Final Frame

End-of-week round-up July 26 2019, and 'Houston, we are looking good'. I want you to imagine seven more power-plants affixed to the above, three more to the lower quad and four to the upper.

For the technically-inclined (and in Imperial measure) the spars are 30-inch by 2-inch by 1-inch by 1/16th, the props 32-inch, the deck 12-inch square, the aperture 9 and 1/2-inch and finally the depth 8-inch (not including the upper set of propellers).

The weight of the basic frame alone is a little under twelve pounds and still too heavy so far as I'm concerned but this will do for a proof-of-concept. The 'wearable' single-deck drone I featured months back weighed just over eighteen pounds, so I've added a further deck and four more rotor-arms and still reduced the weight by a third.

As the load is now spread too over eight arms, there remains scope for lighter alloy sections without reverting to carbon-fibre.

I've run with the simplest arrangement for test-flying though there are two variations to test longer term viz. (a) the decks could be spaced farther apart and the upper set of propellers under-slung instead of over-slung and (b) the upper frame could be inverted to provide an eight-pronged layout when viewed in planform.

The frame is modular in so far as the upper 'drone' can be detached from the lower, and in theory any number could be stacked. The principal problem with 'stretched' airframes of this kind is the increasing torsion, which requires fitting side-panels to act as shear plates.

The mannekin represents a half-scale though the frame will accommodate myself for test-flights. Realistically a commercial version would sub T-Motor U15 motors for the U13 type here (along with 40-inch propellers instead of 32-) yet still remain inside the dimensions of the GoFly competition.

For the record the axial distance between hubs is 48-inch and adding the 32-inch for the arc of each pair of propellers produces an 80-inch span. For roof-rack transport however the propellers can be stowed parallel to lie within the foot-print of an automobile.

We should be moving toward power-plant fitment and shakedown testing in the near future with tethered flights beyond that based on a 48-volt ground-power source.

Thereafter we have to return belatedly to Voltaplex with a spec on the LiPo battery-packs for free flight. My preference is for batteries to be body-worn in a back- (and/or front-) pack so that they can be tailored for the mission, as with current commercial jet-packs.

For the pilots among you the flight-control configuration is likely to call for the lower set of rotors to provide yaw plus collective with the upper set restricted to direction, and initially commanded by standard remote-control hand-sets (whether from base-station or onboard).

I am confident moving forward that the configuration as seen can be made to fly, and frankly having examined every other variation, it's going to have to as I'm betting the farm on it!

Finally to anyone thinking of designing an eVTOL, my advice is... don't.

Ever since Sikorsky helicopters have needed a big rotor on top and a smaller one on the tail, but with electrical flight nowadays that's gone the way of every other Cold War certainty.

Bleri... Oh!

One hundred and ten years to the day since Louis Bleriot flew the English Channel, the remarkable Frank Zapata tries on his jet-powered 'flyboard' but sadly ditches during a refuelling event.

It's only good news however because (a) it IS going to happen and (b) he has a military contract already.

Thus inspired I re-examine the prospects for using what we have in hand to test-fly a foot-mounted drone, seeing as they're all the rage.

Actually a mock-up in the lounge here suggests the prospects are pretty good. It'd be a whole lot easier to build as a simple quad, but lacking redundancy would suit it only to ground-effect operations (for which I've always been convinced there's a market).

The upper-deck is hollowed out ~ because I can ~ and the pair of decks will be separated by tie-rods again just sufficient to provide clearance to the lower set of propellers.

Ideally the upper set of propellers are better under-slung for reasons of safety, but for the test-flights at least we might as well run with the simplest configuration and mount on top.

First time as well that the motor's been anywhere near the frame, and there's nothing like arranging parts in this way to accelerate progress... and this is beginning to look like a flyer.

And although modelled here with the half-scale mannekin, this same airframe fits me too.

Umpteenth Build, Half-Scale #3

This is going to be the end of me.

Too much torsion and looking at it am not at all surprised. The classical way to build a fuselage is using stringers, longerons and a stressed skin and this is missing the latter.

Strength in compression however is remarkable and mainly thanks to that stay at waist-level that is slid into place like a pair of trousers... there's around twenty-seven kilos up top and barely a bend in eight millimetres of tie-rod.

There may yet be an argument to say that the torsion might not matter when it comes to flight. Anyone who's sat at the back of a 757 will have noticed the twisting along the length of the fuselage, or seen the belly of an overly-stretched DC-9 buckling under the load over bumps on the taxi-way.

Or indeed wings, which are able to flex in the case of the 747 over an arc of thirty-five feet.

Problem is, lacking the software the only way to find out is to get it airborne and that's a long way to go down what might be a dead-end.

Umpteenth Build, Half-Scale #2

Tie rods in place. These are M6 threads and they run through the depth of the lower-deck or 'landing module' as I like to call it. I've drilled these without a power pillar, so am pleased that they run generally true.

In a perfect world all those surfaces that they penetrate would have been drilled en-bloc so as to line them up precisely, but then hindsight is a wonderful thing.

These rods will support the upper quadcopter or my 'command module' and enroute they will pass through a stay worn around the pilot's waist that stops them from bowing unduly.

Umpteenth Build, Half-Scale #1

I've gone for a mannekin a little under a metre, which (like the motors, propellers and ESCs) has travelled all the way from China.

Its steel pediment will do fine 'as is' measuring as it does ten inches square.

The tie-rods I bought locally are too short so I've ordered 2000mm M6 rods online.

Never thought I'd ever be ordering clothes for a mannekin online, but I've gone for an olive green flying-suit instead of the red, along with a full-face helmet.

(At least it won't be needing the school fees.)

I need to take him/her with me to a shoe-shop for a pair of boots, which will doubtless label me in perpetuity as the local crank.

"One Small Step for a Mannekin...

One... giant leap for mannekind."

For this is our moon-shot. And not many people know this, but the first man to orbit the moon was not Michael Collins but a Russian mannekin accompanied by two tortoises. The latter were dissected upon return to Earth to see how well they had survived travelling a half-million miles through space. Very well, apparently, or at least until the dissection.

In ways that Neil Armstrong was to benefit from turns of fortune in being the final selection for commanding the first lunar landing, the mannekin here passed the albeit more hurried selection by dint of the fact that our engineer's wife had run out of the 1.20 metre models. The one here therefore represents say a five-year-old child rather than a seven-year-old.

The mannekin commented at the press-conference: "I'm delighted to have been selected for the UK's effort to get people airborne in a flying phone-box and although my own piloting skills will not be called upon as were Neil Armstrong's when 3000 feet above the lunar surface, I have every confidence in the automatics."

And with this he/she was escorted off the podium in order to get suited and booted.

Eureka!

I have to finalise the design on this day of all days, fifty years hence from the lunar landing.

I've fought shy of this layout for some time because I could see no easy way of negating the torsion to which it would be prone along the length of those tubes.

Since the outset however (or not long thereafter) it occurred to me that quadcopters could be fixed both top and bottom, not least because the latter provides means of supporting the vehicle at rest.

Experimenting in the lounge today with lengths of dowelling I realise that by using a support around the mid-riff (and I've built many in connection with the 'wearable' drone) the rigidity of the arrangement is improved sufficient for it to be the basis of a prototype.

Let me recapitulate the course of the design effort to date, with the pros and cons:

(a) A form of flying phone-booth ~ too heavy and too complex at this stage.

(b) A hollow drone worn at the mid-riff ~ essentially been done elsewhere.

(c) A ground-level drone to sit or stand on ~ neither new nor original.

At times in the past I dreamt of helicopters with a rotor both below and above the cabin, as it answered two of the challenges associated with conventional rotor-craft viz. retreating blade and a tail-rotor to counteract torque.

At the time the idea was too complex given the state of the art, but with distributed rotor-arrays used by eVTOL technology it is feasible to build an airframe not dissimilar to the one above, which features a set of propellers overhead as well as underfoot.

And I intend to build it.

Next I have to get over to the electrical engineer's farm where ~ beside everything including the kitchen sink ~ there's an ample supply of full-size mannequins.

Shall build the frame around a junior mannequin at the outset, not least because these are the only type with moveable arms that I can fix around a pair of joy-sticks, as if for flight.

Space Frame

Complete the pair of frames to be used at head and foot of a vertical airframe designed to fly a pilot stood upright.

Been fifty years since the launch of Apollo 11 ~ which I recall watching on TV ~ and occurs to me how tasks do well to be split into component parts.

The space-frame installed between drones above and below an operator can similarly be treated as a module able to be built separately.

The quadcopters here have a base just twelve inches square, on cantilevers thirty inches long.

I've stuck with alloy and rivets because if it works on a 747 then it has to work here too.

For the Record

Actually, see no need for suspending the blog during the course of construction because (a) seems churlish and (b) it's my diary and I'll cry if I want too...

In order to squash a 'flying phone-box' into the allowable dimensions ~ complete with quadcopter top and bottom ~ I've to reduce their foot-print to a minimum.

This involves dismantling the previous build and recycling the parts, which initially involves foreshortening the cantilevers that support the motors and propellers.

With nine inches to come off each, my preference over the jig-saw is to take them to a local engineering workshop for a straighter cut... achieved as above with a band-saw instead.

The veteran (who has broken off from the current task to attend) refuses to accept any cash.

I insist on a tenner however, suggesting he use it for lunch?

Says he'll buy a pint on the way home from work.

My kind of engineer.

Under Construction

Been ham-strung from the outset by the outline dimensions required to meet the GoFly challenge, which inadvertently penalises airframes that extend vertically whilst remaining inside the admissible foot-print.

Nonetheless if this concept has any value at all, it lies in the recognition that most people do their commuting (when they're not sat in a traffic-jam) while standing up.

I'm not going to compromise on this despite my myriad attempts at a work-around, but I do envisage a proof-of-concept at full scale that retains a drone at top and bottom whilst still meeting the criterion of being 102 inches from any one point on the vehicle to any other.

Until that prototype is airborne, though, I won't be updating either the web-page or the blog.

For a single flight (as Boeing 707's test-pilot commented) is worth a thousand speculations.

Garage Drone #6

... and finally the two halves are bonded together to produce the octocopter. The parts here are joined with adhesive and will be secured with tie-bolts that pass through both halves.

Garage Drone #5

... before the centre-space of each module is bulked out with PVC or PU foam block...

Garage Drone #4

For each of the modules, a dozen pop-rivets are used to secure...

Garage Drone #3

Removing the ballast but leaving the top-plate in place, glue the next set of spars to the plate in the same way as previous.

As this plate is no longer available to distribute the pressure of the ballast, I have used a trolley to the same effect.

Wipe the edges with a solvent shortly afterward to ensure the upper "quad" assembly does not get stuck to the lower.

Garage Drone #2

Use the top-plate of the same dimensions as a spreader in order to apply pressure to the first application of adhesive but DO NOT adhere this top-plate to any of the parts at this stage.

Garage Drone #1

On a clean and flat surface lay out the base-plate, in this instance formed of aluminium of dimensions 500 x 500 x 1.5 mm.

Arrange the spars as shown ~ here measuring 1/16 x 1 x 2 x 39 inches ~ and mark up with a felt pen around the margins and the length of each spar as a guide to preparing the surfaces to be joined.

They should then be joined by a suitable epoxy and the choice is between a long cure suited to metal, or here a fast-setting mix to expedite. Super-glue might yet be tried and is likely to be equally cost-effective whilst altogether faster.

Effectively however this process serves mainly to locate the parts, as the plate and sections will be pop-riveted afterward.

Another Saturday Afternoon Gone

When you're designing these things, just when you think you're safe to go back into the water, another shark pops its fin up.

Specifically the two quads will have to be joined with a spacer, a little like the filling in a sandwich, to ensure that the propellers and motors are basically contained within the vertical depth of the airframe... in other words so that they do not stand proud of the deck.

(You can see that better in the profile view of the 1:10 scale model below.)

Exactly how thick that spacer needs to be is as yet an unknown quantity, as the specs for the motor show only its own depth ~ six centimetres ~ and not that combined with the propeller.

This I shall do myself with a ruler, as computer-aided design is well above my pay-grade.

Incy-Not-So-Wincy Spider

On two minds about whether to order the body or the legs first, and opt for the latter. Back to Simmal in Preston who turn the order around in thirty minutes and I come away with a full set of 39" spars.

The section I've gone for is 2" by 1" by 1/16" so we're back into Imperial measures, what with 32" props too.

Reason I've gone for 39" is the fact the alloy retails in 5.0 metre lengths, and given the width of the blade used for cutting it, you can squeeze five of these out of each unit with no waste left over.... I'm that mean.

Happily it means that I've ten spars in all, with two left over to go toward repeat builds.

Bit nervous at the weight loading them into the Jimny, but then remember there's two to split from the pack.

And turns out they weigh less than 625 grams (1.33 pounds) apiece, which is almost half the weight of the altogether thicker spars I used for the original alloy build seen here.

It means I've doubled the number of rotor-arms without significantly increasing the weight, in the way that we've doubled the number of motors, propellers and ESCs to the same effect.

Gryphon incidentally are heavy-drone manufacturers in Korea and we simply share a common UK distributor who in turn imports the T-motors that power both efforts.

They lift cargo, however, whereas we intend to lift people.

Independence Day (2.0)

Been a year as of tomorrow since I formed the company with a view to this venture and I am surprised at how much ~ and equally how little ~ progress I've made.

With the electrical equipment out of the box now, here's an email to my team-mates and shareholders that is as good an update as any:

Gents,

Apologies for what appears to have been slow progress on all fronts but there are times you need to ‘step away from the vehicle’ and reflect on the way forward.

The motors and propellers and speed-controllers (ESCs) are landed from China, and to look at them is to realise that electric flight (and cars) really are the next big thing… the 32” carbon-fibre propellers are so light I figured the boxes were empty, and the motors at 4-inch diameter will raise around 20 kilos apiece (and we’ve eight).

Taking everything into consideration I figure the initial prototype has to take the form of a simple ‘flying carpet’ as per the attached, which is suited to a seat although a podium is not out the question.

With this up and running the fundamental design is adaptable to top- and mid-mounted positions around any form of vertical space-frame, but I like how this looks as a way forward.

Happily the eight-legged layout (basically two quads bolted together) also appears to have been wholly unique from the patent point of view beside promising more stability and redundancy.

The dimensions are well within the fly-off regs (which the ‘flying phone box’ variations generally weren’t) and the airframe is only around five feet square with props stowed as shown and therefore roof-rackable and flyable out the garage door.

We also have barn/workshop/flight-field facilities available to the project with a chance of on-the-spot accommodation… bit like the Wright Brothers at Kittyhawk, but without the sand.

The motors are driven by 48V so we’ve motor batteries available for ground-electrics and tethered tests, whilst the real-deal batteries will be sourced from a bespoke firm in Hong Kong.

A reason for the ‘powered-platform’ design is in fact that their weight (stored as ballast in the base) maybe as much as half my own weight for a flight of ten minutes in ground-effect.

The arrangement also favours the simplest possible flight-control architecture, with each propeller pair driving left-right-forward-back respectively and the throttle adjusting height alone.

Control laws for the engine-out case will have to wait and any failure prior will hopefully take the form of a tight-spiral descent to a softish landing with concomitant damage inevitable.

The lower set of propellers will also be shut down for take-off and landing with the upper set providing those manoeuvres, to avoid prop-strike.

(Ground-effect should assist this up to around one metre by providing a surplus of thrust of between 25% and 50%).

With an airframe complete in July we can then push for first-flight with or without tether and likely a seated mannequin, and having done this the type can be listed in global databases of such project builds.

The biggest players have spent billions of dollars to date on eVTOL types so we haven’t done too badly in the double-garage to date…

Contracts will also need signing this month as otherwise there’s no entry to the ‘pit-lane’ at the event itself in California.

C.

Under Construction

Parts have arrived and we're set to build the first iteration.