Thread Tools
Sep 22, 2021, 02:32 AM
Registered User
Thread OP
Discussion

Proposal for EDF VTOL.


This thread replaces the one on Three Flight Planes( canard, wing, elevator)- same basic aircraft, different approach.

Will this work?
Do I need the support jets to be more widely spaced out, or to use supplementary nozzles?

I am looking at 5kgf thrust, 100mm EDF unit, 3kg takeoff weight.

What are the losses likely to be, going for pressure over velocity, then blowing out through gimbal tubes?
Would curved deflector vanes be better?

How could I use conical lead-ins with bendy tubes, to gradually accelerate the flow?
How should I handle the tailpipe diversion?
I could use flat channels and flapper plates. Is that better?

how can I supports the gimbal balls in a low-friction way, with the applied pressure?
Would they have to be specially fabricated, highly polished steel on PTFE sliders?
How would I grip them for turning?
What is the likely swept area fraction of the 100mm EDF?-do I need 90mm equivalent diameter?
That would be 4 x 45mm tubes.
Last edited by Owen_bern; Sep 22, 2021 at 02:38 AM.
Sign up now
to remove ads between posts
Sep 22, 2021, 03:32 AM
Registered User
Thread OP
Version with 6 gimbal balls and 2 side vents:

1) Seeing as side deflection would be limited, horizontal gimbal frames could be used, and end bearings.
Air seal can just be by close fit, with nice, spherical shapes.
I would need 1 rotary servo and one lever servo per ball.


2) Ball diameter is x 1.3 x the hole diameter, or around 52mm diameter. (2.1")
Maybe "bauble" globes could be used?

3) Some tube taper could be included within the ball outline.

4) If it is desired to fully shut off the ball vent, the tubes need to be within the ball shape.

5) With 6 ball tubes, each one can be less than 40mm in diameter.

6) The balls are not evenly distributed wrt. the COG, so they need to be different sized holes.

7) Partially occluded exit holes may not be very efficient, either.

8) It is probably best to shut the balls off, as the tail vent will be more efficient for forward thrust.

9) A link to an airspeed sensor is needed for auto-changeover from hover to forward flight.

10) 63 oz/sq ft is a bit big. - area can be increased to drop wing loading below 50 oz/sq ft.
Transition speed is likely to be over 40 mph forward speed.

11) If the full 6s 120a 100mm thrust can be utilised, top speed is likely to be 70 mph +
Last edited by Owen_bern; Sep 22, 2021 at 03:57 AM.
Sep 22, 2021, 05:20 AM
Registered User
Thread OP
Revision 2: Scale up by x 1.2, use 2x 90mm EDF units.

I have these on hand.
6s, recommended 100A esc.

I will check this on eCalc.
<edit>
Total thrust = 4670gf. by eCalc,
3000g est launch weight.
345 sq ins wing area.
Very long fuselage at 86.6"
span = 47"

Het typhoon 650-168-1500, sunnysky 90mm edf.
this is roughly an equivalent setup.

Stall speed = 33 mph
top speed = 123 mph.

Thrust/weight = 1.56 static
A = 85
Quoted thrust was 3.6kgf, I think per unit, so 65% - untruths??

Is this enough to work the lift jets, and transition to forward flight?
Last edited by Owen_bern; Sep 22, 2021 at 05:58 AM.
Sep 22, 2021, 06:52 AM
Registered User
Thread OP
Fuselage revision to get 2x 90mm EDF units side by side:

This version is better, because the fans are not masked by the gimbal balls.

This is 210mm across the lead-in rings.
At this scale, the difference is quite small.

However, this is a very large fuselage at 84 inches long.

With the main mechanicals at 1500 + 640 = 2140g,

can the airframe and other parts be made to 760g??

The retract units alone are 60-odd grams ,including tyres, each, I think. x 3 = 180g

There is a fair bit of mechanism in the gimbal balls, maybe 30g each x 6 =180g

add 5 extra servos, probably 5kgf units at 20g each??= 100g
Plus escs, receiver, flight controller. - another 120g
equipment comes to 180 + 180 + 100 + 120 = 580g, remainder = 180g.
Possibly we need to add another 500g to the launch weight, to 3500g.

I think we need about 1.4 actual thrust to weight, to successfully transition to normal flight.
Sep 22, 2021, 05:35 PM
Registered User
Thread OP
Version with 4 ball vents and 3 thrusters.
----------------------------------------------------------

Possibly some weight saving methods can be used on the first version.

1) Non-retract, non sprung UC.
2) Polyester film skinning, with 1/8" balsa stringers.
3) fully built up tailfin and elevators.

A revised version could use more powerful EDF units.

Other vent styles may work better at an angle.
Slats are not much better-they get more restrictive at an angle.

Had anyone seen flexible tube vents?- hopefully less complicated than jet engine bendy thrust tips...
I think they may need to swing around 40 or so degrees before the tail vent can start opening.

Though a helicopter can move forward quite fast with less tilt.- maybe 20 degrees?

Tube vents would need a separate shutoff mechanism- butterfly valve?
It is balanced, and would work with a small servo.

Can the shutoff valve servo be mounted on a tilting tube?
Maybe a smaller ball, with a projecting tube, to seal air leakage- just leaving a tiny gap.

I need to find an appropriate 6S thruster motor, with fairly low KV - maybe 900KV?
I will try some out on eCalc.

This model is rather big, but it uses stuff I have on hand.
Also, extra the vents are quite small, relatively, giving a tidy appearance.

The side thrusters are not on the COG.- does this cause control problems?
They could be moved.
My wheel stowage scheme uses quite a thick wing with the wheels back past the middle of the wing.
I will leave the wheel wells in place- just no retract mechanism for now.

Can I use fore and aft ball vent angling as a balance control strategy?
matched angling reduces the vertical component of thrust.
Angling can be used for yaw control.
I added a nose thruster, as pitch control using the ball thrusters is a bit tricky-you would need big side angles, and full gimbals.
or use a rocking motion of the entire plane.
Last edited by Owen_bern; Sep 22, 2021 at 05:49 PM.
Sep 22, 2021, 08:00 PM
Registered User
Thread OP
Tilting-type tube nozzle.

What are the efficiency figures like for this type of nozzle?
What losses are there from a pressure-type chamber to creating nozzle thrust?
How effective is the bell lead-in ring?
I will do some Google searches on free nozzle efficiency.
What is the effect of chamber swirl and turbulence?
<edit>
A google search convinces me that you need specialist knowhow to even start answering this question.
The google results are almost totally non-relevant, and not at all useful.
Air nozzles are usually jet, rocket engines, or high pressure cleaning jets.
Maybe thruster swivelling nozzles??
Last edited by Owen_bern; Sep 22, 2021 at 08:09 PM.
Sep 23, 2021, 03:37 AM
Registered User
Thread OP
Well, I have enough design preliminaries to put this away now,
and get back to my helicopter.

I get these sudden enthusiasms, and have to come up with a workable design
and small scale drawings, which I put away until later.

This one looks promising enough to upgrade to number two on my list.

I still need to get my more sporty scratch-built learner going, but that is not critical at the moment, as I have a "goer".
The red one just needs a bit of wing maintenance, and a servo set swap, and a receiver swap.

I should post a few pictures of the bits I have.
I will look at that tomorrow.
----------------------------------------------------------------------------------
I am disappointed that no aerodynamic detailing and vent outlet knowhow seems to be available here.


From observation of the "motorcycle with a hole", the helicopter channel floor can be
flattened a bit.
The lead-in should probably be totally flat vertically each side, and no more than 3 degrees convergence each side, and quite a sharp nose radius, similar to a half inch thick wing.
The outside of the pylon profile can be a much bigger radius,
with a nose similar to a 20% section ratio aerofoil.

The rear can converge at about 10 degree included angle or more,
with channel divergence of 5-7 degrees per side.
The diverging side is not as critical as the converging side.

Some electric cars with separate wheel pods use this type of semi-duct design, too,
and I am sure that race-cars use it on their undersides, in the diffuser area.

Please correct me if I seem to be getting this wrong???

For the EDF venting system, I shall develop the swivel tube-butterfly valve
layout a bit more. - a top shroud over the servo, maybe??
30 degrees each side is probably enough.
Sep 23, 2021, 07:12 AM
AndyKunz's Avatar
You really should check out what they've done in videos on YT. I think you'll find a lot of success.

The guy flying the F-35 frame really sounds like it would catch your fancy and answer your questions.

Andy
Sep 23, 2021, 07:19 AM
AndyKunz's Avatar
These two will get you started. Look at their other videos too.

F-35B | AMAZING HOVER in Livingroom with RC F35 MODEL (6 min 27 sec)


F-35B STOVL RC jet 1:9 (6 min 28 sec)
Sep 23, 2021, 07:21 AM
Registered User
Coupez's Avatar
Quote:
Originally Posted by Owen_bern
I am disappointed that no aerodynamic detailing and vent outlet knowhow seems to be available here.
The questions you're asking don't have simple answers.

I'm not an "Internal Aero" guy, but I'm sure the losses in a duct depend on the length and diameter of the duct, the number of bends, and probably the condition of the inside surface. Just like they do for your clothes dryer vent.

The easiest way to get answers might be to get an EDF unit, and enclose it in a section of PVC pipe. Make some adapters and add various lengths and diameters of pipe upstream and downstream. Devise some way of measuring the thrust.

I suspect you'll find EDF units really, really don't like being buried in the middle of a small diameter pipe. They can move air, but they can't generate anything like the pressure differentials that a jet engine can. If you want to get the most out of an EDF, restrict the airflow as little as possible. Plus, ducting adds weight.
Sep 23, 2021, 07:26 AM
AndyKunz's Avatar
Quote:
Originally Posted by Coupez
Plus, ducting adds weight.
And not ducting adds drag (which you alluded to).

Red pill or blue pill? You need to choose one, Owen.

Andy
Sep 23, 2021, 05:18 PM
Registered User
Thread OP
Quote:
Originally Posted by Coupez
The questions you're asking don't have simple answers.

I'm not an "Internal Aero" guy, but I'm sure the losses in a duct depend on the length and diameter of the duct, the number of bends, and probably the condition of the inside surface. Just like they do for your clothes dryer vent.

The easiest way to get answers might be to get an EDF unit, and enclose it in a section of PVC pipe. Make some adapters and add various lengths and diameters of pipe upstream and downstream. Devise some way of measuring the thrust.

I suspect you'll find EDF units really, really don't like being buried in the middle of a small diameter pipe. They can move air, but they can't generate anything like the pressure differentials that a jet engine can. If you want to get the most out of an EDF, restrict the airflow as little as possible. Plus, ducting adds weight.
I have it on authority that installed thrust is practically the same as free air thrust.
The tailpipe is normally about 80% of the fan swept area, which restricts thrust a little, but gives more velocity, and aircraft speed.
running straight into the fan bellmouth at the bellmouth diameter seems to work OK.
for a 90mm fan, this is 105mm diameter.
The fans seem to benefit from being mounted towards the rear of the aircraft.

I am planning on having a permanent vertical split to about the start of the vertical fin, then horizontal after that- for cross-shafts.

This makes joining segments "interesting", but allows easy internal access, and ability to finish the insides of the air passages a bit better.
Maybe very long screws, plug caps over recessed screw heads.

Nuts are not so good- you need tiny long tubular socket drivers. - maybe a hex-socket drive screwdriver-and bigger access holes.

Making the exterior duct panels accessible in the front UC area would be good, too.

The nozzle swivels need covers, and clamps, so they remove to the inside.

Actually using all that fan power really eats the battery, even a 10 Ah one.- predicted flight time is 6 minutes.
This is probably at full speed, though. -it should cruise at 50 mph at half throttle.
at say160A , that is 4 minutes, 16C , so they have factored in part throttle cruise.

I wonder what the large fuselage cross-section does to drag?
it is about 125 x 220mm- more than usual for this wing area.
I could try working frontal area out and plugging it into eCalc.

I would have been better off with a 5000mAh 6S battery, and building to a smaller scale.

You don't gain anything with a bigger battery.- however, I might as well use what I have got.
The battery wasn't too high a price.

All the scale goes into to lifting the battery.

Ducting can be used for torsional and bending strength, too.- though by the time I add about 16 x 1/8 stringers, that is
increasing strength quite well.

Re: trialling:
I am adding another 1000gf 70mm thruster at the front (claimed 1900gf, but likely only 60% of that),
and can rearrange diameters of the nozzles for fore and aft balance.

Very likely internal turbulence and swirl in the main cavity will drain energy. I just have to try it out.

The ideal is to smoothly reduce velocity in favour of pressure, before flow gets to the nozzles.

The fan units likely cannot sustain a lot of pressure, even the 11 blade units.

It is all a bit like a car engine inlet manifold. As big a centre chamber as possible would help.
Maybe extra divergent cones inside, and a reverse flow back to the front nozzles, would help.
The front nozzle set is very close to the fan exits. It will "see" a lower pressure from the high velocity flow.
I can set the fans a little higher in the fuselage for clearance.

I don't have to fully finish it for testing- It can be hover tested without the outer skin- which will be just iron-on
polyester film for this version, anyway.
It probably needs most of the base structure in place.

I can add extra 70mm 6S EDF units if I need to. - they are quite cheap.

The battery has plenty of reserve, and they can be shut off in flight.

This will all take many servo/motor outputs to run, and a big grunty Arduino processor.
-The helicopter will be good practice.

I have been eyeing up an Arduino Pro series unit- maybe $150 or so- it probably needs an extension board for actual IO- it all runs into an 80 pin connector.
There are actual "control" versions, too.
Last edited by Owen_bern; Sep 23, 2021 at 05:43 PM.
Sep 23, 2021, 05:47 PM
Registered User
Thread OP
The f-35 versions use a totally different lift system to what I am planning.
I will have a look, though.
Sep 23, 2021, 09:56 PM
Registered User
Thread OP
The F-35 with the gimballing edf units looks quite ingenious.

It needs a bit of transformer-fu, however, to fold panels apart and put them back together.
If you go the the trouble of full gimbals, movable panels should be simple!

The idea of putting the EDFs back in line is interesting.
would the second one still contribute?
If the centre chamber was enlarged into a "pressure" chamber, it should.
Basically, it supercharges the second stage.
You have to block off all the holes, though.


I could do a similar thing, without gimbals, and seal off the main chamber at the same time.
It requires quite a bit of servo power, as the pivot point is well ahead of the
fan opening in the EDF unit, so it moves downwards as well as rotating.
I would say he is a robotics student!

A good use for 25+ kgf-cm servos, I suppose?


I will do some sketches. and see if I can carry the hole plugs on the EDF unit.
I have more smaller EDF units, so I don't need full gimbals.

It would be nice to nest right into separate air supply tubes, as well,
for the rear diverted flow.
Sep 24, 2021, 12:25 AM
Registered User
Thread OP
Regarding You Tube f35s, there are a lot of different versions.

The one with 2 fully gimballing fans needs a bigger thrust to weight ratio, I think, than I have.

Those EDF units look about 120mm diameter, and the wingspan is so large, the wings need to be removed.
The stated lift capacity is 7000gf,
and the chassis looks around 4kgf.

The ones with one fixed and one rotating fan possibly work like mine , but with less parts.
There are also versions using various propellers.

I am planning on about 1800gf fixed vertical fans which are also gyro controlled (3), and 5500gf
powering the lower tilt nozzles and rear fixed tailpipe.(x2)

The lower tilt nozzles can be manually tilted with one servo output split 4 ways, and transition can be managed by program with reference to a pitot tube module.- 2 servo channels, and one sensor channel.
the two large fan units can run off a single motor signal, plus 3 independent motor signals for the auto EDFs.
I will look tor a suitable one for the Arduino. 3.3v system.

I possibly need an independent hover input pot transmitter channel, for mixed hover flight.

This could be an indirect gyro/ barometric vertical mix- I don't need to consider mixed flight variations with high forward speed and
extra vertical lift.
Last edited by Owen_bern; Sep 24, 2021 at 12:41 AM.


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Video AV8-B Harrier EDF VTOL JET Prototype flies for first time DominicM Electric Plane Talk 4 Oct 24, 2019 02:23 PM
Build Log EDF Harrier VTOL gbolyo19 VTOLs 126 Apr 08, 2018 08:48 AM
Discussion SR-71 BlackBird EDF VTOL HACK VTOLs 18 Oct 30, 2016 05:10 PM
Cool Phan Rulez: High Speed (Transforming) EDF VTOL Application MONOPHAN1 VTOLs 20 Mar 06, 2016 02:40 AM
Cool Simplest VTOL EDF jet shschon VTOLs 3 Dec 11, 2015 01:58 AM