wing loading, span, flying speed, landing speed, charts.

[OwenN]

What, apart from egg-crate, balsa, skin, half-fuselage sections, cutout keel sections. assemble fuselage from sections, wing build flat, one piece?

I am keen on balsa and ply. I thought I would try epoxy coating, as well, squeegeed on under wax paper, to fill the grain and provide a good sanding surface and a bit more stiffness, plus iron-on lightweight monofilm to give some color.

- I need to track down a small iron, and read up on ironing monofilm on to foam.- that should be similar.
For the load bay, I am planning some laminated arches to support the sides, and standard pot-lid style slot-in curved covers, screwed in place with epoxied-in nuts.
- a suitable thread cleaner would be good.

The front wing will be screwed down as well, with a forward rudder threaded through the fuselage and actuator mounted as a unit to the front wing. screw through brass tubes epoxied into corner spreaders, boxed in to spread the loads.

I got a bit of stick about the forward rudder, but I think if it was good enough for Glenn Curtiss, it should work for me.
Anyway, the widely-spaced rear rudders make them a bit suspect for rudder effectiveness. There
is a reason aircraft have their rudders closer to the centerline. They are there as part of the VTOL landing legs.

The rear actuator pair is to be inserted into the wing cavity, facing each other.
Turnbuckle and clevis actuator rods.
As you can see, I have been thinking about this a bit.

Control stabilization, and elevon multiplexing is to be handled via a flight-controller module, running
Ardupilot firmware. That is free and open source, and you flash it in to the flight controller.
It also provides all the "nannies" you would expect with a high-end quadcopter,
as well as center-stick throttle hold, center-stick flight regulation, control override at small inputs to use
as flight-path management instead of direct control of surfaces, and VTOL motor management, thrust
balancing with two props, motor turn assist.

As you may expect, to learn this thing, I will be turning control sensitivity way down, and mainly
letting it fly itself, and doing circuits and touch-and-go.

 

[Timmy]

I found a PDF online for that

Could you share the link?

 

[OwenN]

https://rclibrary.co.uk/files_titles/1433/RC_Model_Aircraft_Design_ALennon_partA.pdf

 

[Timmy]

https://rclibrary.co.uk/files_titles/1433/RC_Model_Aircraft_Design_ALennon_partA.pdf

thanks

 

[Tench745]

I got a bit of stick about the forward rudder, but I think if it was good enough for Glenn Curtiss, it should work for me.
Anyway, the widely-spaced rear rudders make them a bit suspect for rudder effectiveness.

For clarity sake, I believe you mean elevators. I'm not aware of any aircraft with rudders in the front, not even a Curtiss design.

No, that didn't look right. -flattened the tail down to 12 degrees, nose now asymmetric, with a greater curve on top, sharper nose, and furthest projection now down to 7mm up from the flat base. Is there a way of predicting how a random airfoil shape will fit in to the more standard ones? -- I will continue reading Lennon.

Look into the NACA airfoils. The number it is identified by defines the features of each airfoil. Figure out what airfoil matches your idea/ideal and compare.
You can get airfoil data from http://airfoiltools.com/airfoil/index.
There is also a program, x-foil, which will allow your to plot your own airfoil shape and generate polars for it. It's a little beyond my understanding, but may also be something you'd find interesting.

Last thought, The simplified formula for Reynolds numbers given by Lennon is as follows: Rn=speed in mph x chord in inches x K. K refers to the viscosity of the fluid.

 

[OwenN]

OwenN said:
I got a bit of stick about the forward rudder, but I think if it was good enough for Glenn Curtiss, it should work for me.
Anyway, the widely-spaced rear rudders make them a bit suspect for rudder effectiveness.

Q:
For clarity sake, I believe you mean elevators. I'm not aware of any aircraft with rudders in the front, not even a Curtiss design .
A:
I had a good look through. Some of the earlier designs has a little sail-like rudder at the front. He dropped that for later designs.

There are a number of other early aircraft with front rudders, but his were the only ones , I think, that were good fliers
and commercially successful. The variation was still popular after specially made aircraft engines became a "thing".

It should still be stable if the fixed rear fins are enlarged to suit. At high yaw angles, the rear fins overpower the front ones.
The alternative is to mount them right at the rear, in the center of the main rear fins, and between the horn links for the elevons.
The rear body can be lengthened and widened to suit, rather like the Concord tail.
There is plenty of ground clearance with the VTOL legs.
I thought that it made sense to put the rudder where there was plenty of room, and it is not fighting COG location and weight.
I also added an image of a later prototype WW2 era ascender XP 55.
Looking at the Curtiss aircraft, that front vertical surface was not intended to be a rudder, as it is firmly attached to the forward
elevator, but it could be used that way in theory.

 

[OwenN]

I have thought about the front rudder, and it would be over-sensitive. - any yaw adds to the rudder angle. I will investigate
moving it to the rear.

 

[OwenN]

Latest mods for rear rudders-2 views. note-this is still the 24 inch wing, not the latest 34 inch wing.

 

[Hondo76251]

I think you've spent more time on this drawing than I have on any plane ive ever built! 😂

I usually prototype with a pencil, ruler, glue gun, and a piece of foamboard... but to each his own i guess.

Youve mentioned gyro/flight controller, what one do you have in mind?

 

[Piotrsko]

He has spent more time engineering this than total time of anything I have ever had. Not a critique BTW.

I suggest a cheap rc car and a elementary airplane to learn operations on, else the first flight will result in a rather immediate selection of rebuildable materials forthcoming. A gyro and computer flight control system will not compensate for commanded flight into other things.

 

[OwenN]

I think you've spent more time on this drawing than I have on any plane ive ever built! 😂

I usually prototype with a pencil, ruler, glue gun, and a piece of foamboard... but to each his own i guess.

Youve mentioned gyro/flight controller, what one do you have in mind?

Matek 765-This is quite well featured.

He has spent more time engineering this than total time of anything I have ever had. Not a critique BTW.

I suggest a cheap rc car and a elementary airplane to learn operations on, else the first flight will result in a rather immediate selection of rebuildable materials forthcoming. A gyro and computer flight control system will not compensate for commanded flight into other things.

The important thing is pre-flight checking that all controls are wired correctly, and work in the correct direction.

I plan to learn it like you would an actual aircraft. Low control sensitivity, tons of altitude, no trees, no wind, flat terrain, liftoff, circuits, landing approaches, touchdowns, go-arounds. With the vtol and auto-throttle, it should be easy to slow down and accurately land.
I have spent some time on a quadcopter sim, and I will now practice on an RC plane sim.
The logitech gamepad has a similar layout to an actual transmitter-controller.

the RC plane sim is actually harder to use, as it is a jet,
and lands fast and flat-good opportunity for bounces, flips , and other rough,potentially destructive landings.

 

[OwenN]

Latest up-to-date design:
1) mash-up 1:5 and 1:4 scale drawings.
2) Reduce fin side area.
3) Draw front view.
4) Lift thrust line for better tailsitter use.
5) lengthen nose and add fuselage droop to reduce drag and to improve looks.
Re: lots of thought, rework, time taken to design. - I am learningas I go, so come up with a lot of ne ideas.
It is different if you have already made multiple RC planes and flown them.

(this text editor is really not working well-re-edit and inset is playing up. This is a temporary machine, while my new one is replaced.)
The views started off as the same scale. this has changed a bit with the scanning process.

 

[OwenN]

Editor problem solved! It was a caching issue-I just needed to refresh the page twice.

I now have a new thought about gyro-stabilizing of pitch.
The control surfaces are quite large, and any stability re-trimming is likely
to be very small. Can the actuators handle fractions of a degree trim adjustments?
Full size , large supersonic aircraft often have multiple elevons each side to address this issue.

-See Concord - there are 3 elevons each side. I don't think the model flight controller divides up actuator
outputs to that extent.

How do standard actuators work-is it a form of analog-digital conversion?

I have no Idea how modern RC control systems work.

I now know the main transmitter-receiver set is 2.4 GHZ frequency skipping, and has basically an analog signal?

What determines rate of actuator movement and end position normally?

Does the final position echo the control stick position?

I also know that with a flight controller you can set control sensitivity, and ramp motion up to a higher rate as the stick is moved more.

Re: STOL launch control. I have increased the front wing elevator area, to take advantage of air being sucked in by the props.
I will set COG right on the calculated AC.

Any pitchiness can be dealt with by the Gyro stabilizer. I think it includes auto-levelling and flight path smoothing.

STOL launch should take place with little or no forward motion before liftoff. The velocity and volume of air available
allows instant flight capability.-
Over 45 mph , covering 60% of the span.

Keeping the nose up could be the only problem.
I was thinking of allowing controlled blowback up to
an angle past 45 degrees, to help stabilise the effective AC.
The full-up elevons should help pin the tail end down while rotation is happening.

There should be significant pitch,roll,yaw instability, to be controlled by the flight controller doing its gyro smoothing thing.

 

[Tench745]

Sounds like you may need to do some more internet delving as I can't fully answer your questions.
But I can give you this short answer with what I do know: Modern radio systems are a digital system. The signal from Transmitter to Reciever is digital. PWM (pulse-width-modulation) signals are sent from the receiver to tell the servos where in their travel they should be. Modern radios are a proportional control system, so the more stick you give the more the servo moves.
The flight controller adds another layer to this. It takes signals from the receiver and reinterprets them with it's own programming to then tell the servos how to move. It can do things like taking your transmitter stick position and using that to command a roll rate from the aircraft (with all the feedback loop necessary to do so) instead of just a servo position.
You can program multiple different outputs from a flight controller (depending on the flight controller) so you could have three different elevons all actuating differently if you wanted to and could figure out how to program it. You don't need them though.

 

[OwenN]

Thanks. People talk about the video signal being analog, but don't say anything about the control signal.

 

[quorneng]

OwenN
I design and build my own planes so I have some observations on how you are proceeding.

1. In terms of your plane layout why have you got big vertical surfaces on each end of the wing? They would appear to have much more area than is needed to maintain lateral stability.
Does any full size aircraft have such big vertical surfaces positioned like that? I can't think of any unless it was designed to only take off and land vertically.

2. At 56 oz your plane it going to be quite heavy for its size and as you have calculated will have to fly quite fast. Have you thought about its power requirements? Power has a weight penalty.
Speed then poses more questions. Will it be able to withstand the aerodynamic loads? Will it be controllable? Will it survive a landing?

3. The 'Viggen' type configuration was chosen in full size for particular performance characteristics, which includes super sonic speed, not for its natural stability. There are good aerodynamic and structural reason why the vast majority of planes look the way they do.

4. It is possible to build and fly a model plane with an 'exotic' configuration but it is not easy and is likely to generate unique problems that have to be solved along the way.

Designing a plane can be very interesting but to have any chance of success it is always advisable to work from a known 'workable' start point.

Finally unless you have quite some experience in flying an RC plane how will you know what features are likely to make it a good or bad flyer.

Keep posting. I will follow your progress with interest.

 

[OwenN]

The vertical surfaces could be reduced still more. I already chopped a fair bit out. They are only really streamlining the tailsitter legs, and could be reduced easily by 30 %.
The loading on actuators is a bit questionable.
I don't think the standard 2 kg/cm units will stand up to aggressive use.

The structure should be enormously strong.
I am paying a bit of attention to stress raisers.
I doubt it will tear the wings off-for example.
They have good section moments.

It has been specced up with more than enough thrust and battery for VTOL launches, and about 10 minutes flight duration.

I originally started light, but gravitated to 15 amp motors and now 50 amp motors, gradually getting heavier.

I compromised on a normal stall speed of 40 mph, cruise 65 mph, max around 90 mph.

The vtol capability and the high "nanny" factor of flight controller/ardupilot should make it easily landable without dramas.

The main traps are-losing sight contact, flying into trees and other obstructions, and over-confidence in manoeuvering too close to the ground! All happening at a fairly high speed for a rank amateur...

It shouldn't need expert-level controlling for simple maneuvers like take-off, circuits, and landing, due to the gyro-stability features.

I think I can select auto-recovery features like range and height limits, auto-circle, return-to-base, with/without signal,
low battery actions.
The only major possible cause of loss/crash will be component failures.-apart from gross stupidity! ,


It should handle somewhat like a mildly aerobatic pylon racer, scrubbing off a lot of speed in turns, but building speed quickly, too.
It is not going to average over 110 mph in circuits, though!

I now have to stop speculating and get drawing. A full size set of drawings and patterns will take me a while to generate.

 

[mastermalpass]

I assume you have telemetry hardware and other monitoring devices such as ardupilot, to gather data on the performance of the real thing? If there's one thing I can tell you about air at this small scale and relatively slow speeds is that it's very inconsistent.

Think how when in a big passenger liner going 500mph, turbulence typically comes as a vibration. Whereas if you've ever rode a small 4-seater prop plane at 130mph, you find turbulence comes like ocean waves. Scale this down even more and you will struggle to find true airspeed conditions that are consistent enough for you test your hypothesis numbers accurately.

In my experience (and also the experience of the flight instructor I went up in that small plane with), flying is more about getting the feel than getting the knowledge.

Build a concept plane and launch it. Glide-test if you want to be really safe, but get it in the air - ideally with no gyros if you want to examine the behavior of the airframe - and see how it truly flies. The performance of the plane will show you what needs adjusting.

The numbers can come back when you want to maximise its efficiency; the basics can be done by eye. Speaking of doing things by eye - how do you plan to track its speed over ground and true airspeed?

 

[Piotrsko]

I tend to use control surfaces that do not exceed 10% of the total surface area.
The servos are following a pulse train from the receiver that it corrected by comparing that pulse train to the internal servo generated pulse train via a device ( variable pot or cap) attached to the output arm. I guess a 1msec response.

Since the general public doesn't do technical specs well, that metric is reported as degrees per second rotation. Higher response means faster servos. 20 lb-in has been adequate for most of my stuff, although higher torque is available it needs other considerations because TANSTAAFL. @quorneng has many valid points.
People that engineer things for idiots are vary unaware the ingenuity idiots posess.

 

[OwenN]

I assume you have telemetry hardware and other monitoring devices such as ardupilot, to gather data on the performance of the real thing? If there's one thing I can tell you about air at this small scale and relatively slow speeds is that it's very inconsistent.

Think how when in a big passenger liner going 500mph, turbulence typically comes as a vibration. Whereas if you've ever rode a small 4-seater prop plane at 130mph, you find turbulence comes like ocean waves. Scale this down even more and you will struggle to find true airspeed conditions that are consistent enough for you test your hypothesis numbers accurately.

In my experience (and also the experience of the flight instructor I went up in that small plane with), flying is more about getting the feel than getting the knowledge.

Build a concept plane and launch it. Glide-test if you want to be really safe, but get it in the air - ideally with no gyros if you want to examine the behavior of the airframe - and see how it truly flies. The performance of the plane will show you what needs adjusting.

The numbers can come back when you want to maximise its efficiency; the basics can be done by eye. Speaking of doing things by eye - how do you plan to track its speed over ground and true airspeed?

Answer:
Once I get the FPV gear running, I can view the telemetry. For true airspeed, I have the details for a pitot tube module. It shouldn't be necessary, as there is a lot of gyro stuff that will rescue the model from a stall. It may stop suddenly and go into hover mode, however.

Glide testing is a no-no. I don't expect the plane will be at all stable without power and speed. a fast plummet to the ground tells me nothing useful.
I am quite comfortable with being all gyro-ed up.
The main stability questions will be if the stability routines give a good result, or if there are wobbles.
You do a similar tuning process with Segway-style vehicles.