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

[quorneng]

OwenN
To be perfectly honest I think you have so many variables in such a design and some of which are fundamental that the chances of it even flying properly are not good.
If you want a tail sitter take off then I would suggest you start by designing a simple plane (thrust exactly in line with the CofG and frontal area) with a configuration that suits being a tail sitter and fly that conventionally first (with a hand launch?) without installing the high thrust to weight ratio needed for a vertical launch. Very high thrust to weight creates their own problems.
Only when you know the plane flies well start considering the necessary changes to achieve vertical launch and any extra electronic stability that might be required.
It can be done - I have done it!

 

[leaded50]

have you ever seen a jet, with wing/tail control surfaces direct behind the "exhaust" outlet? Front elevators close would restrict air into the EDFs, aka loose power.

 

[OwenN]

OwenN
To be perfectly honest I think you have so many variables in such a design and some of which are fundamental that the chances of it even flying properly are not good.
If you want a tail sitter take off then I would suggest you start by designing a simple plane (thrust exactly in line with the CofG and frontal area) with a configuration that suits being a tail sitter and fly that conventionally first (with a hand launch?) without installing the high thrust to weight ratio needed for a vertical launch. Very high thrust to weight creates their own problems.
Only when you know the plane flies well start considering the necessary changes to achieve vertical launch and any extra electronic stability that might be required.
It can be done - I have done it!

I have made some further improvements to enlarge the new rear wing and reduce the main wing.
This means less lift loss and trim change on a main wing stall.
The only downside I can see are slot effects on the rear elevons, but several planes through history have offset
ailerons with no ill effects (JU87?)
What is your VTOL plane? Any pictures?

Another possible downside of this design is twitchy behaviour on conversion to hover, and at launch.

With a little auto-pitch damping, and attitude stabilisation on no control input, this should be manageable.

 

[Tench745]

For a canard aircraft to fly well the canard needs a higher wing loading than the main wing, and should be designed (airfoil, aspect ratio, angle of incidence, etc) to stall before the main wing does which will drop the nose and let the aircraft recover long before the main wing ever stalls.
In light of this, I'm not sure why you're worried about a main-wing stall.

 

[OwenN]

have you ever seen a jet, with wing/tail control surfaces direct behind the "exhaust" outlet? Front elevators close would restrict air into the EDFs, aka loose power.

1) I think ducted prop is a bit more gentle on control surfaces than jet exhaust.

2) Some model designs have thin control slats mounted right in the duct.

There will be a little thrust restriction at high elevator angles, but most of the inlet air will follow one side of the elevator.
It is hard to put a number on potential thrust drop- I would guess 20% maximum?

If vertical thrust is 2.5 the weight, this would drop to
2.0 times the weight. This still gives a good lift margin. Testing will tell.

Flow "pinch" at one edge is a symptom. any control slats should avoid the parallel edge of the duct, and the duct should be more square
at the control slats. The slats are also a bit draggy, and even the thin profile will occlude the duct a bit, which should be widened locally to compensate.

One of the Moller flying cars used slats at the exhaust end, bending flow through 45 degrees. That is not good.

Also, he was using rotary engines and multiple props in too small a duct. Very inefficient! If one prop won't do it, more props don't help!

Possibly fully swivelling turboprops would work, in bigger ducts, but would jump the machine right out of his target price range,
and be very noisy.
I am sure he knew it wouldn't work-he lectured in aerodynamics after all! - fraud???
I think the picture is of a flying model-those flew alright-It just doesn't scale up to full size!

 

[OwenN]

For a canard aircraft to fly well the canard needs a higher wing loading than the main wing, and should be designed (airfoil, aspect ratio, angle of incidence, etc) to stall before the main wing does which will drop the nose and let the aircraft recover long before the main wing ever stalls.
In light of this, I'm not sure why you're worried about a main-wing stall.

I am deliberately not designing a conventional canard.

There are provisions for partial lift and control surface authority when all wings are stalled.

This should only happen at a high angle of attack, when the aircraft will hang off the prop, anyway.

The only dangerous condition is if you try to fly slowly at a low angle of attack, close to the ground.
You will just dive into the ground when the front wing stalls. You have a similar situation with a conventional wing layout.

 

[OwenN]

I have found a partial solution to the cog being above the thrust line.
Overall the proposed fuselage size is very deep. 220 mm or 8.7 inches by 130 mm wide, or 5.1 inches.

Putting the battery in an "external store" under the front wing looks good.

It is only as deep as the front wing itself. 30 x 42 x 136, or 1.2 x 1.7 x 5.4 inches, and looks in proportion.

The main fuselage will be mainly empty air, so the supporting side walls can be quite thick.

Getting the thrust more in line means it will not drift forward so much on landing.

It is not required to make a perfect tailsitter landing.

 

[OwenN]

Another possible mod for ducts is "front skirts".
Straighten top and bottom edges, and have reverse-hinged flaps on the top and bottom edges.
They can be linked together for stability.
You can also have fixed, shaped side skirts, providing a square entry.

I have seen these on the outlet end, but they should provide up-and-down force on the inlet end, too.
maybe they should be curved and tapered together as well.
The curvature provides a sort of variable Coanda effect.

 

[quorneng]

This is my tail sitter and its true scale as well!

A single 70 mm EDF in the tail and an 1800 mAh 4s in the nose.
With a mid wing and cruciform tail surfaces no issues with the thrust line, CofG and frontal area.
I built a slightly smaller non VTO version first to make sure it could fly and be controlled. I even hand launched the big one at much reduced power for its maiden flight.
At full power it has thrust to weight approaching 2:1 so takes off pretty fast.
It has a 3 axis gyro but operating only in rate mode so it is not stable in hover and relies on the aerodynamic control surfaces. The gyro simply limits the degree of any external influence but retains the full RC control input.
As a plane it actually flies pretty well. It can loop and roll with ease.

 

[OwenN]

IanT said:

Looks like the 9x5 triple blade prop is a good option with 60 amp ESC's you should have more than enough thrust there with two motors ?
and 4s lipo's would be ample.
[end quote]

1) How do you find these bench test charts?
I can't find any on Google.
2) what was the 8 inch prop mentioned? = the chart is scrambled. It seems to give more thrust and less amps.
3) what effect does varying number of blades have on thrust and amps?
[end quote]
Looking at the chart again, the 8 inch prop gives more thrust and less amps. Any idea what the prop was?
Should I reduce my duct size a bit, and would I still get enough thrust?

 

[OwenN]

This is my tail sitter and its true scale as well!
View attachment 187569
A single 70 mm EDF in the tail and an 1800 mAh 4s in the nose.
With a mid wing and cruciform tail surfaces no issues with the thrust line, CofG and frontal area.
I built a slightly smaller non VTO version first to make sure it could fly and be controlled. I even hand launched the big one at much reduced power for its maiden flight.
At full power it has thrust to weight approaching 2:1 so takes off pretty fast.
It has a 3 axis gyro but operating only in rate mode so it is not stable in hover and relies on the aerodynamic control surfaces. The gyro simply limits the degree of any external influence but retains the full RC control input.
As a plane it actually flies pretty well. It can loop and roll with ease.

What is it a model of? It looks like a rocket-powered cruise missile? How does it balance? Are there steering vanes in the
edf outlet duct?

With my proposed layout, there will be issues with thrust-pitch linkage, and it will descend at an angle with some drift,
but I think that is acceptable as it will be a belly lander, and only drag the tail for a short period.

Looking at the motor bench test results again, there seems to be opportunity to reduce the duct diameter
and still get over 2 kgs thrust per motor. but how low can I go???
This would help line everything up better.

Otherwise, the top wing has to be curved or built around the duct, which is tricky.
I don't want a fuselage hump between the wings. It doesn't look good. An inverted "W" shape would be OK.

The wing surfaces get 3D, and are hard to skin properly, plus the curve is a stress raiser of unknown proportions.
Possibly needs to be 'Glassed over??

A straight wing, and a simple inner fiberglass sleeve for the duct, and built-up , planked balsa outer, seems to offer fairly
lightweight construction.

 

[OwenN]

Here is the fuselage revision- dropped 105mm (4.1 inches) relative to wing surfaces.
Also the first showing of my 1/5 scale drawing-layout. The curved section will have radial ribs, and 5 full depth spars,
up from 3 in the main wing.
Rib height is set to allow planking with 1/16 balsa and sanding back, to match up with 1/32 sheeting on the rest of
the wing.
Stepped lap strips to be used.

 

[quorneng]

OwenN
It is a model of a prototype V-2 rocket built right at the end of WWII.

It was flown twice. First failed on launch, second failed at altitude.
It balances at 50% of the root chord.
There are small vanes in the exhaust (as in the full size) but they have little effect once at flying speed compared to the aerodynamic power of the tail control surfaces.

 

[OwenN]

Very Nice!

 

[OwenN]

I have worked out a construction and assembly method for the model.

1) Do the prop shrouds. Make disks, 1/16 covering, glass over, knock out disks.
2) Do the front lead-in profile for the shrouds freehand, from solid. Maybe use some foam for ease
of sanding and shaping.
3) Lay the lead-in side flat on the pinboard. Assemble the fuselage and both curved wing sections vertically,
with flat end frames.
4) Fill in the outer surface of the prop shrouds/ducts.
5) Do the same with rear, front wing section, nose, and glue bulkheads together-doubling them.
The bulkheads need to be well integrated into stringers, shaped keel pieces, and skin.
Thus they won't tear clear first, under load, when bonded face-to-face.
Build the tailplane and front wing flat, first.
6) Build the outer wing panels flat. Don't skin the wing tops.
7) Build outer wing pods, top, and bottom fins, separately.
8) Assemble wing outers to center section, assemble fins to pods, add pods to wings.
9) Thread up flex-push cable and add rudders.
10) Build motor pods, motor mounts, and pylons out from the sides of the fuselage. Cables can be added after surface finishing.
Keep screw heads visible for assembly.
11) Finish skinning the wings and center fuselage.
12) Build other sections in a similar way.

For now, all wing and tailplane panels are built flat, with the rear edge flat on the pinboard. Tip pieces are added last.
Leading edges are propped up off the surface.

How are symmetrical wings assembled? Are the trailing edges blocked up clear of the pinboard?
I have several full-depth web spars planned. some will need to be propped clear of the surface, too,
and the rib sections suspended from the spars and trailing edge.
I am thinking, that there is no need for cambered wing panels with the large amount of thrust,
and camber probably will interfere with 3D manoeuvres .
What do you think? Is this a good wing building method?

 

[Tench745]

How are symmetrical wings assembled? Are the trailing edges blocked up clear of the pinboard?
I have several full-depth web spars planned. some will need to be propped clear of the surface, too,
and the rib sections suspended from the spars and trailing edge.
I am thinking, that there is no need for cambered wing panels with the large amount of thrust,
and camber probably will interfere with 3D manoeuvres .
What do you think? Is this a good wing building method?

Symmetrical wings are assembled exactly as you suggest, the trailing edge is blocked up with any wash-in or wash-out twist that will be built into the wing. You can either pin shims to your building board, or cut ribs with the standoffs included, then cut/sand away the standoffs when you flip the wing over to work on the other side.

 

[OwenN]

News from other threads:

1) motors are now 2806.5, 6 per, ganged in sets of 3 per prop.
Props are back to 9x5.
this gives 128 A for static thrust each prop, up from 54.
This seems a bit odd, given the 3 motors are running at a more efficient speed at static loading.

News from other threads:

1) motors are now 2806.5, 6 per, ganged in sets of 3 per prop.
Props are back to 9x5.
this gives 128 A for static thrust each prop, up from 54.
This seems a bit odd, given the 3 motors are running at a more efficient speed at static loading.

<Rambling number spouting follows-ends at next dotted line>
--------------------------------------------------------------------------------------------------------------------------------------------------

Minds you, this compares the 1800KV motor that could run the 9x5 at around 13000+ rpm (of 28800, 45%)
The loading sounds odd - normally motors won't go below 65% of KV rpm??? that would be 18720 rpm
Maybe rpm was as that high, with 2016gf, 867w, 54 A.
Assuming 48% efficiency at that point, =416 w real at this loading.

Compared with a single 1300KV, which wouldn't run the 9x5 at all.

further 1800KV specs - drop w div by 1.3 = 320 wats "rated" (by me, 77% point)
amps should be / 1.5 = 36 amps. rated.
a check is the Kv/Kt calculation.
Kv = 1800 x 2 x pi / 60 rad/volt-s = 188
Kt = 1/Kv = 5.3 x 10^-3
and Torque = A x Kt = 36 x 5.3 x 10^-3 = 0.19 N-m
Then power = rev x torque, def 77% rev = 22,176, in rad/s = 2322
pwr = 0.19 x 2322 = 441, so my amp calc may be high or my efficiency calc may be low.
This is higher than my assumed real power at full load! - efficiency may be a little higher, say 52 % at that point.

That can be iterated out, depending on only 2/3 of the amp change being related directly to power;
the rest is due to power factor, or inductance.
(may be more like 77%??)
change from 85 % effy to 48 % effy = .37 (plotting effy change vs rpm drop)

so if power rises from est 320 at 85 % efficiency = 376, to 867 at 48 % - increase by 491, or x 1.3(?)

and the real power increases by 1.3, from 320 to 416, then 416/867 = 48%, as predicted.

However, this leave A at rated being high.

another check is V x (2/3)A = 16 x 24 = 384. -this is in between the two amounts - 320, and 441.

then 1/3 of 36 amps = 12 A, 2/3 = 24 A, 24a x 1.3 = 31.2

add back the 12: = 43. 2A, lower than the 54A maximum overload expected.

If 43.2 is used to calculate rated amps, /1.5 = 28.8,

28.8/32 = 90%, and watts = 441 x .9 also, = 396w, this may be a better estimate overall. - the wheel is going round and round, but
the output estimates seem to be converging.- this depend on my observed amp ratio of 1.3 to 1 holding??

that is not 2/3 , 0.667, it is 0.77 - so that could be factored back into the efficiency loop, to... on and on...
where does it end? nobody knows..
this 77% seems to be coming up a lot though! (rated rpm = 77% nominal max. rpm- my estimate)
<end of ramble>
--------------------------------------------------------------------------------------------------------------------------------------------------------

2) Prop ducts are removed, to allow for the fatter motor cowls 106mm tapering to 70mmn opening.
3) The main wing is dropped to be in line with the prop thrust lines, and the front wings.
The fuselage is tall enough to mount gear above and below the wing spars.

4)- The 1/5 drawings have been modified to suit.
All my drawing pens are now stuffed by clogging them with whiteout.
I will buy a couple of each .03, 0.5, 0.8. The fine-tipped taper felt pen still works OK. I will see where that came from.


5) Suppliers are tee'd up for motors,-(iflight direct, $260 nz, free delivery-slow boat??), batteries, cowls, collet prop drives, micro belt drive parts.
These will be on my next "to buy" list,
then I had better get some epoxy resin.
(and bearings, shafting, shaft collars, roll pins etc)

6) Next drawing job: break out layers and dimension motor plates. I have a scale drawing.

Note:
In converting DC to AC, 1/3 of the AC amps does not appear at the battery (motors)
so battery A = 100, and a 30C battery is OK.

I have selected 3 batteries:
2 x Leopard Power 2200 4S 30C , 228 g, $70 NZ.
1x 450 mAh Tattu 3S 45C 42g $24.50

Totals : 488g, $164.50 NZD - a little more expensive and heavier than
the original setup, but this one should actually work! (prev = 320g, $99nz)

Motors have gained weight, too, from 100 g to 500g total- assuming 100g
per prop for mechanical parts.

This means overall weight rises to 1968 g.
Total thrust = 2145 x 2 = 4290 g, thrust to weight ratio = 2.17:1
Still OK for a vtol. I now need to recheck wing area and oz/sq ft.


This is now 54 oz/sq ft. - 40 mph stall speed.
A normal RC model plane would need flaps to land at this loading.
With a 92 mph projected top speed, there is leeway for manoeuvres.

A tailsitter-bellyflopper should still be fairly easy to land, with gyro autopilot assistance.
(Auto throttle trim, tail-down transition, attitude, location damping.)
--------------------------------------------------------------------------------------------------------------------------------------------