EDF VTOL Project

gregchandler1 · Oct 19, 2009, 07:39 PM

Originally Posted by gregchandler1
Gentlemen, I am new to this forum and ducted fans in general. During the past few days I have read the entire 50 pages of comments, much to my wife's protests. There is alot of good information, and I was hoping to get some more. I am working on a project of using a ducted fan for propulsion of a flying car. During the past month, I have talked to or emailed several different knowledgeable people about various questions and ideas about this project. Fred has suggested the 2215 motor and fan assembly, which I have ordered. WW has helped with outlet diameters and other questions. Several people have come up with different requirements of static thrust to all up weight, with a range from 1.2 all the way to 2.0 to 1. This wide range can have a huge impact on design. Does anyone know what the static thrust to weight ratio is for hovering flight? Thanks in advance.

I originally posted this thread with the maker of the EDF unit that I have decided to use. There were several answers, with the majority saying about 1.5 to 1 should be the best bet. The motor I have chosen generates about 21 lbs static thrust and weighs about 2.6 lbs, the batteries should weigh about 6 lbs, and hopefully all of the ducting, chassis, and electronics will bring the total weight to about 13 lbs. I know that there are many aspects that could greatly increase the weight, but I am trying to keep it as light as possible. Does anyone think that this project will work, or have I started down a dead end road? Any and all suggestions are welcome. Thanks.

Greg

Googleplex · Oct 19, 2009, 11:03 PM

It's been done several times before, so a dead end road it is not

.

One of the biggest hurdles of controlled hovering in keeping the inertia moment of the model as compact as possible and as near as possible to the common centre of lift in all axis. Simply put, an object rotates around its center of mass (which should be under of on the center of lift for a hovering aircraft). The bigger the distribution of mass from the balance point, the harder it is to get the aircraft moving when you want to roll / pitch etc, and harder to stop when you want to stop.

Keeping the center of mass below the center of lift (in the case of what you plan to make, the center of lift will be on the plane defined by the fan rotors) will make it more stable in the air as it will want to hang a bit like a pendulum, but making it too far below the center of lift and the pendulum effect will start to cause issues - it's the moment of intertia thing again.

As for the fans, intake rings will boost efficiency higher than a bare fan in its shroud if you can get some to fit.

As for power to weight, the more the merrier to some degree, but too much will be a waste of efficiency on the powerplants. If you can aim to hover at around 65% - 75% throttle on the fans, you will also be getting best efficieny from the motors, as brushless motors usually run the most efficient in the 65%-90% band. I used this rule on my V-22 (there is a thread around here somewhere on it). Running the nacelles at 60% throttle resulted in 700g thrust each, so I aimed to keep the model under 1400g. Then end thrust level was well over 2:1, but when tooling around in hover, I get awesome endurance.

Brandano · Oct 21, 2009, 10:37 AM

Quote:

Originally Posted by Googleplex

Keeping the center of mass below the center of lift (in the case of what you plan to make, the center of lift will be on the plane defined by the fan rotors) will make it more stable in the air as it will want to hang a bit like a pendulum.

No. Again and again, it will not. The vectors of thrust tilt with the plane

gregchandler1 · Oct 21, 2009, 10:28 PM

My current design will have the rear outlet lower than my fan rotor, and the two front outlets about the same height as the fan rotor. I plan on putting the batteries as low as possible, to keep the CG low. I am sure that there will be numerous design changes, but hopefully I can keep them down to a managable number by asking questions and researching different sites. Thanks for the help, and I am still open to all suggestions as I sure don't know much about the aerodynamics of how all of the components will work together.

Googleplex · Oct 22, 2009, 12:17 AM

Quote:

Originally Posted by Brandano

No. Again and again, it will not. The vectors of thrust tilt with the plane

Ahhh, but the vector of gravity doesn't rotate with the aircraft. It always pulls straight down, and as soon as you have the center of mass being offset in any axis from the pivot of support, gravity will always return it to UNDER the pivot of support. It's a pendulum.

The aircraft can be considered a pendulum. A force that resists gravity is applied at the pivot support (center of lift of the wings or fans or whatever) and gravity is applied to the center of mass. When hovering, the center of mass is below the center of lift (the pivot), but when the aircraft rotates from off the hoizontal, the vertical offset of the center of mass from the center of lift means the center of mass moves out from under the pivot in relation to gravity. Gravity will want to move it back under, and the only way to do that is to try and rotate the aircraft back to the horizontal.

It's why a hang glider pilot hangs on the bottom - he controls the aircraft by shifting the center of mass of the aircraft reletive to the pivot (the center of lift of the aircraft), and gravity ensures he ends up back below the pivot by rolling or pitching the hanglider.

hoverman · Oct 22, 2009, 03:56 AM

Not entirely correct, since an aircraft pivots around its Centre of Gravity, not around it's aerodynamic centre. If the hanglider pilot moves his body left he places the CG to the left which means that the total lift production on the right side of the GC will be larger than on the left side, thus the hanglider rolls. However this roll is pivoted around the centre of Gravity!

I always thought it was around the aerodynamic centre, however several books in my ATPL study state its the CG.

If you look around the internet you will find some explanations of this principle, for example that people tend to believe that a high wing aircraft is more stable dan a low wing aircraft because it's CG is lower, which is not the case at all.

Also a helicopter doesn't seem to be any more unstable when flying inverted compared to when it's flying upright. (sometimes it seems more unstable but this is propably due to the pilot:P)

Marijn

Brandano · Oct 22, 2009, 05:22 AM

Essentially the hangglider pilot shifts the vector of lift relative to the CG. This will work as a control method, but if the pilot grabs the control bar and holds it rigidly it will not have any stabilizing effect other than the dihedral of the wing and the wing sweep, both purely aerodynamic effects that work due to the induced sideslip. With a center of lift rigidly attached to the plane's fuselage there is no stabilizing effect at all, you will always need some sort of gyroscopical stabilization. An helicopter without flybar paddles will tilt over even if the rotor is above the CG. BTW, to test this I did swap and invert the blades on a toy coax helicopter. Seems pretty stable when flying inverted, but needs an adjustment in the angle of precession of the flybar

hoverman · Oct 22, 2009, 06:02 AM

Exactly, this pendulum effect doesn't work, the hangglider will only stay stable when it has some dihedral or it uses a gyro, otherwise the pilot needs to give controll inputs to get back in a horizontal flying situation from a turn.

Somewhere on this forum there is a thread about an EDF VTOl which has its electronics etc mounted in a very long tube which is positioned vertically above the fan. The CG therefore is way above the fan, however keeping the craft stable is easy because the model pivots around its high CG, and therefore the fan mounted all the way at the bottom has a long moment arm and therefore gives good controll. it is..kind of comparable to a broomstick which you balance on your hand. If the broomstick would be half the length it would prove increasingly difficult to balance the stick.

Marijn

gregchandler1 · Oct 22, 2009, 09:55 AM

I have not bought the radio, servos, or possibly gyros yet for this project. Not being familiar with all of the radio mixing or on board mixing options, there were a few things that need clarafying. First, I plan on using three nozzles all coming from 1 EDF. The front nozzles will each have 25% thrust and the rear nozzle will have about 50% thrust. I want to controll all three nozzles at the same time to rotate forward and back to transition from hover to level flight and then back to hover, with the flip of 1 switch taking a total of about 10 seconds to allow for forward momentum to gererate lift on the wings. How can I set them up to do this, and still have some independent movement of the front two nozzles and the rear nozzle, which will give me pitch control? To have nose down pitch, I will rotate the forward nozzles back, keeping the rear nozzle down, and to have nose up pitch, I will rotate the rear nozzle forward, and the front nozzles down. I will have one servo for roll controll and one for yaw. I have already figured out these, but it seems a little challenging figuring out the pitch and transition from hover to forward flight control. Does anyone have any ideas about what servos and or mixing would work?

Googleplex · Oct 22, 2009, 09:25 PM

Duh, of course, I was forgetting the sideways thrust component in my modelling, which would apply another acceleration.

Anyway, after the detour around the garden, back on topic...

Assuming the front nozzles are locked together in a pair always (i.e. they stay parallel at all times and only rotate for transition / pitch control), you could do what you suggest on a transmitter like a DX7 or JR X3810ADT and some V-Tail mixers. The DX7 / 3810 has 6 mixers built in, and they can be set up asymetrically (i.e. the different sides of the stick motion can be mixed in different amounts).

I _THINK_ you can achieve the control system by:

On the model - front nozzle pair is driven a V-Tail mixer connected to gear channel (for transition), and flap channel (flap switch is set to have no effect on output via the flap setting menu on the TX). Only 1 output of the V-Tail is used, connected to the nozzle servo (as to which output to use and what input channel the gear channel goes to etc you'll have to experiment to see). In this config, both the gear and flap channel can control the 1 servo.

Rear nozzle is driven by a V-Tail mixer connected to gear channel and channel 7 (AUX2 - Using the setup menu on the TX this channel can be assigned to one of the rocker switches for manual control (a knob in the case of a 3810)). Again, only 1 servo output is used, and the mixer acts to combine the two channels to control 1 servo.

Now on the TX- Set one mix to mix the forward elevator channel travel with the flap channel, and another with the rear elevator travel with AUX 2.

It might take some messing around to get the connections correct and the mixes on the TX going the right way, but it should work.

For controlling the rate of the transition, a servo slow unit would be ideal. Of if you can spring for a more $ radio like a JR 9X or above, you can control servo speeds on the TX itself on the gear channel (it's usually used to make mechanical retracts retract at a scale speed instead of just snap in).

gregchandler1 · Oct 25, 2009, 11:58 PM

Googleplex,
Thanks for the help. I have a friend at work who has a 10 channel JR that he is going to let me borrow and play with. I will try to work out some of those mixes. We will see what happens.
Greg

gregchandler1 · Nov 15, 2009, 03:14 PM

I am still trying to finalize everything, and have come up with the following idea. Could you use one EDF and motor with a driveshaft from this EDF to power a second EDF, thus only using one motor and ESC, and have a craft similiar to a CH-47 with EDFs instead of rotors? I will use nozzles to vector the thrust, and differential thrust to control roll. The driveline will have a gearless transmission to allow the non-engined EDF to either increase or decrease thrust. This should allow for pitch control without all of the complicated mixing. I have a friend who can machine some custom made parts that would possibly allow everything to come together. I am still waiting on a friend's JR radio to play with mixes. Let me know what you think.

Googleplex · Nov 15, 2009, 09:55 PM

You could drive both off the one motor for sure... BUT! At the speed EDF's spin it would require some pretty precisely balanced engineering - especially with the gearless variable transmission. It'll have to spin a lot faster than the forward / rear coupling shaft on a CH-64

.

For my 2c, 2 motors and 2 escs and a quality V-tail mixer would probably be the better config for simplicities sake (there is a new V-tail mixer out that looks really good for VSTOL - http://www.desertrc.com/sm2.html)

Oct 19, 2009, 07:39 PM	#1
gregchandler1 Registered User Join Date: Oct 2009 Posts: 16	EDF VTOL Project Originally Posted by gregchandler1 Gentlemen, I am new to this forum and ducted fans in general. During the past few days I have read the entire 50 pages of comments, much to my wife's protests. There is alot of good information, and I was hoping to get some more. I am working on a project of using a ducted fan for propulsion of a flying car. During the past month, I have talked to or emailed several different knowledgeable people about various questions and ideas about this project. Fred has suggested the 2215 motor and fan assembly, which I have ordered. WW has helped with outlet diameters and other questions. Several people have come up with different requirements of static thrust to all up weight, with a range from 1.2 all the way to 2.0 to 1. This wide range can have a huge impact on design. Does anyone know what the static thrust to weight ratio is for hovering flight? Thanks in advance. I originally posted this thread with the maker of the EDF unit that I have decided to use. There were several answers, with the majority saying about 1.5 to 1 should be the best bet. The motor I have chosen generates about 21 lbs static thrust and weighs about 2.6 lbs, the batteries should weigh about 6 lbs, and hopefully all of the ducting, chassis, and electronics will bring the total weight to about 13 lbs. I know that there are many aspects that could greatly increase the weight, but I am trying to keep it as light as possible. Does anyone think that this project will work, or have I started down a dead end road? Any and all suggestions are welcome. Thanks. Greg

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Build Log EDF VTOL Project	mnfiero	Electric Ducted Fan Jet Talk	88	Mar 15, 2009 04:35 PM
Build Log "...EDF VTOL PROJECT"...- CONTINUE...	Fon	Electric Ducted Fan Jet Talk	0	Jan 29, 2007 06:31 PM
Discussion EDF VTOL Project	mnfiero	VTOLs	1	Jan 31, 2006 02:08 PM
electric vtol project?	ttm	Electric Plane Talk	2	Jul 17, 2001 01:42 AM
EDF VTOL Project	blswinney	Electric Ducted Fan Jet Talk	14	May 19, 2001 07:36 PM

Oct 19, 2009, 11:03 PM	#2
Googleplex It's a really big number. Join Date: Jan 2001 Location: Brisbane Intl, Australia Posts: 1,161	It's been done several times before, so a dead end road it is not . One of the biggest hurdles of controlled hovering in keeping the inertia moment of the model as compact as possible and as near as possible to the common centre of lift in all axis. Simply put, an object rotates around its center of mass (which should be under of on the center of lift for a hovering aircraft). The bigger the distribution of mass from the balance point, the harder it is to get the aircraft moving when you want to roll / pitch etc, and harder to stop when you want to stop. Keeping the center of mass below the center of lift (in the case of what you plan to make, the center of lift will be on the plane defined by the fan rotors) will make it more stable in the air as it will want to hang a bit like a pendulum, but making it too far below the center of lift and the pendulum effect will start to cause issues - it's the moment of intertia thing again. As for the fans, intake rings will boost efficiency higher than a bare fan in its shroud if you can get some to fit. As for power to weight, the more the merrier to some degree, but too much will be a waste of efficiency on the powerplants. If you can aim to hover at around 65% - 75% throttle on the fans, you will also be getting best efficieny from the motors, as brushless motors usually run the most efficient in the 65%-90% band. I used this rule on my V-22 (there is a thread around here somewhere on it). Running the nacelles at 60% throttle resulted in 700g thrust each, so I aimed to keep the model under 1400g. Then end thrust level was well over 2:1, but when tooling around in hover, I get awesome endurance.

Oct 21, 2009, 10:28 PM	#4
gregchandler1 Registered User Join Date: Oct 2009 Posts: 16	My current design will have the rear outlet lower than my fan rotor, and the two front outlets about the same height as the fan rotor. I plan on putting the batteries as low as possible, to keep the CG low. I am sure that there will be numerous design changes, but hopefully I can keep them down to a managable number by asking questions and researching different sites. Thanks for the help, and I am still open to all suggestions as I sure don't know much about the aerodynamics of how all of the components will work together.

Oct 22, 2009, 03:56 AM	#6
hoverman Registered User Join Date: Sep 2006 Location: Zwijndrecht Posts: 17	Not entirely correct, since an aircraft pivots around its Centre of Gravity, not around it's aerodynamic centre. If the hanglider pilot moves his body left he places the CG to the left which means that the total lift production on the right side of the GC will be larger than on the left side, thus the hanglider rolls. However this roll is pivoted around the centre of Gravity! I always thought it was around the aerodynamic centre, however several books in my ATPL study state its the CG. If you look around the internet you will find some explanations of this principle, for example that people tend to believe that a high wing aircraft is more stable dan a low wing aircraft because it's CG is lower, which is not the case at all. Also a helicopter doesn't seem to be any more unstable when flying inverted compared to when it's flying upright. (sometimes it seems more unstable but this is propably due to the pilot:P) Marijn

Oct 22, 2009, 05:22 AM	#7
Brandano Registered User Join Date: Oct 2004 Posts: 1,385	Essentially the hangglider pilot shifts the vector of lift relative to the CG. This will work as a control method, but if the pilot grabs the control bar and holds it rigidly it will not have any stabilizing effect other than the dihedral of the wing and the wing sweep, both purely aerodynamic effects that work due to the induced sideslip. With a center of lift rigidly attached to the plane's fuselage there is no stabilizing effect at all, you will always need some sort of gyroscopical stabilization. An helicopter without flybar paddles will tilt over even if the rotor is above the CG. BTW, to test this I did swap and invert the blades on a toy coax helicopter. Seems pretty stable when flying inverted, but needs an adjustment in the angle of precession of the flybar

Oct 22, 2009, 06:02 AM	#8
hoverman Registered User Join Date: Sep 2006 Location: Zwijndrecht Posts: 17	Exactly, this pendulum effect doesn't work, the hangglider will only stay stable when it has some dihedral or it uses a gyro, otherwise the pilot needs to give controll inputs to get back in a horizontal flying situation from a turn. Somewhere on this forum there is a thread about an EDF VTOl which has its electronics etc mounted in a very long tube which is positioned vertically above the fan. The CG therefore is way above the fan, however keeping the craft stable is easy because the model pivots around its high CG, and therefore the fan mounted all the way at the bottom has a long moment arm and therefore gives good controll. it is..kind of comparable to a broomstick which you balance on your hand. If the broomstick would be half the length it would prove increasingly difficult to balance the stick. Marijn

Oct 22, 2009, 09:55 AM	#9
gregchandler1 Registered User Join Date: Oct 2009 Posts: 16	I have not bought the radio, servos, or possibly gyros yet for this project. Not being familiar with all of the radio mixing or on board mixing options, there were a few things that need clarafying. First, I plan on using three nozzles all coming from 1 EDF. The front nozzles will each have 25% thrust and the rear nozzle will have about 50% thrust. I want to controll all three nozzles at the same time to rotate forward and back to transition from hover to level flight and then back to hover, with the flip of 1 switch taking a total of about 10 seconds to allow for forward momentum to gererate lift on the wings. How can I set them up to do this, and still have some independent movement of the front two nozzles and the rear nozzle, which will give me pitch control? To have nose down pitch, I will rotate the forward nozzles back, keeping the rear nozzle down, and to have nose up pitch, I will rotate the rear nozzle forward, and the front nozzles down. I will have one servo for roll controll and one for yaw. I have already figured out these, but it seems a little challenging figuring out the pitch and transition from hover to forward flight control. Does anyone have any ideas about what servos and or mixing would work?

Oct 22, 2009, 09:25 PM	#10
Googleplex It's a really big number. Join Date: Jan 2001 Location: Brisbane Intl, Australia Posts: 1,161	Duh, of course, I was forgetting the sideways thrust component in my modelling, which would apply another acceleration. Anyway, after the detour around the garden, back on topic... Assuming the front nozzles are locked together in a pair always (i.e. they stay parallel at all times and only rotate for transition / pitch control), you could do what you suggest on a transmitter like a DX7 or JR X3810ADT and some V-Tail mixers. The DX7 / 3810 has 6 mixers built in, and they can be set up asymetrically (i.e. the different sides of the stick motion can be mixed in different amounts). I _THINK_ you can achieve the control system by: On the model - front nozzle pair is driven a V-Tail mixer connected to gear channel (for transition), and flap channel (flap switch is set to have no effect on output via the flap setting menu on the TX). Only 1 output of the V-Tail is used, connected to the nozzle servo (as to which output to use and what input channel the gear channel goes to etc you'll have to experiment to see). In this config, both the gear and flap channel can control the 1 servo. Rear nozzle is driven by a V-Tail mixer connected to gear channel and channel 7 (AUX2 - Using the setup menu on the TX this channel can be assigned to one of the rocker switches for manual control (a knob in the case of a 3810)). Again, only 1 servo output is used, and the mixer acts to combine the two channels to control 1 servo. Now on the TX- Set one mix to mix the forward elevator channel travel with the flap channel, and another with the rear elevator travel with AUX 2. It might take some messing around to get the connections correct and the mixes on the TX going the right way, but it should work. For controlling the rate of the transition, a servo slow unit would be ideal. Of if you can spring for a more $ radio like a JR 9X or above, you can control servo speeds on the TX itself on the gear channel (it's usually used to make mechanical retracts retract at a scale speed instead of just snap in).

Oct 25, 2009, 11:58 PM	#11
gregchandler1 Registered User Join Date: Oct 2009 Posts: 16	Googleplex, Thanks for the help. I have a friend at work who has a 10 channel JR that he is going to let me borrow and play with. I will try to work out some of those mixes. We will see what happens. Greg

Nov 15, 2009, 03:14 PM	#12
gregchandler1 Registered User Join Date: Oct 2009 Posts: 16	I am still trying to finalize everything, and have come up with the following idea. Could you use one EDF and motor with a driveshaft from this EDF to power a second EDF, thus only using one motor and ESC, and have a craft similiar to a CH-47 with EDFs instead of rotors? I will use nozzles to vector the thrust, and differential thrust to control roll. The driveline will have a gearless transmission to allow the non-engined EDF to either increase or decrease thrust. This should allow for pitch control without all of the complicated mixing. I have a friend who can machine some custom made parts that would possibly allow everything to come together. I am still waiting on a friend's JR radio to play with mixes. Let me know what you think.

Nov 15, 2009, 09:55 PM	#13
Googleplex It's a really big number. Join Date: Jan 2001 Location: Brisbane Intl, Australia Posts: 1,161	You could drive both off the one motor for sure... BUT! At the speed EDF's spin it would require some pretty precisely balanced engineering - especially with the gearless variable transmission. It'll have to spin a lot faster than the forward / rear coupling shaft on a CH-64 . For my 2c, 2 motors and 2 escs and a quality V-tail mixer would probably be the better config for simplicities sake (there is a new V-tail mixer out that looks really good for VSTOL - http://www.desertrc.com/sm2.html)