My friend Koen Van de Kerckhove has been busy sketching away, and his ever-inventive brain continues to come up with interesting designs.  I’ve taken one of his sketches, made some RTFM-type changes, and this is what I came up with.
It is a Flying Flea, but a canard-variant.  i.e. the front wing is smaller than the rear wing.  Axel Darling proved that this configuration is superior to the Mignet formula (which had the larger wing in front). By reducing the front wing a bit, it puts the CG roughly in the correct position (at about the pilot’s belly button).  Making the front wing smaller moves the CG rearwards.

The rudders (two of them) are mounted to the rear wings.  That is going to make a very striking package.

Engine?
Well, it needs to be light, in order to get the CG in the right place.  Possible candidates are the Eos 150 (4-stroke, 30hp, 20kg, 2.5l/hr); the Bush Dawg 4 Stroke EFI Paramotor From BlackHawk (4-stroke, 40hp, water cooled, EFI); the Vittorazi 185 EFI (the only EFI 2-stroke – 15kg, 26hp).  I’m not a great fan of 2-stroke engines, but the EFI on this one gived almost 30% greater fuel economy over a non-EFI 2-stroke, the exhaust system is extremely quiet, and there is little to no power band to speak of.  This is the first 2-stroke I’d consider flying behind.

One of the advantages of buying a paramotor is that it comes with everything – all of the above, for example, come with everything already wired up.  Just mount the digital readout, fix the battery in place, mount the throttle somewhere convenient  and that’s it.

Another interesting powerplant is the electric OpenPPG electric motor, or here:

The question is – can you settle for 60-minute flight durations?  If you can, then the electric option is a real contender.

And while we’re about it, why not opt for the automatically variable pitch prop from Scout Aviation?  30% better fuel economy isn’t something to be sneezed at.

Well, there you have it.  Lot’s to think about, and lots of fiddling, drawing, re-drawing and calculating.  But this is looking very promising.

Most Fleas have a clutter of struts to keep the front wing in place.  This is both messay, unsightly and very draggy.  It’s only benefits are that it is light, and very secure.  Here is a typical example – of a Croses design.

Interestingly. Emile Croses then went on to design a far cleaner looking airplane.
So much neater.  This is how I’ll do things on the Tiny Cedar Flea.  So the question then becomes, how does one build these struts?

And the answer is simple:  build them the way one would build the wing spars.  A box spar built from Paulownia, with a balsa airfoil shape around it, and glassed.  It will be super-rigid, very strong and look beautiful – certainly in keeping with the rest of the Cedar strip construction.

These two pillars are then bonded directly to the firewall.

I’ve been working on the wing arrangement for the reverse pou.  Mmmm  that’s not a good name.  What about Duck pou?  Sounds too much like Duck poo.  Pou Canard?  Clumsy. Axel Pou – after the aerodynamist Axel Darling, who worked out the aerodynamics of this configuration, and showed it to be significantly superior to the Mignet (et al) designs.

Quite apart from it’s rather racy lines, this configuration (in Axel’s words) IS VERY INTERESTING:
Consider the stability of such a planform… the N.P. is going to be much further back than in a standard Pou planform, even equal span, equal chord.

For instance, with 4.4M forespan and 5.8M aft the N.P. is
~1.3M as apposed to .997M for your planform. That’s
41.3% as opposed to 31.65% of total chord – all else in
form, fit and function as a real Pou, not a tandem.

In a complete analysis several things crop up but the most
obvious are foreplane loading is greater, say about 1 lb. /
sq. ft., (Better) but is still mushes at only about 44 km/h
(+20 degrees incidence.) AND it always remains in laminar flow with proper entreplans [ed: interplans = wing gaps vertical/horisontal]. In this case, the foreplane is lifting at about CL=5 in the center and like most other laminar scenarios in the Mignet/d’Escatha formulas, the total lift coefficient for the wing system exceeds CL=2, which is over twice that of a conventional Mignet planform.

An interesting aside is that as lift distribution across the spans is different because in the larger foreplane planform the aftplane is severely and almost equally suppressed from tip to tip, the smaller foreplane leaves lift bumps at about the vortex locations [Ed: at 89% wing span], which when the aeroplane yaws, allows a much greater yaw/roll force and much stronger dynamic lateral stability and much better and immediate pilotage in turns. This undoubtedly is the reason why the mad professor exclaimed that his backward Pou flew so much better than the reverse.

So, the first 25% scale model I’ll build and test will be the “Axel Flea”.

The plan is to cut a fuselage out of 30mm blue foam (of which I have a number of sheets already, and to then 3D print the two front wing supports. (just because I need an excuse to use my 3D printer for something useful).  Standard RC servos, and control linkages to mimic full-sized Fleas will be used.

I’ll CNC carve the wings (NACA 23112 airfoil), and slide in a CF rod for stiffness.

I actually can’t fly RC planes, so I’m going to try to tether the plane to a stake and fly it (with my RC controller) in circleslike the Control Line guys do.  The aim of the exercise will be to observe:
(1) the slow flight characteristics (how does it stall, and what about parachutal descent?)
(2) What effect will fast flight have on aircraft controls?
(3) Using very small capacity batteries, how long will straight and level flight last (i.e. a crude way to measure aerodynamic “clean-ness”).  Axel Darling claims laminar flow of both wings.  This should be interesting to observe with wool tufts to verify on the video camera.

And there you have it.  Let me know what you think.

After a few weeks of intensive research, I have finally narrowed down the design candidates for RC (25% scale) testing.  They are:
The jean de la Farge “Pulga” variant:

Main features:

1. Main wings of equal span
2. Wings same chord (1m)
3. Tail wing also swivels (about 1/3 area of main wings)
4. Both wing swivel (front: 20 deg, rear: 40 deg, tail 60 deg)

Expected flight characteristics:

1. Very low stall
2. 5x speed range (stall/S&L flight)

Axel Darling did pioneering aerodynamic research into the so-called “Reverse Pou” and sometimes referred to as a Canard Pou.  So called because the smaller wing was in the front.  Axel calculated (and verified) that if the Eh and Ev (horisontal and vertical wing gaps) were identical, both wings would experience laminar flow.  In addition he writes: (Pou Renew ed 41)

“the smaller foreplane leaves lift bumps at about the vortex locations, which when the aeroplane yaws,
allows a much greater yaw/roll force and much stronger
dynamic lateral stability and much better and immediate pilotage in turns. This undoubtedly is the reason why the mad professor exclaimed that his backward Pou flew so much better than the reverse”

Main features:

1. Smaller wing in front, larger in rear
2. Vertical and horisontal wing gaps equal

Expected flight characteristics:

1. Superior handling, esp recovery from disturbances
2. Very low drag
3. Slower stall/landing

Keeping to the Axel formula, I’m going to try a low-wing Pou.  Why not?

First, I’ve decided to use the well-respected NACA 23112 airfoil – as recommended by nearly all modern Flea designers.  It is a much thinner airfoil that the NACA 747a315 I had been considering, but it has a much higher max Cl (1.5 vs 1.35), which lowers the stall by 5 kts.

Second, I’ve been agonising over how to pivot wings which need to pass through the fuselage.  Why do they need to pass through the fuse?  Well, if they didn’t, I’d have to secure them to the airframe at three points (at the pivot point, and on the pivot line outboard just like other Fleas.  Three external tubes on each side.  And that would spoil the aesthetic completely.  So in order to make them clutter free, the main spar would need to pass though the fuselage like all other cantilevered wing airplanes.

If there were only one wing, I’d simply close in the D-tube, and that would provide the required stiffness and torsional rigidity.  But check out what would happen at thew rerar wing.  The D-tube would go right through the pilot’s bum.  Not a pleasant prospect.

My solution is to have a super-wide main spar (100mm in fact).  Closed in on both sides, this would form a very stiff replacement for the usual D-tube, and protect the pilot’s bum at the same time.  The main spar will be attached to a 25mm tube, which rotates in two bearings fixed to the bulkheads.  Secure, simple and effective.

The outer panels don’t require this oversized main spar, and will revert to the usual 20mmx20mm spar caps.

If you take a piece of 3mm foam (say) and bend it – it does so easily.  Now bond ordinary paper to each side, and suddenly the foam becomes extremely stiff.  Why is this?  It is because as one tries to bend the foam/paper sandwich, the paper on one side is being stretched, while of paper on the other side is being compressed.  And the force required to stretch paper is quite large.  And since the foam core will resist any buckling effect, the composite test piece becomes very stiff indeed.  Increase the thickness of the foam, and the stiffness increases exponentially.

Wings operate in exactly the same way.  A wing spar consists of two strips of material (called spar caps) separated by what is called a shear web.  This is usually (in a wooden wing) plywood.    And this web need not be massive.  The venerable Cub, for example, kept the two spar caps apart with 1.5mm plywood (1/16″).  Under load, the top spar cap is stretched, while the bottom spar cap is compressed.  If there were nothing tying the top and bottom together, the load would cause the wing to collapse.  But the shear web ties top and bottom together, and absorbs the differential forces.  At some point, the shear web’s ability to resolve the opposing forces will be exceeded and the wing will collapse.

So when it comes to designing a wing, most folks use spruce for the spar caps, and plywood for the shear web.  Plywood has a shear strength of about 2Mpa – and this has proved quite sufficient.  But it is also quite heavy (680kg/m^3).  Paulownia, on the other hand has a shear strength of 5.3Mpa, and weighs about a third that of plywood (260kg/m^3).  It turns out that Paulownia is by far the better material for shear webs.

As for the spar caps themselves, while spruce has been the wood of choice in the past, it is more than twice as heavy as Paulownia (550kg vs 260kg per cubic metre) but has almost identical shear strength.  And so it turns out that Paulownia is again far preferable to spruce for the spar caps.

In conclusion, the Tiny Cedar Flea wings will be built from Paulownia spar caps, and Paulownia shear webs.  A no-brainer, really.

All airfoils tend to nose over – that’s just a fact of life.  It’s part of the aerodynamic mystery that surrounds flight.  So “normal” single-wing aeroplanes have an elevator – a tailplane which can compensate.  The wings lift, the nose wants to dip, and the elevator applies an opposing force.  Result?  The plane flies level.

Essentially, Fleas (like flying wings) need airfoils which have as little pitching moment as possible, because they have no tail-plane.  The way to achieve an airfoil with a low pitching moment is to curve the trailing edge of the airfoil upwards slightly.  The airflow off a wing is directed (initially, anyway) downwards, by about half the wing’s angle of attack.  So if the wing is set at 12 deg, then the downwash immediately behind the wing is about 6 deg.  Why not 12 deg you ask?  Because the airflow UNDER the wing meets the air coming off the top of the wing, and they balance each other out, more or less.  This is why most Flying Fleas set the rear wing incidence to 6 deg – basically resulting in the rear wing being aligned with the airflow.

But this presents a problem, because when the front wing is settled back to (say) 3 degrees in straight and level flight, the rear wing is still sitting back there ar 6 deg – no longer aligned with the oncoming airflow, but now at positive 3 deg angle of attack, and it is lifting “too much”.  This results in the Flea’s tail lifting, and the pilot once again has to pull back on the stick to maintail level flight.

But that’s another issue.  What I’d like to talk about now is the choice of airfoil.

When dear old Henry started out, he got a bit carried away and not only designed a new plane, with a new control system, but he also created his own airfoil.  Letc we be accused of unkindness, let’s just say it wasn’t a great airfoil.

Soon after the initial teething problems with the control system, he switched to a recognised airfoil – the NACA 23012.  Later designers moved to the NACA 23112 which is the same as the 23012, but with more reflex to lessen the nose-down pitching moment.  And until recently, that’s what has been recommended.

Then along came Richard Fraser.  He created a new airfoil for the Flea.  And noone used it.  Actually, not true.  Someone did use it, and it failed miserably to live up to its hype.\

So, are we stuck with the NACA23112?  Not exactly.  The NACA 747a315 is an excellent candidate for Flying Fleas.  It has extremely low drag, extremely low pitching moment, but it also has quite low lift.  So the question is – higher lift?  Or lower drag and pitching moment?

I’ve taken the step to put the 747 into the design spreadsheet.  With 8m^2 of wing area, the TCF will have a stall of 40kts.  Can I live with that?  I plugged in the lift for the 23112, and suddenly I get a 35kts stall.

Mmmm  Time to rethink my airfoil choice, I think.  Maybe those old buggers knew a thing or two…

It’s fair to say that most if not all folks interested in sport flying (that’s you and me) are interested in what’s happening as far as electric propulsion for aeroplanes is concerned.  However, the truth is “not enough”.  Sure, there are a number of light airplanes flying all-electric, but you’d have to sell your house in order to afford them.  Slightly more promising is the option to replace your trusty gas engine with an electric powerplant – and a few of these are becoming available as turn-key installations.

For example, in conversation with Chip at Aeromarine, he tells me that he will have a plug-and-play electric alternative by the end of the year.  30Kw, 18kg, but no price.  Endurance?  No info either.  Or maybe he was just too busy filling orders to have the time to reply more comprehensively to my questions.

Meanwhile, across at Skyleader (based in the Czech Republic): they offer the following info:

Q: How long is the flight time on one charge?
A: Flight times average 1½ hours per charge at a constant cruising speed.

Q: What type of motor is used in the ElectraFlyer?
A: The motor is a custom built DC 18hp (13.5kw) permanent magnet motor.

Q: What type of batteries are used in the ElectraFlyer?
A: Each battery pack is made up of Lithium-Polymer (LiPo) cells which provide the greatest specific energy density of any battery type. The battery packs come in 3 sizes, from 2.6kwh to 5.6kwh capacities depending on the flying style you may choose.

Q: How much do the batteries weigh?
A: The largest battery pack (5.6kwh) weighs 78 lbs (35.5 kg) including all packaging and wiring.

Mmmm… the equivalent of 18hp at 35.5kg?  90 min on a clean airframe at constant speed?  That probably translates to 45 min on “ordinary” planes under “normal” flying conditions.  I’m not WILD about the numbers.  But they do provide pricing:

It looks like for the biggest battery pack, it is going to cost $13,995 USD.  When compared to either the Aeromanine or Hummel v-twin engines ($7,500) producing double the power – I think I’ll pass for now.

According to Denis Carly at Aerolite:
“The complete electric propulsion system, which includes the motor, motor mount, 2 batteries, 2 battery chargers, controller, throttle, cables, etc (everything you need to install and run the system) is $9,750. The complete system with 4 batteries and 4 chargers is $13,950.”

The motor runs off one to four lithium-ion battery packs, each weighing about 35 pounds.  So four batteries will weigh 140lbs (63kg),  Dennis states that “a typical ultralight flight profile, with a full power take off, a reasonable climb and a 40-45 mph cruise will result in about an hour of flight time”.  However, that’s in Part 103 aircraft.  Expect far less in your typical home-built creation.

So where does this leave us?  On the one hand, there is some movement on the R&D side of things, and the first of the plug-and-play electric powerplants are starting to emerge.  The downsides are: at least double the cost, and half the power.

Oh well, it’s fun to dream.

The above  set of sketches illustrate three wing configurations.  The first is what is called a Nanadovic config, in which both wings are fixed.  Prof Nanadovic was a professor at Belgrade University’s Faculty of Mechanical Engineering.  His research was in determining the optimal biplane wing configuration.  He found that “one can see improvements over a monoplane of the same profile of 25% less drag, 15% more lift, and 51% better speed range.”

The Flying Flea configuration (2nd sketch) has a pivoting front wing, with a fixed rear wing.  Generally, the rear wing is set at a 6 deg incidence, because that is the angle of the airflow off the rear of the front wing (at max deflection)

Finally, the Pulga – a design by Jean de la Farge in Argentina a one-time collaborator and partner of Mignet.  He pivoted both wings, and while achieving significantly increased lift, he also experienced significant nose-over torque.  So he added a third wing on top of the tail to counter this.

In all of these designs, the basic wing positioning is the same:  front wing at 30% chord above the rear, and with close to zero horisontal gap between the wings.  What is different is in the articulation (or lack of it) between the flying surfaces.

From my perspective, the Tiny Cedar Flea will provide a very suitable experimental platform because it conforms to the basic wing layout, of all three designs, and with some creative control linkages, could be made to conform to the Nanadovic (fix both wings in place), the Flying Flea (allow the front wing only to pivot) and the Pulga (allow both wings to pivot)

I plan to build a 25% scale model of the Tiny Cedar Flea and try all three configs to compare flight characteristics.

After many teething problems, the rtfm-aero.com website is now live.  It is the reincarnation of a much older site I created a good number of years ago.  Today I started importing pages from the old site to the new.

In the process, I saw that a significant amount of information was now out of date, and I’ll be updating this over the next few days.  But the good news is that the Tiny Cedar Flea now has a home, and a place you can visit to add your comments.

I have also created a Facebook page for the little plane:  https://www.facebook.com/groups/2005021576329962

So please free to pop across, comment, subscribe etc.  If I can get 1000 subscribers, I would be super-pleased.  I plan to get a dedicated YouTube channel also, because starting in 2025, I plan to video document the building of the Tiny Cedar Flea in 15 minute weekly episodes.  Now that is both a challenge and I’m sure it’s going to be a lot of fun.  Right from building the formers, to the first test flight.

Regards,

Duncan