Angled Ducted Fans: A New Way to Improve Stability?

[Gigabyte111]

Ok, here is an idea I have been playing around with for about 2 months now:
i am design a RC aircraft that is lifted verticaly into the air by 4 ducted fans. however an issue with stability came up, so instead of trying to rely on a low center of gravity alone for the stability, I came up with the idea to mount the fans on an angle so that the thrust would be angled away from the plane, which would work much the same as a dihedral in a wing.
this is a diagram to help clarify what i am trying to get accoplish
http://img382.imageshack.us/my.php?image=fanmovement0cu.jpg"

that is my theory,

if anyone could give me any insight as to whether of not this will work, that would be very helpfull.
thanks, zac

 

[FredGarvin]

Yes that will work in theory. It's obviously not the most efficient use of the thrust created, but you are trading efficiency for some stability.

The only thing I would mention is that no matter what you do, it is going to be unstable. Even the big aircraft have stability issues that are dealt with in their control systems via computer or mixing. I would suggest that you look into the small gyroscopes that are used in RC helicopters to help you with your problems.

 

[enigma]

It possible that it won't tip over, but it will wobble horrendously. Have you taken any classes in controls?

 

[Gigabyte111]

It possible that it won't tip over, but it will wobble horrendously. Have you taken any classes in controls?

what do you mean by classes in controls?

do you mean classes in how to fly an rc plane, because if you meant that then here's your answer, yes i do know how to fly a model airplane, actualy i hacve about 6 of them up above my garage, all of which i can fly with well engout to not crash them, but not well enough to do dificuilt stunts,

or did you mean something else that i am not thinkinging of?

zac

 

[enigma]

No, I'm talking about https://www.amazon.com/gp/product/0471445770/?tag=pfamazon01-20

It would teach you how to fix the stability issues. Basically, the problem you're going to have is this:

When the craft pitches to one side (from your diagram, let's call it counterclockwise) the thrusters will impart a clockwise torque. That torque will stop the counterclockwise motion and start the thing turning clockwise. The clockwise torque will continue until it is righted, but at that point, it will still have a clockwise motion, and the same situation will occur on the other side.

For stability, you want the clockwise torque and the clockwise motion to approach zero at the same time.

Depending on your gain, it will either continuously wobble, or it will eventually increase the amplitude until it flips over. You may be able to manually control it, but it's not going to be an easy bird to fly.

 

[Gigabyte111]

i see what your saying and i have thought about that, and i have drawn a couple of plans to address that issue but none of them are very plausable. one calls for a servo to pull a weight along a track similar to the ones used in those game things where you try to grab the stuffed animal with a claw, but pull it in the opposit direction of the tilt so as to off set the C of G.

zac

 

[Gigabyte111]

keep in mind that i am on a very tight budget, and i am still in high school, so i don't have the time to build proof of concepts, so i never know it they will work or not unless somone has already built one

 

[Danger]

I wasn't aware that model choppers have gyros, but Fred has a good idea there. Maybe you can mount all 4 fans on gymbals with two-axis servo controls operated by the gyros. That would give you pitch as well as roll stability. If you plan to use the same fans for forward (and lateral?) thrust, you'd need to set your R/C receiver up to provide a manual override without actually disabling the gyro input.
Have you considered how this thing will handle when it comes to banking into turns or pitching up/down?

 

[Gigabyte111]

i was thinking about for a propulsion method, i would simply adjust the C of G forward or aft, left or right. this would create an imbalance causing the leading edge to dip down and the fans on the leading edge to take up all of the vertical lift. then the trailing edge fans will provide the forwad momentum. (see diagram posted previousely). my calculations predicted a very fast and maneuvarable craft. i also think that this idea is patentable, so i am claiming all design rights to this at 3:13 wensday january 18, 2006

zac

 

[Danger]

But it's only 1:20.
You should do a very thorough patent search before going to any great expense. A lot of it is probably covered already by things like the Moller M400, the Harrier, various helicopters, etc.. I don't mean to discourage you, but be forewarned that you're not exactly in virgin territory.

 

[Gigabyte111]

its one twenty where you are but it is 3:37 where i am (EMT)

 

[FredGarvin]

I wasn't aware that model choppers have gyros, but Fred has a good idea there. Maybe you can mount all 4 fans on gymbals with two-axis servo controls operated by the gyros. That would give you pitch as well as roll stability. If you plan to use the same fans for forward (and lateral?) thrust, you'd need to set your R/C receiver up to provide a manual override without actually disabling the gyro input.
Have you considered how this thing will handle when it comes to banking into turns or pitching up/down?

The gyros used are quite nice and really allow the R/C rotary wing guys to do the really weird stuff real helicopters can't do. They really do add an incredible amount of stability and they are in a nice, neat package that is designed to be integrated into the control system right away.

do you mean classes in how to fly an rc plane, because if you meant that then here's your answer, yes i do know how to fly a model airplane, actualy i hacve about 6 of them up above my garage, all of which i can fly with well engout to not crash them, but not well enough to do dificuilt stunts,

I hate to break it to you, but that will only help a little bit. If you can, try to talk to someone local who flys R/C helicopters. It's an entirely different skill set and talent. I gaurantee you that you would be nothing but frustrated without aa lot of experienced help.

 

[Gigabyte111]

just for informational purpuses, how much would the gyros cost me, can i hook up 8 electric engins to them, how much do they weigh, and how big are they.

zac

 

[enigma]

http://www.spp.co.jp/sssj/crs03cataloge.pdf

That's just one of the first that I found online. With a 5V power source outputs + or - 2.5V to measure angular velocity. They didn't list the cost. You'd have to contact the manufacturer, but they're pretty about 1" square by 2/3"

http://www.futaba-rc.com/radioaccys/futm0807.html that has some that are about the same size. Again, no price listed that I could find, but they do have some FAQs. I didn't dig too far into them.

Both of those are designed for use in model aircraft, so they aren't going to cost exorbitant amounts of money.

As for hooking them up, the worst case scenario will involve getting a microcontroller and learning to program it.

Give one of the companies a call. I'm sure someone would be glad to answer any questions.

 

[FredGarvin]

The standard is like this:
http://www2.towerhobbies.com/cgi-bin/wti0001p?&I=LXVA42&P=ML

The pricing here isn't too terribly bad. IIRC, this one hooks up in line with the servo and the receiver.

Here is an article. It's a bit dated, but it would be a good first read on gyros for RC applications.

http://www.heliproz.com/jwgyros.html

 

[vampire67]

http://img382.imageshack.us/my.php?image=fanmovement0cu.jpg"
that is my theory,
if anyone could give me any insight as to whether of not this will work, that would be very helpfull.
thanks, zac

Unfortunely Gigabyte111 were not referring the pending thread:
http://www.rcgroups.com/forums/showthread.php?t=460061"
In short, the discussion is wether the thrust vector dihedral/angle does stabilise or the offset of the center of gravity (CoG) vs. the center of thrust/lift (CoL) is responsible. Or if there is no stabilising effect at all.
I already did a lenghy calculation in RC Groups showing the supporters of thust vector dihedral, that when CoG and CoL have no offset the thrust vector dihedral has zero stabilising effect. So all you do with a thust dihedral is lowering the lift effectivity.
So can support somebody her one or the other point of view with some facts ?
regards Peter

P.S. here some picture to clarify the differences:
Stabilising by CoG-CoL offset:
http://people.freenet.de/peter.dittmann/pic/CoG-CoL_offset.png"
http://people.freenet.de/peter.dittmann/pic/Offset_stabilising.png"

Thrust dihedral (I assume no stabilising effect here based on calculation)
http://people.freenet.de/peter.dittmann/pic/Thrust_dihedral.png"

 

[FredGarvin]

That is all well and good for static applications, but it is not taking into account any dynamic inputs. You will most certainly have to have some kind of constant feedback loop to continually make corrections.

It appears, on the surface, that an adjustable thrust dihedral would be almost equivilent to the compensation of CL vs. CG. I don't think that the adjustable part has been mentioned as of yet.

 

[vampire67]

That is all well and good for static applications, but it is not taking into account any dynamic inputs. You will most certainly have to have some kind of constant feedback loop to continually make corrections.

Well, Nobody believed a dynamicall feedback could be obmitted. The idea was (if I remember it right) that a added statical stability would make life easier for any dynamical feedbach (aka. gyro stabilising).

It appears, on the surface, that an adjustable thrust dihedral would be almost equivilent to the compensation of CL vs. CG. I don't think that the adjustable part has been mentioned as of yet.

You mean instead of adjusting the thrust level to use a servo to control the trust direction ?
That's sounds indeed a new idea.

The problem originally was the lateral stabilising of the VTOL plane/copter.
Experiments already showed a poor control stability as the ducted fans (EDFs) generating the thrust are rather slow responding. This make life of any gyro-based dynamical stabilising difficult.
It was also reported that the control throw of the EDF must be rather high to get a real stabilising effect. So any disturbance (wind) would almost shut down on thruster while the other went to max (a little exaggerating ).

But I have no idea how you think controlling thrust direction will help here.
The main thrust must stay vertically (its hovering). So with a thuruster left and right of CoG I get almost no resulting momentum if I change the thrust direction. Only the thrust differential make a big impact.
The only sensible quick response I could imagine would a lateral movement of the thruster in relation to CoG (like an artist on a rope moving a long pole left and right to compensate the momentum).

Peter

 

[Gigabyte111]

Unfortunely Gigabyte111 were not referring the pending thread:
http://www.rcgroups.com/forums/showthread.php?t=460061"
In short, the discussion is wether the thrust vector dihedral/angle does stabilise or the offset of the center of gravity (CoG) vs. the center of thrust/lift (CoL) is responsible. Or if there is no stabilising effect at all.
I already did a lenghy calculation in RC Groups showing the supporters of thust vector dihedral, that when CoG and CoL have no offset the thrust vector dihedral has zero stabilising effect. So all you do with a thust dihedral is lowering the lift effectivity.

i was aware of your imput in the rc forums website, as it was i who posted the origional idea of the vectored thrust on that website. i was hoping to get a couple of people with a high level degree in this stuff to comment on my idea (not to say that you don't have degree, I am sure you do), but this forum is more specific to why it would or wouldent work. (some how, with the mixed imput i am getting, i think the only way to know is to try it out, so i am coming up with a chart that will tell me what angle is most effective to set the fans at.

 

[Gigabyte111]

Formula

can anyone tell me what all of this means.

the formula that i used to find this was

A=the angle the fan is set at
E=the number of ducted fans
R=the percent of throttle remainin to accend
W=the weight of model, excluding the fansA/90 ->set answer to equal->B
B*100 ->set answer to equal->C
C/2 ->set answer to equal->D
100/D ->set answer to equal->F

WR ->set answer to equal->V

F(W+V) ->set answer to equal->T

T/E ->set answer to equal->H

T= the total thrust needed to lift the model
H= the lift each fan must produce to lift the model. the chart that it produced is attached.

keep in mind that the amount of thrust needed to hover will be 15% less than what is listed. this is to ensure that i have ample thrust to make the craft accend. the cells that have a color background are matches.

based on my stastics and a litle reasearch, i would want to use a http://www2.towerhobbies.com/cgi-bin/wti0001p?&I=LXHHR0&P=7" at 9.6V in a 130 gram model, at a 60* angle. any objections?also, if anyone comes across a ducted fan that produces more than 117 Grams of thrust after the weight of the motor has been subtracted. please let me know, because that would be very cool.

plus if anyone finds a mathmatical error in either my formula or my chart, let me know. thanks

Zac

 

[vampire67]

can anyone tell me what all of this means.
the chart that it produced is attached.

Can't open the word attachment. Claims it's broken or parts missing.
Rather use pdf or other less DANGEROUS formats

Peter

P.S. Do you had external links or specific libraries in the doc ? It almost brings my system to a halt for minutes

 

[Gigabyte111]

Can't open the word attachment. Claims it's broken or parts missing.
Rather use pdf or other less DANGEROUS formats
Peter
P.S. Do you had external links or specific libraries in the doc ? It almost brings my system to a halt for minutes

all it contains is a chart that i created in excel, i will just create a pdf file like you suggested.

zac

ps on an unrelated idea, is there anyway to use a gyro to hold altitude pitch, yaw and roll, without any lifting surfaces. eg. wings

 

[Gigabyte111]

here is the chart, for real this time