Originally Posted by samwei1950

Kelly
many thanks for the great comments
let me catch your paces again...
1.move elevator/rudder to upper control surfaces with 2 servos
2.move YAW to lower control surfaces but only with 2 servos & 2 vanes
3.no skirt but spoilers may help
4.reduce some rate on gyros.especially upper control surfaces gyros.
I'm willing to do the tests to see what benifits the ball and come back to you again.
one question: do i needed to mix upper & lower control surfaers?
BTW:could not speak well but i can read Japaness,any secific topic you need my hands?
best regards
Sam

Hi Sam,
Everything above is correct except part of #3, You may want to leave the skirt on, but since you've all ready removed it that's okay too.

Mixing is not required.

You could have all four lower vanes used for yaw axis control too if you don't want to remove the servos. That maybe easier and quicker.

The idea is to reverse the controls to move the pitch and roll to the upper vanes, and the yaw axis to the lower. The CG may need to be adjusted too.

Before your progress to the test above please try this simple suggestion.

Since you've already removed the skirt, I'd suggest that you move the battery downward or add small amounts of weights to the very bottom of the sphere and test fly, add a bit more weight and test fly again etc, then make a new video for us to all enjoy! Let us see what the affect of moving the CG downward has on the craft.

Removing the skirt affected the Center of Pressure and the CG. So if you drop the CG back down with some small weights or battery placement, we then can tell if the removal of the skirt affect the stability (CP change) and or it was a combination of the removal of the weight of the skirt and the re-positioning of the CP.

Center of Gravity is easy to measure, but CP is a dynamic function that represent a greater challenge. It can be simply approximated by an old method used in model rocketry. If you're interest I can elaborate on that later.

Brandano has made many good comments on the CP/CG subject on several threads, and it maybe helpful to look over some of his postings. We had an interesting discussion on this subject last years, just click on his blue name badge on of his postings and read up on his suggestions. I'll try to recap his main theme, and perhaps he can add more to this thread?

For Hover: you need the CG and CP to be at the same location, or perhaps have the CG below the CP.

For Horizontal (high speed) flight The CG needs to be forward the CP.

So it depend on what you expect from your sphere. If you only want to hover and have it very stable in hover then drop the CG downward, but don't expect it to be able to fly in sprint/horizontal mode very well.

For horizontal high speed flight the reverse is true, CG moved forward, it will be very stable but will not hover well, thus we read in the Japanese Sphere patent the need for a mechanical movement of the CG.

Now I think they figured out that they could leave the CG forward for high speed flight and use the spoilers to remove or reduce the oscillations in hover. The spoiler would damping the pitch/roll motions, eliminate the complex mechanical CG moving device, make it lighter and easy to manufacture.

So that leaves you with a few choices. Just hover? or both high speed flight and hover, or like in the case of the Crazy Ball that NewGuy found, mostly high speed flight, but I have to say the Crazy Ball may hover very well with the addition of spoilers and two yaw axis control vanes just below the propeller.

For just hover; maybe just lowering the CG location, or maybe the re-installing the skirts and a lower CG, or maybe the addition of spoilers, or maybe switching around the controls to be a canards. It's all very hard to guess, but the correct testing procedures will clearly indicate what will work. Just make one change at a time. I'd try lowering the CG, then try adding back the skirts, then switch the controls around to a canard configuration then add spoilers.

I think the skirts are for the purpose in increasing the wing area. This will lower the wing loading allowing for a lower speeds in horizontal flight and less power required for horizontal fight. The skirt would have an affect on the CP, just remember that the Japanese sphere is configured in a canard configuration and the wing is down low, thus the skirt functions as more wing area. Right now your craft is configured backwards so the skirt represent a slight different function more along the lines of increased tail surface area as represented in normal aircraft.

I was thinking that a search for a Japanese patent on the spoilers maybe helpful, or perhaps if you could find out the inventors name from one of those news article. It maybe possible to contact him and ask him to look at this thread and add some details, or maybe he could point us to the patent(s). One would hope that this would not represent a patent infringement just experimenting with these ideas, but it's always good to find out if this spoiler idea is covered and seek permission to experiment with their ideas for a noncommercial fun use. Just a few thoughts on ways to further our knowledge about this exceptional craft.

If you need any of this clarified please let me know.

I'm looking forward to your next update and hopefully new video,

Regards,

Kelly

The spoiler system is genius in it's simplicity. It is amazing that someone thought of it. I couldn't even figure it out watching the videos.

Azarr

Originally Posted by corocopter

Hi Sam,
Everything above is correct except part of #3, You may want to leave the skirt on, but since you've all ready removed it that's okay too.

Mixing is not required.

You could have all four lower vanes used for yaw axis control too if you don't want to remove the servos. That maybe easier and quicker.

The idea is to reverse the controls to move the pitch and roll to the upper vanes, and the yaw axis to the lower. The CG may need to be adjusted too.

Before your progress to the test above please try this simple suggestion.

Since you've already removed the skirt, I'd suggest that you move the battery downward or add small amounts of weights to the very bottom of the sphere and test fly, add a bit more weight and test fly again etc, then make a new video for us to all enjoy! Let us see what the affect of moving the CG downward has on the craft.

Removing the skirt affected the Center of Pressure and the CG. So if you drop the CG back down with some small weights or battery placement, we then can tell if the removal of the skirt affect the stability (CP change) and or it was a combination of the removal of the weight of the skirt and the re-positioning of the CP.

Center of Gravity is easy to measure, but CP is a dynamic function that represent a greater challenge. It can be simply approximated by an old method used in model rocketry. If you're interest I can elaborate on that later.

Brandano has made many good comments on the CP/CG subject on several threads, and it maybe helpful to look over some of his postings. We had an interesting discussion on this subject last years, just click on his blue name badge on of his postings and read up on his suggestions. I'll try to recap his main theme, and perhaps he can add more to this thread?

For Hover: you need the CG and CP to be at the same location, or perhaps have the CG below the CP.

For Horizontal (high speed) flight The CG needs to be forward the CP.

So it depend on what you expect from your sphere. If you only want to hover and have it very stable in hover then drop the CG downward, but don't expect it to be able to fly in sprint/horizontal mode very well.

For horizontal high speed flight the reverse is true, CG moved forward, it will be very stable but will not hover well, thus we read in the Japanese Sphere patent the need for a mechanical movement of the CG.

Now I think they figured out that they could leave the CG forward for high speed flight and use the spoilers to remove or reduce the oscillations in hover. The spoiler would damping the pitch/roll motions, eliminate the complex mechanical CG moving device, make it lighter and easy to manufacture.

So that leaves you with a few choices. Just hover? or both high speed flight and hover, or like in the case of the Crazy Ball that NewGuy found, mostly high speed flight, but I have to say the Crazy Ball may hover very well with the addition of spoilers and two yaw axis control vanes just below the propeller.

For just hover; maybe just lowering the CG location, or maybe the re-installing the skirts and a lower CG, or maybe the addition of spoilers, or maybe switching around the controls to be a canards. It's all very hard to guess, but the correct testing procedures will clearly indicate what will work. Just make one change at a time. I'd try lowering the CG, then try adding back the skirts, then switch the controls around to a canard configuration then add spoilers.

I think the skirts are for the purpose in increasing the wing area. This will lower the wing loading allowing for a lower speeds in horizontal flight and less power required for horizontal fight. The skirt would have an affect on the CP, just remember that the Japanese sphere is configured in a canard configuration and the wing is down low, thus the skirt functions as more wing area. Right now your craft is configured backwards so the skirt represent a slight different function more along the lines of increased tail surface area as represented in normal aircraft.

I was thinking that a search for a Japanese patent on the spoilers maybe helpful, or perhaps if you could find out the inventors name from one of those news article. It maybe possible to contact him and ask him to look at this thread and add some details, or maybe he could point us to the patent(s). One would hope that this would not represent a patent infringement just experimenting with these ideas, but it's always good to find out if this spoiler idea is covered and seek permission to experiment with their ideas for a noncommercial fun use. Just a few thoughts on ways to further our knowledge about this exceptional craft.

If you need any of this clarified please let me know.

I'm looking forward to your next update and hopefully new video,

Regards,

Kelly

many thanks for the good info & advices,and very much appreciated for telling me why to do it in detail,I'm curious about the upcoming testing results and will come back with video soonest.
the designer is a Japaness military officer,and they are kind of
"conservative" I must confessed that the task is considered
"mission impossible" in the real world......should I give a try?
have a nice day
regards
Sam

the designer is a Japaness military officer,and they are kind of
"conservative" I must confessed that the task is considered
"mission impossible" in the real world......should I give a try?
have a nice day
regards
Sam[/QUOTE]

Sam, I agree that is most likely "Mission Impossible". You'd probably find a better use of you time than that adventure.

Maybe the inventor is already watching this thread with great humor, as we try to unravel this marvelous device. He certainly has earned my respect!

It would be interesting to know if the spoilers are covered by a patent. I would guess that they applied for a separate patent for these devices, since they are employed in such a unique manner.

I'm very impressed that NewGuy found the original patent for this sphere. That was a huge help. Gary also contributed a great deal to this cause. Can you think of anyway that you could find out if there is a spoiler patent, perhaps NewGuy maybe able to find that one as well?

Gary, have you had any further luck with your sphere? I was impressed with how stable your craft was with the Open Pilot CopterControl.

I'm looking forward to your post flight reports.

Kelly

Originally Posted by corocopter

the designer is a Japaness military officer,and they are kind of
"conservative" I must confessed that the task is considered
"mission impossible" in the real world......should I give a try?
have a nice day
regards
Sam

Sam, I agree that is most likely "Mission Impossible". You'd probably find a better use of you time than that adventure.

Maybe the inventor is already watching this thread with great humor, as we try to unravel this marvelous device. He certainly has earned my respect!

It would be interesting to know if the spoilers are covered by a patent. I would guess that they applied for a separate patent for these devices, since they are employed in such a unique manner.

I'm very impressed that NewGuy found the original patent for this sphere. That was a huge help. Gary also contributed a great deal to this cause. Can you think of anyway that you could find out if there is a spoiler patent, perhaps NewGuy maybe able to find that one as well?

Gary, have you had any further luck with your sphere? I was impressed with how stable your craft was with the Open Pilot CopterControl.

I'm looking forward to your post flight reports.

Kelly[/QUOTE]
I read the Japaness paten applicational document in the thread twice,
unfortunately...
1.the patent did not mentioned anything about the spoilers.
2.neither how many servos engaged nor how they functioned
maybe they are in the other edition as you mentioned earlier.
here's an Japaness ball drone info I just got from local website
S2-1
brief translation.....
1.the ball idea came from VTOL but which have problem when landing
2.reason to use spherical shape and what benefits.
3.weakness of sphereical shape
4.developing stages
5.relationship between CG & CP
6.the Japaness ball drone spec
7.practicability of the drone
8.ideal future usages
9.challenges ahead
please not hestitated to tell me which charts need further detailed translation.
best regards.
Sam

Sam,

What a great find! Now that is helpful and it will take me some time to digest these photos and I'll get back with you later today or tomorrow with requests for translations.

They gave use some nice historical photos to learn from. Great work!

Tracking down a patent for just the spoilers maybe very difficult without help from within the defense department. I'm still convinced that the spoiler are covered in some manner. I have a hard time believing that they would showcase such a clever concept as the spoiler to the public with out a patent pending. Note: Version #7 shows the spoiler, yet no mentions of these devices. That is very odd. You'd think they would try to cover that bit of art and science.

The problem I've found over the years is trying to figure out what sort of odd names used in the process of describing the patent. They could have used just about any sort description.

I think we have a fairly clear understanding of the spoilers, it's just a matter of getting the size and distance setup correctly and test flying.

Nice work Sam,

Thank you,

Kelly

No to be honest I have been flying other platforms with my CC but now I see this moving forward again and lots more for me to think about but I might have another crack!

So now I have read the previous posts and come upto speed. That PDF is fantastic.

I feel foam cutting fun coming on again!

Great News Gary! I'm excited to see your next creation.

I injured my right hand yesterday, so typing is just one handed for some time. I will take me a few days to get some voice recognition software up and running. In the mean time I'll be watching.

Kelly

Originally Posted by corocopter

Great News Gary! I'm excited to see your next creation.

I injured my right hand yesterday, so typing is just one handed for some time. I will take me a few days to get some voice recognition software up and running. In the mean time I'll be watching.

Kelly

take good care of your right hand, we needed you.
best regards
Sam

Hello gentlemen,

I now have the voice recognition software up and running, it may take a few days for its accuracy to improve, so please excuse my occasional errors.

I have a couple of new ideas to express about the control systems on the sphere. After looking at the video of the control surfaces moving, it appears to me that the upper vanes and the lower vanes work together for pitch and roll control in this manner.

If you consider that the craft has about equal amounts of control surfaces ahead of the center of gravity and aft of the center of gravity, it is possible to make that assumption that the upper the control surfaces would move in the opposite direction of the lower control surfaces. For instance if you wanted the craft to roll to the left, the upper control surfaces on the canard would move to the right pushing the top of the craft to the left, while the lower control surfaces would move to the right pushing the lower section of the craft to the right, thus initiating a pitching motion of the entire sphere to the left. I guess you could think of this as a form of as power steering.

It also appears that the upper and lower control surfaces work together for Yaw control. And input from the pilot for a yaw motion to the left, will make both the upper and lower control surfaces turn an equal amount in the same direction and with an equal amount of deflection.

For simplicity of design and testing this would not be required during testing and learning how to copy the craft, it's too bad we don't have information from the designers on how they learned to optimize their control system.

Kelly

here are flight testings conducted this AM
phase 1
1 no skirt
2.reverse the control surfaces,thus,upper part for ele/rudder,lower part for
YAW,it's a Canard configuration now.
3.moved CG downward by adding extra weight,it's about 1/3 on lower part of
the craft.
4.lower YAW rate but remained same rate for upper surfaces
the testing results of phase 1
-the craft became uncontrolable,it swung & bounced back from wall,
crashed and damaged.
-fixed the craft then moved the CG back to it's original place & tested again.
-same results,so I stopped the testing
phase 2
I put back the skirt and gained a bit of stability,but still I have to called it
uncontrolable,as such,no vid this time.
the findings:
YAW is not an issue even with 2 vanes in action.
the upper control surfaces were calling for help, I'm pretty sure we need to mix the lower control surfaces for "stability compensations?" on the canard
configuration.
so I needed to mix the upper/lower control surfaces,they work together as a team,but how........
nevertheless I'll remove the skirt and put spoilers on the lower part of the craft and test again to see what will happens.
have a nice day
Sam

Good morning Sam,

Thanks for the post flight review.

If you want to setup spoilers on the sphere they need to be set either level or slightly below the level of the propeller, or in your case I would imagine approximately set between the two propellers.

I would imagine that you need to set the control sensitivity of the upper vanes to a lower rate. The reason for this is the velocity of the air exiting the propeller is greater and is acting on the upper vanes with more force than the lower vanes, perhaps reducing the control deflection on the upper vanes may help make the craft more stable. You may be over controlling it in the current configuration.

For clarification, there's two things to consider with the control vanes very close to the propeller, there's the rate of the Gyro's, and the control surface deflection, both of which may need to be reduced when you switch to a canard configuration.

Yaw control:

From my experimentation's with VTOL's I have found that two Yaw vanes are adequate, and in the case of the TurboPlane, only one Yaw control vane was required for adequate control.

Re: mixing: since you're using external gyros, it will require the use of external mixers, unless you want to offset the gyros at an angle between 45 and 65°, this off-setting of the gyros will allow you to use an internal mix within the transmitter. I would recommend the use of two external mixers one to mix the pitch and roll, and the other one to mix the Yaw into two of the vanes at the lower end.

I'm rather short on time this morning, if you have further questions about this please ask.

Kelly

Hi Sam,

I've been thinking about the canard configuration of the Japanese sphere, it appears that this craft is a multipurpose device and one of the considerations of the designers was how to extricate itself if it was trapped in a hole or up against a wall or a log. the use of the canard control vanes and the control vanes on the lower part of the craft creates a huge torque that can allow the craft to rotate around on the ground. For what we're trying to accomplish I think copying the design of the Japanese ball may be over complex. It appears that the canard configuration alone from your report may be less stable than your original setup.

My guess would be the simplest and easiest thing to test would be to start by using a single electric motor and one large propeller, this would allow you to use the spoilers effectively.

As for the control vanes you may find that using two Yaw vanes at the top, with to servos and utilizing two servos at the bottom to control the pitch and roll vanes would give you that best control, the least amount weight, the least amount of trouble configuring the control surfaces.

It would be very nice to stay away from using any type of mixing or mixers if possible. It may also be very difficult to try to copy the Japanese ball, without utilizing a sophisticated specialized stabilization control board.

Go back and look at the crazy ball videos on this posting, notice that this craft only uses elevons for control. This device can be reconfigured with just with pitch and roll control vanes using two servos.

The Yaw control axis can be handled by two Yaw control vanes, using two servos just below the propeller. also notice that there's a shroud around the propeller on the crazy ball. The shroud could be removed and a set of eight spoilers added to the crazy ball and I think you'd have a nice line aircraft. You could wrap the outer surfaces with your carbon fiber rods to give it a structure that would be similar to the Japanese ball.

I'm a strong proponent of using the fewest possible parts to make any type of aircraft. I think the Japanese ball as we are observing it is overly complex for the task of simply flying around mostly in hover and with the addition of the spoilers perhaps a reasonable amount of Sprint or horizontal flight.

Sam,

Here's a few photos of my latest project I'm working on. I'm using a Tower Hobbies .75 engine. As you can see the anti-torque vanes and the motor motor mount were carved from a 2 x 4 Cedar block. The vanes are cut to an angle of 15° but they will need to have additional material attached to increase their area by 100%.

I have a special testing stand that has a bearing support that I strapped the device to determine the amount of surface area of the anti-torque vanes to offset the engine torque at full power. It will only require one small movable vane that can be positioned either between the two vanes or attach to the lower side of one of the vanes to control the Yaw axis.

You can see that I have made cardboard mockups of the lower pitch and roll control surfaces in order to be able to get a visual idea how to place various controls. This cardboard will later be replaced with Coroplastic

I thought I'd show you these photos to give you the idea that the torque vanes don't have to be placed perfectly symmetrically under the propeller. I chose this configuration because I can form it into a box structure which will later support a small riser that will hold an FMA copilot which has an infrared sensor that will detect the earth's horizon. This form a stabilization has proven to work very well on helicopters so I thought I'd give it a try on this device.

I only need one Gyro to control the small Yaw control vane.

Regards,

Kelly