Atom Autogyro

This page last updated: 4 June 2017.

On Saturday 21 January 2017 I went to the London Model Engineering Show at Alexandra Palace.  Having not been for many years I found that I was in heaven.  I really mean it, I was smiling the entire time and it gave me ideas for several Things I Must Do.  One of these, after watching the British Model Flying Association demonstrating flights in a very small area of the show indeed, was to build a model autogyro.  I had watched a documentary about these machines many years ago and thought they were fascinating.  Flying my Piper J3 Cub RC model hadn't really worked for me: staring into the sky, with watering eyes, at a dot in the far distance that I cared deeply about (after 6 months of construction work), yet was in imminent danger of crashing... this didn't float my boat (though I can re-use the 35 MHz RC sender unit).

An autogyro, on the other hand, flies much more slowly, is less difficult to build and should, I think, be more fun.  I verified with a guy on the British Model Flying Association stand that autogyros work at model scale and then found what appears to be a modern classic, the Atom, on the internet: a set of autogyro plans on four sides of A4.  What could be easier than that?

Find below the various phases of the project:


First off, I purchased a digital copy of the Autumn 2014 Special Edition of RCM&E (it doesn't seem to be defined anywhere what the "M&E" stands for, I guess Modelling and Engineering?) which contains the plans.

Autumn 2014 Special Edition article on building the Atom

Then I found the two web locations where Richard Harris discusses his design with many builders: and

And finally I found the location where he stores some additional useful PDFs, most importantly a sheet on making the rotor blades (since, the original source of the blades for this design, no longer trades) and the sections of the forum where some kind users explain how they made their rotor blades from scratch in useful detail.

Here is my parts list collated from those locations, total cost around 150:

Item Quantity Source Notes
Medium grade balsa sheet 3.2 mm x 102 mm x 915 mm 2
Fuselage sides, tailplane/fins and sheeting.
Medium grade balsa sheet 6.5 mm x 102 mm x 915 mm 1 See above.
Fuselage base and some formers.
Medium grade balsa sheet 6.5 mm x 76 mm x 915 mm 2
See above. Rotor blades.
Soft balsa sheet 9.5 mm x 102 mm x 915 mm 1 See above. Fuselage base and balsa in-filling.
Birch plywood sheet 3 mm x 300 mm x 300 mm 1 See above. Undercarriage mount and servo mounts.
Lite plywood sheet 3 mm x 300 mm x 300 mm 1 See above. Several formers.
Strip of spruce 6.5 mm x 6.5 mm x 1200 mm 1 See above. Boom.
Strip of spruce 6.5 mm x 12.5 mm x 915 mm 1 See above. Mast.
Strip of spruce 6.5 mm x 6.5 mm x 915 mm 2
See above. Rotor blades leading edge.
Glass fibre delta plate (0.8 mm thick)
1 Head plate.
Other CNC cut glass fibre parts 1 Various, see here.
Mylar hinge strip

Adjustable clevis
To make attaching the rudder coupler easier.
Nylon clevis
For tail push rod.
Spare landing gear plates
I found the CNC cut fibre glass ones to be too weak.
Push/pull control wire set
To control the rudder.
Piano wire, 3 mm (10 SWG), 915 mm BUT SEE NOTE OPPOSITE.
Undercarriage. NOTE: 10 SWG is actually 1/8 inch, so 3.25 mm and hence won't go through the 3 mm bore collets identified below.  Either accept this and drill the collets out with a 1/8 inch/3.3 mm drill or order 11 SWG wire instead.
Piano wire, 1.2 mm diameter (18 SWG), 915 mm
1 See above.
Push rods and rudder coupler.
C30 rotor head assembly 1
Brushless motor 1
Stand-offs for motor mount, 10 mm long, 3 mm bore

Speed controller and battery eliminator, 30 Amp
1 The brushless motor won't work without one, also supplies 5 Volts to receiver and servos.
Servos 3
Radio control receiver unit (35 MHz)
Comes with pairs of crystals, not including channel 72 (I wanted to stick with my sender unit's channel 72).
35.120 MHz (channel 72) crystal
Not many available, everyone has moved to 2.4 GHz.
1500 mAh 3 cell LiPo battery, 25C discharge rate 1 127 gm, 73 x 33 x 27 mm.
3.5 mm bullet 3-wire connectors
For connecting the motor to stuff.
Seemed a sensible addition.
APC propeller 10 x 7
Light wheels, 70 mm diameter (for 3 mm axle) 3 Undercarriage wheels.
Collars, 3 mm bore 3
Wheel retainers and bell crank retainer.
Push rod linkages
Part of the rudder control.
Neodymium magnets
To hold the false cockpit floor in place.
Transparent plastic sheet (e.g. over-head projector film)
For the canopy/cowl, 0.25 mm thick, 230 mm x 300 mm.
Solarfilm covering with matching Solarlac paint
Your choice of colours.
According to the drawings here.
Optional: a model figure ~150 mm tall

To sit in the cockpit; but you may be better off making your own for scale reasons.

Note: the strange wood sizes are simply imperial sizes converted to metric, which seems to be how this timber is sold in the UK.

Tools/consumables required were:

Cutting Out

I ordered everything and began cutting out.  The first thing I noticed was that the pre-cut fibre glass delta plate was slightly larger than the version on my printed PDF of the plans, which made me realise that I had to print the plans at 107% scale on my printer to get the size correct; for reference, this makes the longest side of the mast 25 cm long.  Also, the centre-page spread was, of course, printed on two separate sides of A4 so I used the spacing of some lettering that happened to appear twice, one of which crossed the gap, to figure out what the overlap should be (virtually none) and then cellotaped the two sides of A4 together.

100% printer scale
105% printer scale 107% printer scale Determining the gap in
                the middle of the page

I cut out all of the paper shapes roughly and grouped them into their materials: lite plywood, birch plywood, 3 mm balsa, 6 mm balsa and misc (no shapes for the fibre glass pieces, since I already have those pre-cut, or for the tail booms and rotor blade materials since they don't need patterns).  Note: when you get the birch plywood and lite plywood sheets mixed up ('cos they look pretty much the same) the birch plywood is the heavier one (180 gm versus 150 gm).

Paper shapes, grouped by material

For cutting out the balsa I chose to use the "pins" method, viz, place the paper pattern over the relevant material, stick in clothes pins at < 1 cm intervals along the edge, remove the pins/pattern and then cut along the line of pin-holes with a sharp modelling knife, using a steel rule as an extra guide on the straight bits and trying to keep the knife perpendicular to the material at all times.  Don't forget to note the light grey lines on the paper patterns which indicate what the underlying grain direction should be.  Take advantage of any nice straight edges/corners you have available on a piece of material by cutting the paper pattern to the edge on one or more sides and aligning it with the edges.  And, of course, once you've put in sufficient pins (at the vertices) to hold the paper pattern down you can simply move around the edge making pin holes at convenient intervals with a single pin.  Where there are semicircular cut-outs they are all 5 mm diameter and can be marked with a few pins and drilled before the rest of the cut-out is done; I put a piece of cellotape over the exit-side and, as soon as I could hear the tip of the drill bit begin to break through it, I turned the piece over and drilled from the other side to prevent bits of the surface pulling away.  That said, masking tape would be better as the drill would be able to break through it neatly.  If chunks of balsa break away at the edges while you are cutting, keep them and glue them back into place with PVA.  Keep the paper patterns somewhere safe in case you need to re-make a shape for repairs (they are also useful when cutting out some of the covering pieces).

Pins detail Pin holes
                remaining Cutting along
                steel rule for straight bits Cutting between
Pattern on
                material with pins stuck in
Final cut shape

For the ply, the pins needed some considerable pressure so I used them only for the corners to hold the paper pattern down and then used a hand-held pin-holding collet-thingy (apparently called a pin vice) I found in my general toolkit to make pin marks at the usual intervals; on reflection I could have done the same thing with the balsa and needed fewer clothes pins.  I drew pencil lines over the pin marks so that I could clearly see the pattern.  I cut the pieces out with a fret saw.

Pinning out the pattern on
The pattern marked out in
                pencil on the ply

Finally, I cut the lengths of spruce and shaped the mast with a fret saw.

Some notes on the cutting-out phase:

Balsa (6 mm) in-fill For pieces F3 and F4 you will need to cut some 6 mm balsa in-fill as well as ply (though leave drilling the holes in the balsa for later when you've made the sandwich that they fill); I don't know whether it matters or not but I chose the grain direction of the balsa to be opposite to that of the ply.
F1 and F2 grain direction
It doesn't say so on the plans but the assembly drawings show the bulkhead pieces F1 and F2 as having opposite grain direction.
The T3s glued either side
                of the tail and glued before drilling
Don't drill the holes in T3 yet, wait for the T3s to be glued into place in the construction phase and then you can drill through them and the holes will line up.
Tail fin extension
Consider making the tail fins 8 mm longer (see below).


Here are some pictures of the bits I bought for the electrics to show how they connect together.

LiPo battery
Battery, overall view
1500 mAh, 3 cell LiPo battery.
LiPo battery power connector: XT60, which is compatible with 3.5 mm bullet connectors.
Charging connector
LiPo battery balance connector (for charging only).

Switch (with integral charging connector)

Electronic speed controller and battery eliminator.
                speed controller
30 Amp electronic speed controller.
Bare wire
Power input: bare wires suitable for soldering into 3.5 mm bullet connectors.
Control input and 5 Volt output: JR/Hitec "S" connector.
Bare wire
                  output to motor
Power output: bare wires suitable for soldering into 3.5 mm bullet connectors.
35 MHz radio receiver.
Radio Receiver
35 MHz, 8 channel radio receiver.
Receiver outputs and power input
Control output and power input pins: JR/Hitec "S" connector spacing.
Socket for crystal
Socket for crystal.

35.120 MHz (channel 72) crystal.
                MHz crystal (channel 72)

Servos (three off).
                metal gear micro servo 1.6kg / .13sec / 11.4g
BMS-373MG metal gear micro servo 1.6 kg/0.13 sec/11.4 gm.
Servo body
Servo body.
                "S" Connector
JR/Hitec "S" connector.

Electric motor.
Note: mounting bracket and spindle attachments not shown.
                  Aerodrive SK3 - 2826-980kv Brushless Outrunner Motor
Turnigy Aerodrive SK3 - 2826-980 kv brushless outrunner motor.
3.5 mm
                  bullet connectors from motor
Power input: 3.5 mm bullet connectors.

3.5 mm bullet connectors
Blocks of three 3.5 mm bullet connectors.

I soldered the 3.5 mm bullet connectors to the electronic speed controller/battery eliminator leads and mated them with the switch and the motor:

3.5 mm bullet plugs
3.5 mm bullet plugs.
Speed controller cables
                  soldered into place
Wires from speed controller soldered into plugs.
Heat-shrinkable sleeving in
Heat-shrinkable sleeving in position.
Sleeving shrunk
Sleeving shrunk.
Plugged into battery XT60
Connected to battery XT60 connector.
3.5 mm bullet sockets and
                  3-socket housing
3.5 mm bullet sockets and 3-way socket housing.
All connected.
Wires soldered into sockets, sockets fitted into housing, motor leads plugged into sockets.

And here is everything all wired up.  I fitted the bracket to the back of the motor so that I could hold it in a vice while it rotates freely then I got out my old radio control sender unit and made sure that everything basically worked: servos rotate, motor spins.

All wired up Radio
        sender unit

General Construction

Note: for all glued joints I used PVA unless otherwise stated.

First I glued F1 to F2, UM2 and UM3 to UM1, and the balsa in-fill to F3 and F4 (using the 6 mm-wide mast to hold the balsa apart the correct distance). I held the pieces in a soft-jawed vice while the glue dried, though, when doing F3/F4, I made sure to slide out the mast after closing the vice so as not to accidentally glue it into place as well.  With the balsa in-fill glued into place I drilled through the 10 mm holes.  I left cutting out the servo holes in F3 for now as my servos aren't the ones used in the original design (those were out of stock) and I wanted to make sure the servos I had ordered fitted well when they arrived.

F1 and F2
Laminating F1 and F2
The undecarriage mount
Undercarriage mount parts
Using mast for spacing
Balsa laminated with F3/F4
10 mm holes drilled through

Then I turned to the base, B1, glued B2 to one end of it and then B3 (placing this as centrally as you can) followed by B4 to the other end of it to form a sandwich, either side of which the tail booms will later be slotted as the final part of the sandwich filling.

B2 glued to one end of B1
B3 glued to the other end of B1
B4 glued on top of B3
The B1, B3, B4 sandwich

Then I glued each of D1 into place on the fuselage sides.  The positioning of D1 is really important: the side of D1 has been cut to be about 5 degrees adrift from the vertical, the angle at which the F3/F4 sandwich, and hence the mast, must be mounted for the rotor blades to gain lift.  The correct angle is achieved by placing D1 on top of B4 with the bottom of the edge of D1 lined up with the end of B4, while the fuselage side is simultaneously flush with the bottom of B1; the F3/F4 sandwich will later lean against the angled edge of D1.  It doesn't matter if D1 sticks out a little above and beyond the fuselage side.

D1 and the B4
D1 and B4 alignment
Fuselage side flush with bottom of B1
D1 glued into position

Having ascertained that my servos are actually of the same dimensions as the ones in the original design, I cut out the holes for them in F3 by first drilling a 5 mm hole, threading the fret-saw blade through it, mounting the blade on the fret-saw and then cutting out, removing quarter-sized chunks first to give me room to manouvre.  Based on a suggestion in the forums, I also cut some cable runs in the balsa (rather than drilling more holes in F4 to let the cables out and back in through the holes already in F4), making sure to leave plenty of room as you will need to be able to tease those servo connectors down the channel and out through the holes later. Note: I could, in theory, have not drilled one of the holes in F4 since the holes are now only required to get the servo cable from the tail servo through to the controller; however, teasing the connectors on the end of the wires from the two servos attached to F3 down through the balsa tunnel between F3/F4 and out is quite difficult and so I chose to keep both of the holes to allow me to push the connector through the exit hole in F3 more easily.

Hole drilled
Fret saw blade threaded
                through hole
Cable run in the balsa

I spent a little while making room in the balsa glued to F4 for the backs of the servos, adding 3 mm birchwood ply strips for all of the servo mounting screws to bite into (I later found that these would be better to be 6 mm thick), making sure that the servos fitted snugly, that the associated wires could be routed and then drilling 1.5 mm pilot holes for the servo mounting screws to make sure that the ply won't split.

F3 with birchwood ply
F4 with cut-outs in
                balsa for servo backs
Servos mounted in F3/F4
Tail servo hole with
                birchwood ply strips
Servo mounted in B4

With that done, I glued the rear half of one fuselage side into place against the straight part of the base (B1) and B4, applying glue to the lower edge of D1 also.  I made sure that the ends of D1 and B4 lined up.  A couple of G cramps were needed to hold things in place while the glue dried; I also put some paper beneath the assembly to avoid glueing it to the workbench.  When the glue had tried, I turned the assembly over and applied a little more glue into any gaps, wiping off the excess.  It looks disturbingly like I'm building a tiny coffin.

Glue fuselage side to B1/B4
Alignment of D1 with B4
Turn over and apply glue to
                any gaps
One side glued

I placed a piece of coarse glass paper on the work surface and then rubbed the base of F4 (and its balsa) against the glass paper to make a roughly 5 degree angle. I also made sure that both sides of F4/balsa were abraded flat and straight, test fitting it into the fuselage along with the other fuselage side to make sure it fitted nicely. Then I glued it to the base, the fuselage side, B4, B3, and D1, leaning it against D1 to achieve that critical angle.  I test fitted the mast and used a file to angle the bottom of it in the same way.  Then I repeated what I did for F4/balsa with F3/balsa, checking that it fitted within the fuselage sides neatly and had a 5 degree angled base.  I glued F3/balsa to the balsa of F4, the base and the fuselage side, using a pair of G cramps to hold the assembly while the glue dried; I kept the mast in place while positioning things then pulled it clear and wiped off any glue that had got stuck to it.  The coffin now has a headstone at one end.

Base of F3/F4 at 5 degree
F4 leaning against D1
Base of mast at 5 degree
F3 and its balsa being glued
                into place
F3 and F4 glued into place

I glued the rear portion of the other fuselage side into place, following the same procedure as for the previous fuselage side, then glued the remainder of both fuselage sides to the base (placing B5 into position to dictate the spacing); given that the fuselage sides are already aligned with the base, I found that it was possible to do this upside-down and then I could add glue to the gaps.  I made sure that the sides were square with the base while the glue was drying. I glued the F1/F2 laminate on the end, and UM1/2/3 underneath, up against F2 and beneath B2.  I used quick-drying Araldite rather than PVA for this; as F1/F2 holds the motor and the landing gear is housed beneath, I guess it is a particular stress point.  This completes the glueing of the fuselage.

The other fuselage side, rear portion, glued
Glueing the remainder of the
Make sure the sides are
                square with the base
F1/F2 glued into placeF1/F2 glued into placeUM1/2/3 glued into place
All of the fuselage glued

Next I needed to make it aerodynamic and pretty, with outlines like this:

Aerodynamic outline

Starting at the rear, I trimmed off the bits of D1 that were sticking up above the fuselage on both sides.  Then I cut two pieces of 3 mm balsa, 65 mm by 45 mm (with the grain running along the shortest dimension), strapped them to either side of F3/F4 using an elastic band with a stick pushed through it so that I could tighten it up, used a G cramp to hold the bottom of the pieces in place and then dampened the outside (only) of the 3 mm balsa with a wet cloth in the hope of it expanding to force the shape.  It did a little but not a lot.  Anyway, I used the same arrangement of elastic band and G cramp, with the addition of some greaseproof paper to stop it being stuck to the balsa, while glueing the balsa pieces into place.  When the glue had dried thoroughly I marked the shape I wanted in pencil and trimmed with a sharp knife, the front being trimmed flush with the front of F3.

D1 before trimming
D1 after trimming
Balsa bent and wetted
Balsa pieces slightly bowed
Glueing complete, pencil
                line drawn

At the front I decided to cut a new piece for B5 (the old one was rather smaller than the space between the fuselage sides) and I glued this into place.  On top of it I glued four pieces of the 9.5 mm thick balsa, cut into lozenges 58 mm long and 40/45 mm at each end.  This gave more than the 30 mm recommended height, leaving plenty of room for finishing.  Then I put the 9.5 mm balsa sheet underneath the fuselage, up against the undercarriage support, drew around it and cut it out. After glueing this in place I made pencil marks all around the fuselage representing the desired curves.

B5 re-cut and glued into
Four lozenges of 9.5 mm
                balsa glued into place
Fuselage ready for shaping

  I shaped the fuselage at the front and the base with a razor plane, removing all material in one direction first, then turning to the other, so that I wouldn't remove my pencil lines straight away.

Planing in one direction
One direction done
The other direction done
Front view

Then I used medium abrasive paper to finish it off and, finally, some Wonderfill (lick your finger, dip it in the powder, apply) to fill in the gaps.

Abraded front
Abraded front
Filled front

Here are the same blue lines superimposed on the result:

Aerodynamic result

I made the false cockpit base, pretty much a free form thing that you can make however you wish, following the basic pattern from 3 mm balsa for the bottom, which needs cut-outs for the servos at the back, and bits of 6.5 mm balsa for the rest.  Glue and then shape/abrade/fill to a pleasing form that fits neatly inside the fuselage.  Later I will add a canopy, a pilot and a method for holding it in place.

Cockpit balsa cut
Cockpit balsa glued together
Cockpit, end view
Cockpit abraded and filled

The tail boom struts were slotted into the remaining part of the B2/B3/B4 sandwich, not glued, just shoved in (and it was a pretty tight shove) until they hit F4.  The ends were then brought together with T1 between them, T2 at the other end, all glued with Araldite as I reckoned these could be under some stress. No part was glued to the fuselage; the tail could potentially be removable if you tried hard.  The tail skid was glued into place with more Araldite, along with the two T3s placed either side of T1 and the fibre glass tail skid from the CNC cut Coolwind parts.  When the glue had dried I drilled the 3 mm hole through the T3s and tidied up the end of the boom with a file and some filler.

The boom inserted
T1 glued into place
T2 glued into place
Tail skid
The tail boom fitted

The rudder crank is a somewhat adhoc arrangement.  Taking the small bag of parts that came with each servo, I cut one limb off the four-armed connector and then drilled the centre of it out to 1/8th inch.  I filed the arm down to about half thickness and drilled the furthest hole out to 2 mm so that the push rod linkage could be mounted on it turning freely (using an additional nut to lock it).  I cut a 35 mm length of the 10 SWG wire, found a couple of washers (a plastic one for the top and a metal one for the bottom, not sure why, it just felt right) and drilled out two of my 3 mm collets with a 3.3 mm drill in order that the 10 SWG wire fitted through them.  Then I screwed the assembly together (don't tighten the lock-nut up yet though, you'll need to undo it later).

The small parts that
                came with each servo
The parts that made up
                my rudder control
The parts bolted

I had a look at the tail and, following the advice at the top of the page here, since I had spare material I decided to recut the tail fins (not forgetting grain direction) 8 mm longer so that they matched the pivot point of the rudder crank and hence gave the widest angle of rudder throw.  There was no need to modify the rudders, they actually matched more closely with the extended fins.

Offset of rudder crank pivot
                point versus original end of tail fin
Proposed extension of tail
Re-cut tail fin
Re-cut tail fin in position

Before attaching the tail plane and fins, which would make the model more difficult to manipulate, I decided to make and test-fit the undercarriage and mount the motor.  The piano wire is quite tough stuff; I needed to use a very solidly held vice with metal teeth to hold it while bending and hammering it into shape.  I didn't cut the wire first; it is only possible to bend it with ease when you can work with the whole length.  After bending and cutting off one undercarriage piece, I just made the second to match the first (better that they are the same than matching the plan).  I made sure that the short section at the end was truly at right angles to the rest (improving on the picture below), as the legs wouldn't clamp into place properly otherwise.  I drilled the holes in UM1 right through the balsa base and B2, using a drill of the appropriate size for my choice of piano wire.  Then I positioned the fibre glass undercarriage plates from the CNC cut Coolwind parts, drilled some pilot holes and fixed everything into place with four short wood screws.  I later found the fibre glass plates to be too weak and so ordered some plastic ones from Hobbyking instead (they are on my parts list above).  I found that the ends of the undercarriage didn't lie flat on the table (they need to do so if there is going to be enough clearance between the tips of the propeller and the ground) unless I bent the wire approximately an extra 5 degrees.

Bending the wire
Wire bent
Undercarriage fixed into
Extra bending

The motor seemed, to me at least, to be a somewhat peculiar thing.  It is sort of double ended: one end has a shaft sticking out and the other end has three mounting holes and you are provided with a threaded bolt and plain nut which attaches there.  I wanted to use a prettier propeller retainer and all the ones I have seen are designed to attach to a shaft rather scew onto a thread, however the motor mounting holes are on the side where the shaft sticks out, so it's not a great deal of use.  You can, I believe, force the shaft to come out of the other end of the motor through certain delicate adjustments using a vice but I didn't fancy trying that so I settled for using it as is and therefore I needed some offsets to stop the shaft fouling the fuselage.  I happened to have a motor mounting plate which was already offset somewhat but the flat one that comes with the motor is perfectly fine, there is enough clearance if you use 10 mm long offsets.  I made some pilot holes and used four long woodscrews to mount the motor.  Finally, I drilled a 10 mm diameter hole in F1/F2 for the wires to pass through.

Motor back wiht mounting
                plate and shaft sticking out
Front of motor with
                mounting holes for bolt
Motor mounted
Shaft clearance
Hole for wires

Next I took a look at the rotor head.  Following the instructions here I took the whole thing apart and discarded all of the white and transparent pieces.  Then I attached the rotor plate from the CNC cut Coolwind parts in their place, using Loctite on the M2 nuts and making sure that the M4 lock nut is positioned upside-down on the M4 bolt and not tightened, 0.5 mm clearance being left so that the rotor plate can flex.  I also tightened and then applied Loctite to the other M2 nuts on the assembly.  I did up the other bolts/lock-nuts; not tight but so that they did not rattle.

Rotor head, as delivered
Discarded parts
Fibre glass rotor plate in
                head assembly
Spacing of reversed lock
Apply additional Loctite

I used Araldite to glue the fibre glass reinforcing plates from the CNC cut Coolwind parts onto the sides of the top of the mast; note that these should be the two completely rectangular pieces, not any of the three pieces with rounded ends (which are for reinforcing the rotor blades) that I mistakenly chose.  Then I drilled the two 2 mm holes for the rotor head bolts and attached the rotor head to the top of the mast (the right way around: the control rods need to be above where the servos mount) then added Loctite to those nuts also.

Fibre glass reinforcers
Assembly mounted on mast
Assembly and mast

I really wanted to get on with mounting the tail plane and fins next but before that I wanted to test-fit the electrics and the switch I had ordered from Hobbyking (Hong Kong) had not arrived yet, so I went back to work on the cockpit and canopy while waiting.  I made myself an instrument panel, using a few pieces of scrap aluminium sheet, and have ordered a 6 inch tall action figure off ebay (Mr Fantastic from the 2006 Fantastic Four film), which I reckon might be modifiable to look very slightly like Sean Connery.  I'll wait for that to arrive before I make the cockpit steering bits.  I made recesses in the front of the false cockpit and the fuselage then used Araldite to glue four Neodymium magnets in place (in order to get the North/South aligned, you will need to scrape the glue off the self adhesive backing of one pair and glue them in backwards); this, along with a couple of pieces of scrap ply cut and glued to F3 to form a slot, is, I hope, enough to hold the false cockpit in place.

Control panel Magents to hold the
                false cockpit floor in place
A few bits of scrap
                of ply to form a slot

The very next day the switch arrived.  It was somewhat larger than I expected but I was able to make a cut in the extra piece of balsa glued to the base and then a further, smaller, cut through B1 itself and the switch was mounted reasonably neatly in the base.

Hole cut in the extra
                balsa glued to B1
Smaller hole cut
                through B1
Switch in hole
Switch in cockpit

Back to the tail; I glued the fins to the tail plane, using some large blocks of wood to keep them square while the glue dried, then shoved the boom up against each fin and marked the position of the hole in the push rod linkage against each rudder. I cut a small horizontal slit, centred on this mark, in each rudder and glued the fibre glass horns from the CNC cut Coolwind parts into the slots with Araldite.  I also shaved off either side of the straight edge of each rudder to make a V shape.

Tail fins glued into
Marking positions of
                horns on the rudders
Horns glued into

A pair of mylar hinges were used to hold each rudder in place following the technique described half way down the page here. I cut mylar strips about 25 mm long and 10 mm wide (i.e. the width of the mylar tape cut in half), marked the desired positions in pencil on both rudder and fin then, very carefully, inserted a modelling knife to make each slot.  Before inserting the mylar strips into the slots I drilled a 1.5 mm hole, roughly in the centre of where each hinge would end up, in fin and rudder.  After inserting the mylar strips I checked that the rudders were centred and flexed sufficiently, while leaving no appreciable gap, then applied Loctite to the holes. The Loctite seeped onto the mylar strip and through the balsa, holding the hinges in place. Once the glue had dried I turned the whole thing over and applied Loctite to the holes on the other side as well.

Mylar strips and positions marked
Slot cutting
Holes drilled
Hinges half inserted
Hinges fully inserted
Hinges glued

I cut a 210 mm length of the 18 SWG piano wire,  removed the push rod linkage from the crank at the back of the tail and threaded it onto the piano wire.  Then I pushed a nylon clevis onto each end of the piano wire.  With some blocks of wood holding the rudders in line with the fins, I adjusted the position of each clevis on the wire such that the peg of each clevis would go through the holes in the control horns; I locked both clevises into the horns at this point but actually they can't be in place while covering and are difficult to undo once locked, so it is best not to lock them in place until later.  I used Araldite to glue the tail plane into position on the boom/T1, being careful to get the tail as square as possible, not so much with the tail boom as with the line of the fuselage body.  Then I attached the push rod linkage back onto the crank, tightening up the grub screw and then the lock nut, putting a dab of Loctite onto the lock nut and also on the points where the piano wire enters and leaves each nylon clevis.

Piano wire with nylon clevis's
Nylon clevis
Adjusting length of
                clevis plus piano wire
Clevis locked into horn
Tail plane glued into
Overall view

I decide to fit the undercarriage, wheels and motor at this stage.  I drilled the wheel centres and the collars out to my chosen size of undercarriage wire, fitted them and applied Loctite to the grub screws.  Getting the motor wires to go through the hole and connecting them to the electronic speed controller was really fiddly, since the motor wires are quite short: I ended up cutting the block of three 3.5 mm bullet connectors that I had fitted at the end of the electronic speed controller into individual connectors so that I could push them through the 10 mm hole in the front of the fuselage (from the inside) one at a time, connect them to the leads on the motor, and then push the wires back inside again.

Outer side of wheel
Inner view of wheel

Mr Fantastic arrived: he was on the small side, 8 or 10 inches tall would have been better but I would have had a real scrunch getting such a figure into the cockpit space; on reflection it might have been better to make my own, larger, head and shoulders to achieve a better scale match (e.g. following one of the drawings here).  Anyway, after a test fitting I cut him in half, cut the "4" from his chest and made a vague attempt at a dinner jacket outfit with black and white model paint.  He may not look much like James Bond but he does look better in a dinner jacket.  I glued him onto the false cockpit with a generous quantity of Araldite and added a back rest made from 6.5 mm balsa.  The false cockpit now needed matt black paint and the canopy added but that I left for later.

Mr Fantastic
Cut to size
Glue into place

Aside from the
mast not being fixed into place (I'll left that to the end to stop it getting in the way during covering) and the false cockpit not being completed, this concludes general construction.

          construction completed

Rotor Blades

To make the rotor blades I followed the suggestions here and here and Richard Harris's drawing here.  First I took the two pieces of 915 mm long balsa and spruce, each 6.5 mm thick (the balsa 76 mm wide) and cut them in half.  Then I glued each piece of spruce to the edge of a piece of balsa with Araldite and immediately taped the two together.  The resulting four blades were laid flat on some greaseproof paper, more greaseproof paper placed over them, then a large flat wood panel plus a weight placed on top of that to ensure that the blades remained absolutely flat while the glue hardened completely overnight.

The material
Spruce glued to balsa and
Weighted down to ensure
After glueing

From the drawing, the blades needed to be 60 mm wide and, being generous, around 450 mm long (but see below) so I began by cutting them down to this size.  Then I adopted the approach described at this link, making a series of marks along which I could trim each blade in a structured manner, viz:

This leaves the leading edge, which I planed into shape by eye with a standard plane (the razor plane is too weedy for spruce) making sure to leave the lower edge untouched.  The aerofoil shape that you need to have in your mind's eye is this:


Note that the height of the spruce leading edge ends up lower than the height of the balsa portion. When planing I found it useful to put a wooden block (the length of the blade) underneath the blade in order to get the plane at the correct angle without it hitting the work surface.  With the planing done I finished the balsa portion of the blades using abrasive paper on a block and I shaped the much harder spruce with an electric sander, weighing the blades every so often so that I would achieve a similar weight at the end.

Stage 1
Stage 1: markings
Razor planing stage 1
Stage 2: planing
Stage 3
Stage 3: more markings
Stage 4
Stage 4: balsa planed
Leading edge roughly
Leading edge planed
Blade completed
Aerofoil shaping completed
Blade on jewellers scale

The four rotor blades ended up weighing 19.6, 19.4, 19.5 and 18.5 gm, so it was clear which one should be left as a spare.  The ends that attach to the rotor head have a cut-out which, as far I could tell, needs to obey no particular shape.  I found that the large reinforcing plate from the CNC cut Coolwind parts was the perfect size to use as a template.  So, continuing with the three rotor blades that most closely matched in weight, I turned each one upside down (so that the flat side is up), placed the large reinforcing plate such that it was flush with the leading edge, then turned it to its mirror image and used its outline to help mark the cut-out.  Then I glued the large reinforcers in place in the original position with Araldite.

Reinforcer placed
                into its intended position
Reinforcer used as
                template for cut-out
Cut-out marked
Cut-out cut out

For the remaining steps I needed to find the balance points of each rotor blade in two directions.  Terminology-wise, the two directions are chord and span, as follows:

Chord and span

The chord balance point is required in order to know how far back from the leading edge to drill the fixing hole (and glue the reinforcer for the fixing hole).  The method used was to push a pin in each end of the rotor blade, making sure the pins are parallel to the leading edge, and then see if the rotor blade balances level when resting on those pins.  If not, move the pins and repeat.  With the chord balance position established, I made a mark 1 mm back (towards the trailing edge) from the balance position and this is where I glued the small reinforcing plate from the CNC cut Coolwind parts onto the top of each rotor blade.

Establishing the chord
                balance point
Top reinforcer glued into

To establish how far in from the end of the rotor blade to drill the fixing hole, there is a useful drawing near the top of the page here, which suggests that it should be 10.5 mm inwards.  This made the distance from the fixing hole to the outer end of my blades 440 mm, which is 10 mm longer than the 430 mm in the drawing.  I decided to leave the blades as is, rather than trimming them down to this length, though I did make sure that each blade was of identical length (using the electric sander).  I used a centre drill first, to make sure the fixing hole was accurately positioned, then followed up with a 3 mm drill.

Fixing hole drilled

The drawing calls for a 4 mm wide, 0.4 mm thick, piece of shim to be glued to the trailing edge of the lower reinforcer in order to angle the blades downwards slightly once mounted.  I cut away one laminate from the birchwood laminated sheet, abraded it to roughly the right thickness and then glued strips of that in place using Loctite.

Cutting one sheet from a
                laminated sheet
Measuring the abraded sheet
Shim glued into place

Next came the somewhat tedious process of balancing the blades.  I weighed each blade and marked them in descending order of weight, I, II, III.  Then I balanced each one span-wise on a screwdriver shaft and marked the balance point. The idea is to make the balance point of the two lighter blades the same as the heaviest blade by adding weight on the appropriate end of the blade.  Weight is added by applying blobs of Balsaloc at the light end of the blade, making sure to add the blobs evenly about the chord balance point so as not to move it.  I shoved a piece of metal bar through the fixing holes to make sure that all three blades were at the same reference point.  I was lucky in that my heaviest and lightest blades had the same balance point, so I only needed to add Balsaloc to the middle blade to move its balance point to the correct position.

Balancing a blade on a
                screwdriver shaft
The balance points of the
                three blades
Balsaloc on the blade

Then I weighed the blades again and added Balsaloc (this time around the span balance point, though I realised afterwards that it should be around the intersection of the chord and span balance points in order not to throw the chord balance off) of the lighter blades until they each weighed the same as the heaviest.  As an idea, this amount of Balsaloc smear:

          0.12 gm smear of Balsaloc

...weighed around 0.15 gm (it dries transparent) and the final weight of each blade was 21.40 gm.

The three blades
Blades test fitted

It only remains to cover the blades, which is dealt with in the next section, noting that there is a further balancing stage once the rotor blades have been covered.


I wanted to have matching colours of covering material and paint, so I went for Solarfilm and Solarlac in the colours of the original Atom: red, white and dark blue.  I also ordered the original Atom decals from Model Markings.

While waiting for these materials to arrive, I checked the model for dents/depressions and filled/abraded the surface until it was satisfactorily smooth, taking the time to build up a few layers in a couple of places. Then I applied Sand 'N' Seal all over to seal the wood grain and gave that a gentle rub down.

I looked up general advice on how to apply covering and it went like this:
There are also detailed instructions from Solarlfilm.  After some consideration, I chose the following covering order, of which the first step was the one where the over-allocation of hands was required:

Single piece of blue film over the fuselage nose and sides: I made sure this was considerably larger than the area to be covered as the angles were odd and the shape of the nose distorted things (it took me three attempts to get both the size and the initial position right).  I started ironing from the middle of the nose and worked around one side and then the other.  I trimmed the excess and then worked out to the edges, used a hot air gun to stretch out the wrinkles before finally sealing the lot with the iron.  There were a few remaining wrinkles, which will be covered by the canopy/cowl, and some breaks in the colour, which I patched with small pieces of blue film later. The oversized
                materialMain work of covering
                completedDetail of front of
                noseBreak in the colourRear viewBottom view
Red film on the fuselage base: I cut a 320 mm by 70 mm piece of covering material for this and didn't use any Prymol on the film overlaps as the edges will be further covered in the next step.  As you can see, I used the box from the iron plus some rags to trap the fuselage while working.  The darker blue film underneath showed through but I didn't mind that on the base where no-one is likely to see it. Material for
                covering fuselage baseBase coveredFront viewRear viewSide view
Strip of red film 20 mm wide on the lower fuselage sides: overlapping that on the base, coming up as high as the bottom of the curve at the rear of the fuselage, this time with Prymol applied underneath to improve bonding.  I found that the contrast between the blue film and the red film underneath showed through so I applied a second identical red film strip over the top of the first to improve matters. Red film on
                fuselage side, first coveringRed strip on
                fuselage side, second covering
Strip of white film 25 mm wide on the fuselage sides: mounted just beneath the edge of the cockpit and meeting or slightly overlapping the edge of the red film, a coat of Prymol applied underneath.  I washed the cover of the iron with a hot water and a bar of soap after this as it had started to deposit red and blue marks (which can be scraped off gently with a finger nail) on the white strip. The white strip
                appliedRear view
Red film over all of the tail fins/rudder on the outer side plus overlap: I took the paper patterns for these parts, used a pin to mark around them on the red film with a ~10 mm overlap on all sides but the hinge side (not forgetting the extra 8 mm on the hinge side of the tail fin) and then used scissors to cut along the pin marks, giving me the two pieces of covering.  The patterns need to be used the correct way up for each side being covered.  I tried to make the overlap as neat as possible while applying the covering. Marking
                outlinesOuter view
                of coveringOverlap
Red film over all of the tail fins/rudder on the inner side, no overlap: I used the same technique as above to pin the outline into the film, this time exactly on the lines, then cut along the pin marks with scissors.  I made a slot for the push rod linkage in the rudder covering and cut the tail fin covering in two to go either side of the tail plane, removing a slice to allow for the thickness of the tail plane, afterwards covering any small gaps with additional bits of red film, Prymol beneath. I washed the cover of the iron once more at this point. Inner side
                angleTail fin
                and rudder covered
Strip of white film 25 mm wide on the tail fins/rudder, outer side plus overlap: a strip 300 mm long was sufficient, cut into four pieces, Prymol applied underneath.  It was mounted just above the tail plane. White
                stripe on outer sides
Strip of white film 25 mm wide on the tail fins/rudder, inner side, no overlap: again a 300 mm long strip, cut into four, was sufficient, with Prymol applied underneath and the strip positioned to match the one above.  I cut a slot in the rudder portion for the push rod linkage. White
                strip, inner side
Strip of blue film on the tail fins, meeting white film, outer side plus overlap: I turned the model on its side, placed some blue film under the tail fin and cut around it with a modelling knife, leaving a ~7 mm margin. Blue strip on
Strip of blue film on the tail fins, meeting white film, inner side, no overlap: I followed the same procedure as above but leaving no overlap this time.  There was a spot of red left on the rudders above the white strip which I painted blue with matching Solarlac paint later. Blue strip on inner
Red film on the tail plane underside: I used the paper pattern for the tail plane and a pin to help me out, just as I did for the red film covering of the tail fin/rudder above, and cut myself two pieces, one for each side of the tail plane, trying to ensure a small overlap on the fin-facing edge but not on the leading and trailing edges. Red film on
                underside of tail plane
Red film on the tail plane top, with overlap: again, I used the paper pattern for the tail plane and a pin to help me out, including a ~10 mm margin on leading and trailing edges and no margin (but no gap either) on the sides. Overlap on tail
                plane undersideTail plan covered
Red film on both sides of the tail skid: I used the paper pattern to help me mark the shape out and left an overlap at the rear. Skid covered
Red film on the top then the bottom of T2: I used the paper pattern again, leaving an overlap at the trailing edge, and I also took the time to cover the raggedy portion on the lower rear of the fuselage beneath T2.
Lower rear of fuselage
                before tidyingLower rear of fuselage
                after tidyingCovering of T2, lower viewT2 covered
White film on the underside of the rotor blade, no overlap: this was pretty easy, pin method to the rescue again, Prymol applied afterwards around the edges where the overlap will be.
Rotor blade
                underside covered
White film on top of the rotor blades, plus overlap: I marked out the shape for the covering using the pin method, leaving 10 mm overlap on all sides then, for the trailing and leading edges, I supported the rotor blade on a block (the length of the rotor blade) so that I could roll the iron over the edge; there was still some inevitable gathering at each end, the inner edge of which won't esaily be seen and the outer edge of which I felt was passable.
Covering cut out
                with overlapsRolling the ironOuter edge of
                covered rotor bladeInner edge of
                covered rotor blade
Rotor blade
                covered, undersideRotor blade
Blue and red stripes on the ends of the rotor blades: I went for 15 mm wide strips, blue on the outside, a gap of 15 mm, then red on the inside, Prymol applied underneath, a strip 60 mm in length for the underside applied first, then a strip 75 mm in length (i.e. with overlap) for the top.
Blue and red
                stripes applied

Here is the body with the covering completed.


The remaining surfaces were painted in the matching Solarlac colours and the false cockpit in matt black.

Painted, front view
Painted, rear view
Painted, main fuselage view

The nylon clevises of the rudder control rod can now be locked into the push rod linkages.  I could have glued the mast into place at this point but decided that it was really useful to have it removable so instead I drilled two 3 mm holes through the centre of the F3/F4 assembly and the mast such that I could use two 25 mm long M3 bolts to fix it in place.

Bolting the mast into place

That's the somewhat lengthy process of covering and painting done; I left the application of the decals until later to avoid them getting messed up as I continued to manhandle the fuselage.

Canopy And Cowl

In order to get the canopy correct, I decided to mount the servos first.  The servo for the rear was easy but, for the servos at the front, I found that I couldn't push them fully into the hole I had dug when they were mounted with the servo spindle upper-most (meaning that the servo cable, which exits at the spindle end, had to run back down behind the servo and so required extra depth).  This worked when I test fitted them but the addition of paint around the hole probably didn't help, the orientation of the cable down the back was impossible to control and I didn't want to force anything.  I could have tried harder but instead I decided to mount them with the servo spindle lowest, which would also make the mechanisms less visible.  This meant quite a large cut-out in the false cockpit floor, making it fairly weak.  On reflection, I should have made the cross-pieces to which the front servos are attached thicker (6 mm rather than 3 mm), then there would have been more room for the cable at the back, the servo spindle could have been upper-most and the cut-out in the cockpit floor much smaller.  Anyway, since I had a little bit of space to fill on the edges of the false cockpit floor anyway, I glued a spare length of wire to each side with Araldite to strengthen it.

The servo arms for the front servos had one side cut off and then the outer hole drilled out to 2 mm so that a push rod linkage could be inserted into each one; no lock-nut this time, instead a washer and some Loctite added to the end of the thread after gently tightening the nut.  Having the grub screw of the push rod linkage facing backwards made it a bit difficult to tighten but the result is neater so best stick with it I thought, and used the modelling knife to tighten it up when the time came.

The lower mounting screw of the front servos was quite difficult to insert so I decided to leave it out; there was absolutely no chance of either servo coming loose.

Rear servo mounted
Front servos mounted
Cut out in false
                cockpit floor
Strengthener added

For the canopy, considering that the initial flights of this machine are likely to involve more crashing than flying, I didn't fancy taking the time to make formers and fiddle with pastic bottles that might not last the distance, so instead I chose to use some transparent plastic sheet.  Since the power switch I purchased included a charging connector, I won't need to be removing the false cockpit to get at the battery all of the time (just around every 10th charge to balance charge the battery, as that requires access to the balance connector), so I decided to improve aerodynamics and keep construction simple/strong by making a one-piece cowl and canopy.  First I wrapped the backing paper from the transparent plastic sheet around the fuselage and drew a line on both sides where I would like the canopy to end.  Then I used this as a template to cut out the transparent plastic sheet using a modelling knife.

Marking out the canopy size
Cutting the transparent plastic
Test fitting

After getting the shape roughly right, I installed the push rods between the front servos and the rotor head (without fixing them finally in place) so that I could cut the rear of the canopy into a shape which wouldn't foul them.  Then I kept the transparent plastic sheet in place by hand, with an elastic band over the nose-end to help me out, while I marked (with an indelible pen) and then trimmed (with a scissors) the nose backwards in a step by step fashion in order to achieve an end that was as small a diameter as possible, ending at around where the propeller fits.  When I had got it close to what I wanted I put some greaseproof paper between the false cockpit floor and the fuselage body and then used Araldite to glue the transparent plastic sheet to the sides and the front of the false cockpit floor (only), holding everything in place with pieces of wood and weights until the glue had set completely.

Marking cut-out
Trimming nose
Glueing the canopy to the false
                cockpit floor
Canopy glued to false cockpit

With the false cockpit floor/canopy fitted to the fuselage I placed a clamp in a position where I will later add a couple of small fixing screws.  Then I made a longitudinal cut along the top/centre of the transparent pastic sheet in the cowl area and used Araldite to glue it into an overlap (holding things in place with elastic bands and weights again while the glue dried), such that the diameter of the end of the cone, when everything is held together, is now close to that of the motor.  Then I made a mark with an indelible pen on both sides of the transparent plastic sheet in a pleasing shape for the lower edge and front of the cowl, along which I cut with a scissors; this leaves the bottom of the cowl open to dissipate heat from the motor.  I drilled two 1.5 mm holes through the transparent plastic sheet and into the fuselage, low down in order to bite into ply rather than balsa, and used a couple of 10 mm long sheet metal screws (should really be hex headed to look neat) to fix the canopy/cowl in place.

Slit in cowl area glued
Cut line for bottom of cowl
Canopy cut and mounted
Front view

All that remained was to cover the canopy/cowl areas, hiding the glue marks in the process.  The chief problem here was avoiding air bubbles between the transparent plastic sheet and the covering film, which build up really easily.  I started by making a template in greaseproof paper for the covering of the canopy sides, which will be in blue Solarlac film.  I applied the film to the transparent plastic sheet and smoothed it into place well with my fingers to remove all the bubbles I could.  With both sides of the canopy covered in this way, I applied a 30 mm wide strip of blue Solarlac film across the front of the canopy area in the same manner.  A slight bulge developed in the transparent plastic sheet between the canopy and the cowl as the film shrank on top of it so I made an additional rounded cut in the transparent plastic sheet to make it less obvious.

Canopy covered, side
Top view

For the cowl, I covered the whole area in blue Solarlac film first, one piece covering from the leading edge up to the end of the longitudinal cut and then a separate strip beyond that to finish off.  Then I applied Prymol to the areas where white and red strips were to be laid on top.  The white strip I tapered from 25 mm wide down to 20 mm wide at the front, then the red strip was layered twice to prevent the blue showing through.  Initially I left the gap between the canopy and the cowl areas alone but then I decided it looked a bit strange and so I covered that area as well.  As you can see the cowl has a slight tendency to splay out so I kept an elastic band around it in the hope of persuading it otherwise and that seemed to do the trick (the picture below right was taken a day after the others).

Cowl covered, side view
Cowl covered, front view
Area between cowl and
                canopy also covered


With the switch, motor and servos already mounted it was easy to connect all the electrics together in the body of the fuselage and string the antenna cable out along the tail boom.  For the record, things were connected to the receiver as follows:

Right servo (as viewed from the front of the autogyro).
Electronic Speed Controller (and hence motor).
Rudder Servo.
Left servo (as viewed from the front of the autogyro).

It was quite a tight fit getting everything into the body.  I placed the battery between the electronic speed controller and the receiver in the hope that it might act as a shield.

Electronics stuffed into
        the body

The following tasks remained to finalise things:

Push rods glued into
            rotor head

Cables at tail
                    end Cables at
                    servo end Cables
The "Atom" decal Peeling away from the cardboard
                  backing Applying decal to body Peeling away carrier "Atom" decal applied Both decals applied
Rotor blade balancing
DIP switch settings on
            the RC sender unit
And there she is.  Ready for her maiden crash.  Click to see her starting up.


I measured the angle at which the boom meets the ground while I suspended the model with my fingers (holding the ends of the bolt where the rotor head is attached); I made the angle about 18 degrees, which is at the high end of the expected range.  The blades were attached firmly but not very tightly, allowing them to move in the event of a crash.  I planned to find a deserted location, on grass, and have the machine run along the ground without taking off, just to check that I can make it move in a straight line.

The collated advice I could find on flying was as follows, with all the emphasis being on the importance of giving plenty of time for the rotors to get up to speed:

I posted the above on (here at the bottom of the page) asking for confirmation and any further advice.

Back to Meades Family Homepage