final part giving the future rudder something hard to be hinged from and adding a little stiffness to the whole ensemble.
Where it gets a little more complicated is the elevator. The plans call for an all moving tailplane (AMT) and for the control mechanism to be hidden in the tail and fuselage. An AMT in a moulded glider is straightforward: the tail tends to be hollow and there’s plenty of room for the bell crank (the 90 degree control horn the centre of which things pivot round, made from 1.6mm PCB board) and the associated strengthening and linkages. With the Orcrist we face a few trip hazards:
- The tail is quite narrow
- It’s made of balsa
- It’s solid
My solution was to use the Dremel router to carve out a rebate where the bell crank would be located. A thin shim was then fashioned to fit over the bell crank which helps to keep things aligned. Lastly, a spruce strip was rebated vertically into the back of the tail to offer a more substantial brace against the AMT wobbling over time as the hinge wore down the balsa. This was further stiffened with a section of brass tube to act as a bearing and with additional spruce fairings of the same section as the tail surfaces on either side of the brass tube. I hate it when AMTs wobble!
The final step was to skin the whole vertical stabiliser in 0.6mm ply, adding stiffness and robustness, while hiding all the AMT linkages. Obviously the linkages and brass sleeve were fitted and commissioned before the skin was glued in place!
The Rudder
With a near enough completed tail, my attention turned to the rudder. By this stage I was conscious that the tail was a little heavy. With hindsight, I could have added a few lightening holes to the vertical stab with no ill effects. In an attempt at weight saving, the rudder was therefore built up from (mostly) scrap as opposed to using a single solid slab of balsa. At the bottom, inboard end, additional wood was retained to give the control horn a good surface to be glued too.
As with the vertical stabiliser, a spruce post was attached to ensure a good strong anchor point for hinging. Once shaped, the whole rudder was skinned in the same 0.6mm ply and the spruce post bevelled to allow movement once the cyanoacrylate (CA) hinges were in place.
All Moving Tail
The AMT mechanism required some imagination, some woodworking and a dash of engineering. The AMT surfaces were a cake walk by comparison! A single 3" wide plank of hard / stiff balsa with only two cuts and two glued butt joints yielded the two complete surfaces. To these, a pair of routed rebates were made, not quite all the way through, to accept a pair of brass tubes. With these epoxied in place, the leading edge (LE) and trailing edges (TE) were given a sanded profile; rounded to the LE, tapered to the TE.
Top Tip Packing tape makes a great mask to avoid sanding areas you want to preserve.
Lastly, the two AMT surfaces were skinned in 25g/m² (0.75 oz/ft2) glass/epoxy to add a little chord-wise stability (the grain ran span-wise) and sanded smooth ready for a primer/filler base layer.
AMT Finishing Touches
The next step was to finalise the fairings either side of the vertical stabiliser that added to the rigidity of the AMT. A secondary benefit being it adds clearance between the AMT surfaces and the fuselage should I ever modify things to extend the AMT throws. (Google rc glider mad stab to see some of the mods people do!)
Getting the Horn
The last piece of work on the tail was to make a control horn to match the rudder and the control run fitted in Part I.
There could potentially be some quite large forces applied in flight, so I was keen to ensure there was a large surface for gluing, but at the same time, I wanted to minimise the change in angle of the control rod as it exited the fuselage and in turn, had to restrict the height of the horn from the surface. You can see below the evolution from mock-up to finished horn, again made from 1.6mm PCB board cut and shaped with the bandsaw and filed by hand.
Elevator and Rudder Servo Fit Out
I have a fuselage with an AMT and rudder fitted, with control rods and linkages in place — what should I do now?! Fit some servos and enjoy my fruits with a bench fly, of course!
With nose weight in mind, a servo tray of ply was fashioned with the servos towards the front of the available space, ensuring both that the control rods for the elevator and rudder were long enough and that the servo wires could reach where I intended the battery and receiver to live.
There was a secondary benefit to adding the servo tray too: it provided an excellent way to stiffen a section of the very thin ply I’d used for the fuselage sides. Remember from Part I where I realised I’d used the wrong ply?
A Diversion — The Canopy
Taking a break from the work on the tail, I turned my attention to the canopy. The plan left things a bit open to interpretation (so I thought initially) and I ended up making it up as I went: lots of balsa stock and then shape with lots of sanding back to what I thought ‘looks right’. With hindsight, I could have just used the same flat sheet and triangle section, but the result would have been a little more square than the rounded section mine ended up as. I’m happy. What do you think?
With the shape there, the final step was to separate it from the fuselage. An accurate cut through the two formers and it was liberated! From there, I added thin ply facing, doubled internally for structural rigidity (this is a well handled part) tin the same way as the other ply sides were in Part I.
Now that I had a structurally sound, properly shaped canopy with a layer of 25g/m² glass and epoxy. I added a few layers of spray filler with a quick wet sand between each layer with 240 grit paper which resulted in a nice smooth surface. I couldn’t wait to give it a top coat of French Blue (seems appropriate) to see what the end result might look like.
Attaching the canopy for flight use was a head scratcher for me, I have seen complicated spring clip systems. I’ve seen clips and slide arrangements too, but none seemed to lend themselves to this shape and arrangement. In the end, I used three magnets, spaced evenly from nose to tail of the canopy giving a firm but hand-releasable fit. Alignment is taken care of with small doublers attached to the formers. According to the magnets packaging, 6kg of force is needed to release the three pairs — should be secure in most of the situations it will end up in!
The Wings
The wings of the Orcrist are a semi-symmetrical blend of SC17 at the root to a SC17s section at the tip. The plan included templates for a hot wire cut foam wing core.
I’ve cut a few cores over the years but I’m no expert, particularly when it comes to wide spans and differing chord at the root and tip. To help with the cutting, I built a swing arm/drop arm mechanism a few years back to help keep things correctly ordered (with a shorter tip chord, the rate at which each end of the hot wire moves is not equal, the shorter chord tip needs to move slower than the root). There are many videos and forum post dedicated to swing arm hotwire cutters. Google foam core swing arm hot wire | foam core gravity hot wire cutter | foam core feather cut hot wire or similar and you will quickly get the idea.
To use the swing arm cutter, you need a template for each end of the core for the hot wire to follow. As the name implies, the wire is hot, so the material these templates are made of needs to be heat proof (or at least resistant from 200C through to about 400C). Plastics are great to work with and easy to shape, but they’d be toast in this use case. My preferred material for this is aluminium sheet, but I have also used ply and have read about people using melamine sheet.
The idea is you have either an outline of the section and the wire traces top and bottom in two passes, or you have two templates for each end, top and bottom and make two passes. I opt for the latter as I find it makes alignment easier to have a lead-in/lead-out to each cut and that’s not possible with a single template.
Hot Wire Foam Templates
I printed two sets of plan templates for each end and glued them onto the aluminium sheet. The bandsaw makes light work of the rough shaping resulting in an ‘upper’ and ‘lower’ template for each end. All the while, ensuring the bottom of each upper/lower is constant.
The roughly shaped templates are then fine tuned by hand with various files and eventually 80 and then 240 grit emery paper. The aim here is an accurate representation of the aerofoil section, but also a smooth edge for the wire to transition along; any rough parts or snag traps will cause issues in the cut and may ruin the whole core. Very minor ripples can be sanded out of the cores, but if the shape is wrong it’s toast!
Foam Selection
My initial attempts at cutting cores for the Orcrist were made using some EPS polystyrene that I had in the workshop. Alas, it was poor quality and the resulting core was very flexible and very low density. A few Google searches later and I located a supplier of grey XPS foam offered as a replacement for the blue and pink foams many of us have used for years. (N.B. You cannot get blue and pink foam in the UK any more.)
First Skin
A pair of cores were cut, remembering to make two different cores, not two the same (yes, I have been there, done that in the past!). The first order of business then was to add a lower skin. This helps with keeping the delicate trailing edge intact and provides a reference surface when cutting spar and joiner channels in the foam later.
The skins are a one piece wood veneer similar to Obechi. They are coated with epoxy, a 10mm x 0.8mm carbon strip, a 25g/m² glass layer (wetted with the epoxy) and a 100g/m² rectangle around where the joiner will go. This lower skin is then sandwiched between the core’s outer carcass, the core and the top carcass, all weighted down with what ever I could find while the epoxy cured overnight.
Spars
With the cores skinned on one side I began the process of laying out the spars referencing the plans, drawing out with sharpie and a long ruler. Once happy with the layout, I used a combination of craft knife and razor saw to carefully score the foam to make a channel twice the width of the spar. The foam cuts easily, but I used several passes, rather than trying to make a single cut to full depth. I needed to be careful NOT to cut through the lower skin. With luck (skill? ha!) I found the carbon strip at the bottom…perfect!
The next step was to remove the material between the two scored lines. If you have made any foamboard aircraft (looking at you, FliteTest), you will recognise the process of removing foam here! I found a small flat bladed screw driver worked well to eat away at 3–5mm sections and left a nice neat rebate ready for the ply spar.
The spar was made by laminating two strips of 1/16" ply and then gluing in the channel with slow epoxy. The lower skin provides a firm base when weighting the spar , ensuring a good bond line to the carbon strip that becomes a lower spar cap. By using play for the spar the vertically aligned grain of the centre ply layer acts a a shear web too.
Once cured (again, an overnight wait!), the top surface of the spar was trimmed flush to the core’s surface with a block plane. (I find a good, sharp block plane is a very therapeutic thing to use!) The plane made light work of the spar and once again, I used packing tape to mask the delicate foam as I got close to the surface. You can hear when you’re approaching the last few skims: the sound gets a little ‘crunchy’ as you start to plane through some of the epoxy that’s escaped when the spar was slotted in.
A light sand with some 80 grit paper gives a slightly roughened spar top and a slight rebate either side for the carbon strip that’s the top spar cap.
This is where adding the lower skin first helps keep things true: if the gap you cut was slightly too wide or not quite wide enough, you may end up altering the curve of the lower wing section. The veneer helps avoid this, especially when combined with the carcass and weights during the cure stage.
Joiner and Pockets
The plans show a 10mm joiner but on a recent local hobby shop visit, all I could find was an 8mm solid carbon rod and a 10mm carbon tube, with a ~1.6mm wall thickness. The fit is good, but I may still run a line of CA along the rod near the end of the build.
With my joiner and tube at hand, I set about marking up the joiner sockets in the wing cores. They sit snug to the spar and have a ‘false spar’ (not sure of the engineering term — it’s a second spar) to form a box for the joiner socket. The area of foam is cut and excavated out in the same way as the spar rebate, leaving the lower skin quite exposed and delicate. Once done on both wings, I used the extending dining table in the house to lay out the full wingspan (wing upside down to allow a very slight dihedral) end-to-end, so as to be able to confirm the joiner’s alignment and the correct wing sweep of 25mm measured at the tips.
The more observant of you will note that one wing is covered on top, the other underneath — yes, I did get it wrong and cover the top of one core — doh! Luckily its no biggy. The construction is the same except for the servo pocket reinforcement (see Second Skin section below) so I’ll just put it down to distraction!
With the joiner and wings aligned, a couple of small foam offcuts and a dab of Clear Gorilla Glue held the rod in place while I poured slow epoxy with 10% carbon millings added into the trench with the tube in it. Once cured, the whole thing was flipped over and the process repeated for the newly exposed trench in the other wing (did I mention I skinned the wrong surface of a wing!?)
Once cured, the rod can be cut and the roots trimmed back to flush with the rest of the core. The surface of the new joiner pockets can be planed back flush (if proud) or scuffed ready for the skin bonding (if recessed).
Second Skin
Time to add the second veneer, to the top sides of the cores (or the bottom if you messed up the first time like me!)
The order of the sandwich is (with additional 100g (3oz) glass over the joiner/spar area top and bottom):
Getting all the layers ready before hand helps to avoid forgetting a layer and given potentially short pot life of some epoxy, it makes sense to be organised. below you can see the dry mock-up, with everything cut to size and ready to go, in order.
With the epoxy mixed, time to wet out and make the sandwich. As before, I wetted the veneer first, followed by the foam core’s layup, which was followed by another wetting of both halves, before the last step of adding the veneer and clamping under the core carcass.
Weight
The target all up weight (AUW) weight is in the region of 1700g — 2000g (60oz —70oz ). So far as photographed below, its at 1680g. That excludes the receiver battery, four wing servos, any balance weight that’s needed and coverings of course. I think I should still just creep in under the upper target!
N.B. There’s also around 5–10cm (~4") of wing span to shorten to fit the plans, that should shave a little too.
Joiner to Fuselage
We’ve got this far… a fuselage, tail plane, a pair of wings with a joiner… would be a shame not to have a quick table fly wouldn't it?! Just one problem! there’s nowhere for the joiner to go in the fuselage. 10 mins with some very careful measuring and drilling — the joiner fits both the fuselage and also makes a nice ‘pop’ sound when removed from the wing pockets.
So, without further ado, here’s the table fly!
See you next month for the next instalment of my Orcrist project. And by all means, if you have any questions, please do not hesitate to leave your questions in the Response section below and I’ll do my best to answer them.
©2022 Marc Panton
Resources
- Orcrist | A 2.5m VTPR Glider — Part I: Picking the design, making plans and getting the build underway.
- What a Tool! Servo Templates for Dremel Rotary Tools — My previous RCSD article where I explain the finer points of routing with Dremel tools.
- Dark Grey Styrofoam — The foam is listed as ‘styrofoam’ with a density of 33kg/m³ (~2 lbs/ft³). They do other colours, but not in the ‘large size’ sheets — however, it’s all the same density.
- Koto Wood Veneer — This veneer is very similar to Obechi.
- 25g/m² Ultra Lightweight Close Weave Glass Cloth, 950mm Wide — I’ve had both glass and epoxy from this supplier — friendly and quick service. Their epoxy hardener has a slight blue tint which is helpful when mixing which isn’t noticeable once cured.
- EL2 Epoxy Laminating Resin — I’m using this laminating resin with the ‘fast’ hardener, pot life 15–20 minutes in my shed at 15C. Cured in ~10 hours (ie. overnight). The cloth samples are handy for odd jobs and repairs where you don’t need a meter of material, such as the servo pocket reinforcements.
- My LHS: Addlestone Model Centre — A proper model shop with materials, kits, and parts. Don’t forget to support your LHS (local hobby shop)!
- Orcrist 2.5m on RCGroups — The RCGroups thread that proved to be the source of so much valuable information.
All images by the author unless otherwise noted. Read the next article in this issue, return to the previous article in this issue or go to the table of contents. A PDF version of this article, or the entire issue, is available upon request.
