Electronic Gyro Rocket Roll Stabilization System

[gyro32]

Hello everyone,

I would like to introduce you to a gyroscopic stabilization system which I made.The system is designed to prevent the rocket from rolling.
The idea is to use it with rockets that have a built-in camera or other equipment that does not withstand the rolling of the rocket during its flight.
The system uses a MEMS gyroscope and digital circuits.RC servo motors were used for this test model.
The electronic system can be customized for all types of rockets.

I would love to hear your opinion on this stabilization system

The video shows a static test of the system.

 

[Kelly]

Looks very cool.
I've never designed or used anything like this, so I can't give much valuable critique, but my initial questions would be:
What is the speed/responsiveness of the system? Controlling a rocket in flight has much more rigorous requirements than controlling roll in the gentle breeze that that fan is providing. Are the servos strong enough as well? How is the control algorithm tuned for an actual flight?
Is this a unit that you plan to commercialize? Will you be posting a build log, software/schematics?

 

[FredA]

Cool - Mate it to a ring-tail and go fly.....

 

[OverTheTop]

There are some other comments on the Australian Forum on this as well:

https://forum.ausrocketry.com/viewtopic.php?f=32&t=6752

 

[gyro32]

Looks very cool.
I've never designed or used anything like this, so I can't give much valuable critique, but my initial questions would be:
What is the speed/responsiveness of the system? Controlling a rocket in flight has much more rigorous requirements than controlling roll in the gentle breeze that that fan is providing. Are the servos strong enough as well? How is the control algorithm tuned for an actual flight?
Is this a unit that you plan to commercialize? Will you be posting a build log, software/schematics?

This is just a static test model.It is NOT intended for any dynamic testing.In this static test model the gyroscope cannot withstand over 100 G acceleration.Inside electronics don't follow COTS and MIL-STD rules.Classic RC servo motors were used (BMS-620MG).

If used as a finished product ready for installation in a rocket, custom made or industrial servo motors would be used, gyroscope and electronics would be made according to COTS or MIL-STD standards to be able to meet the requirements required of one such rocket system.The control algorithm is in fact an electronic rolleron.Which works according to all the laws by which rolleron works.

This actual test unit is not intended for commercialization,because the built-in components do not meet the requirements required for such a system.
Since I have no intention of making and selling finished hardware products,for the reason that I do not see that demand for such a system that would justify the cost of making the finished hardware product for sale.The only thing that would come into consideration is the sale of licenses (schematic,PCB and software) for the production of finished products.My main project I am currently working on is the vertical active stabilization system,therefore this project is not my primary one and I would do it exclusively on an individual order.

 

[OverTheTop]

Don't knock the quality of regular servos, or other electronics for this purpose. If you are looking at normal HPR flights they are in the region of 5-20G typically for peak acceleration. It is only the highly optimised speedbirds that really peg the high accelerations. Things have to be highly optimised to achieve 100G you discuss above.

You do need to check the commercial hardware for quality though. I had a power converter go short circuit on the first flight of my Vertical Trajectory System. The boost dislodged a solderball and shorted out my 12V-5V converter. The servos were fed 12V and promptly died, some going to the mechanical limits (I had set only 4 degrees as the endstops on that flight I think). I would highly recommend using mechanical endstops for your servos but program them to not hit the endstops in normal operation. That keeps current and power dissipation down.

 

[gyro32]

You are all right about what you are saying, but for this test model I have shown, there is no need to use high end components.This is a simple static model.
It's not a problem for me to make an electronic circuit that can withstand firing from a howitzer.The only question is whether it is necessary to spend money on such a simple static test.Each assembly has its own purpose and requirements, so the necessary components are used. Everything else beyond that is a waste of resources and money.
I can advise you to potting all electronics in the future with low viscosity epoxy/PU resin, preferably in a vacuum chamber.

 

[OverTheTop]

For static testing the chepest $1.50 servos from China are fine . Been there, done that.

 

[Kelly]

I was more interested in the torque ratings for the servos, than the G ratings. How much torque does it take to rotate and hold a fin in the airstream when you are going Mach 1?

 

[Loathing]

Interesting idea. I would be concerned about the amount of authority that the fins have and the data acquisition rate. I could see that you could get behind the curve pretty rapidly. Would smaller changes in the fin angle be better? I think the smaller fin angle would allow for slower data acquisition rates. While this is a "proof of concept" - it looks really interesting.

 

[Dan Griffing]

Hello everyone,

I would like to introduce you to a gyroscopic stabilization system which I made.The system is designed to prevent the rocket from rolling.
The idea is to use it with rockets that have a built-in camera or other equipment that does not withstand the rolling of the rocket during its flight.
The system uses a MEMS gyroscope and digital circuits.RC servo motors were used for this test model.
The electronic system can be customized for all types of rockets.

I would love to hear your opinion on this stabilization system

The video shows a static test of the system.

Can your same concept be extended to use four small canard fins and additional control mode following the spin stabilization for verticality and directional control?

Conceptually it’s only adding a second mode of control where opposing fins turn at the same angle N/S and E/W instead all turning together at the same CW/CCW angle.

But there is considerable math complexity in translating the feedback from 3 mode gyroscope and magnetometer sensors, (with dead time delay compensation) into PID angular position changes in each of the canard fin servos.

Several others on TRF have been working to develop similar systems for over a decade, so you may leverage from their work and learn from their experiences as well.

 

[JimJarvis50]

Hello Gyro32,

I've been flying what I now call an "orientation control system" for about 6 years now. It does yaw/pitch and roll control and actually works quite well. One thing I can tell you for sure is that it's great fun! Keep at it.

With respect to your videos, a couple of comments come to mind. First, the canard deflection is much higher than what you will actually want to use in flight. When you turn the model like in the video, you won't want a deflection of more than a degree or two. Second, if you plan to fly more than a model rocket, you will need to figure out how to support the servo axle. They are not made for lateral forces and will fail quickly at higher speed. Protecting the servos from flight and landing forces is perhaps the most difficult single thing needed to make a system that works.

I have a thread here on TRF that documents most of what I have done in the development of my system. It's been a long journey!

https://www.rocketryforum.com/threads/i-could-use-just-a-little-guidance.122042/
I have also posted probably 20+ videos of flights. You can search YouTube for jiminaus50 and you'll find them. Most are attempting to control to a vertical orientation starting from some launch angle. More recent flights, though, control the tilt and bearing of the rocket through the entire flight. An example is attached.

Jim

 

[jgavlik]

Gyro32 & JimJarvis50,

Thank you, both, for attempting (and succeeding) to advance the state of the art of this hobby!!!

Joe Bernard...are you watching??

-john

 

[schworer]

In 2016 and 2018 I flew a single servo roll control project/experiment at Balls. The roll gyro was an inexpensive
R/C stabilizer with the rudder/yaw output to the servo. The stabilizer has some drift, next year it will fly with a much more responsive older R/C helo pizzo gyro. Below the photos is a link to a YouTube of both test flights, and I've attached
an edited version of a much longer presentation I made for NOVAAR a while back.

Bill Schworer

 

[jgavlik]

In 2016 and 2018 I flew a single servo roll control project/experiment at Balls. The roll gyro was an inexpensive
R/C stabilizer with the rudder/yaw output to the servo. The stabilizer has some drift, next year it will fly with a much more responsive older R/C helo pizzo gyro. Below the photos is a link to a YouTube of both test flights, and I've attached
an edited version of a much longer presentation I made for NOVAAR a while back.

Bill Schworer

View attachment 493676View attachment 493677

Bill,

Thanks for this experiment and especially your excellent data analysis. This is what the hobby needs more of...john

 

[gyro32]

@ Kelly
That should be calculated.
But take the IRIS-T missile as a reference.The picture has information that might interest you.


_____________________________________________________________________________________________________________________________________________________________________________

@ Loathing

This static model is just this is a proof of concept.Therefore, some things are omitted because they are irrelevant in this type of test.
_____________________________________________________________________________________________________________________________________________________________________________

@ Dan Griffing

I don't think it's possible to use the current principle of rolling stabilization for vertical stabilization and directional control.A completely different management system will be used for this vertical stabilization project that I am currently working on. This is a simple system to prevent the rocket from rolling, nothing more nothing less.
______________________________________________________________________________________________________________________________________________________________________________

@JimJarvis50

This is just a static test model, the canard deflection is large due to the low air velocity generated by the fan.The change in canard deflection must be linear with the acceleration of the rocket.
I will read all the pages you have published.
______________________________________________________________________________________________________________________________________________________________________________

@ jgavlik

I think the results would be better if there were no stupid legal restrictions on the use of some technologies.For example for vertical stabilization of rockets could be used laser beam riding.Aim the laser at the sky, and the rocket flies in a straight line.
One of the projects I'm working on outside of these shows is a laser point tracking system (Quadrant photodiode used).Which would be used in robotics or an automated guided vehicle.But it could be used to accurately return the rocket to the desired location if there were no legal limitations ...


______________________________________________________________________________________________________________________________________________________________________________

@ schworer

I think the RC heli gyro is not adequate for use in rocketry.
Since I'm not currently able to fire high power rockets,if you're interested, I could make (Lite version) rocket roll stabilization system at my own expense.
It's just up to you to put it in the rocket and fire it.

 

[schworer]

<SNIP> I think the RC heli gyro is not adequate for use in rocketry.
Since I'm not currently able to fire high power rockets,if you're interested, I could make (Lite version) rocket roll stabilization system at my own expense. It's just up to you to put it in the rocket and fire it. <UNSNIP>

Unlike the R/C stabilizer arrangement that doesnt require a receiver's output, the heli pizzo heading hold gyro requires a continuous pulse wave modulated input from an onboard R/C receiver in order to power up the servo and to send centering and correction signals. Once the receiver on the rocket receives the initial handshake signal from the R/C transmitter the canards center and the transmitter can be shut off. From this point on, as long as power remains continuously applied the receiver will output PWM control signals to the pizzo gyro and the rocket will "hold heading". I've bench tested this several times, but haven't flown it yet. I'd be interested in test flying your control board sometime in the spring, PM me and we can discuss off forum. Tnx, Bill

 

[FredA]

I think the results would be better if there were no stupid legal restrictions on the use of some technologies.For example for vertical stabilization of rockets could be used laser beam riding.Aim the laser at the sky, and the rocket flies in a straight line.
One of the projects I'm working on outside of these shows is a laser point tracking system (Quadrant photodiode used).Which would be used in robotics or an automated guided vehicle.But it could be used to accurately return the rocket to the desired location if there were no legal limitations ...

I don't think you could technically do this but I also don't know of any laws stopping you.
Can you show this how/why you think this is illegal?

 

[JimJarvis50]

I don't think you could technically do this but I also don't know of any laws stopping you.
Can you show this how/why you think this is illegal?

Having been through the process, I can tell you that the powers-that-be would not be happy with the use of a laser, as it would allow (in theory) a rocket to fly towards a target. Don't do this. There is no problem, however, with controlling the orientation of the rocket.

Fred, great idea about using a ring fin. Never thought of that.

Gyro32, it is quite possible to use RC-quality equipment for these systems. Mine is based on MPU-6000 gyros (better are now available) and RC-type digital servos. I fly on N motors and the system works reasonably well.

Jim

 

[gyro32]

I don't think you could technically do this but I also don't know of any laws stopping you.
Can you show this how/why you think this is illegal?

Today when technology is available everything is possible, the only question is how much is $.Semi-active or beam riding guidance is nothing special, and can be achieved without large investments.Depending on the country you live in and what the regulations are. Somewhere such technology for civilian purposes is not regulated and somewhere yes.
_______________________________________________________________________________________________________________________________________________________________

@ JimJarvis50

When it comes to rocket systems, I still apply different standards than in the classic hobby. It may be excessive, but I think that in rocket technology there are still some other forces in relation to the classic hobby.Maybe I look at some things with different eyes.
This model I showed uses a gyroscope that costs 250$.Someone will say it’s too much, maybe it is but it works.
For that reason for IMU I would not go below ADIS16477.

 

[cautery]

I was more interested in the torque ratings for the servos, than the G ratings. How much torque does it take to rotate and hold a fin in the airstream when you are going Mach 1?

DISCLAIMER: I have zero experience in active attitude control in rocketry.

Intuitive thought brings me to these conclusions (mostly in relation to the SAS and stabilator used on UH-60 Blackhawks):

1) Torque requirements are not linear with respect to velocity. Torque required to achieve a given control surface deflection (should) increase as velocity increases. In mach transition or above speed of sound at a given altitude, control surface location/size/shape will likely also figure in prominently as to their location with respect to the wave front.

2) At a given velocity, torque required to move control surface definitely would depend on control surface design (area, balance above/below shaft station, thickness, profile, et al., ad nauseum.

3) As velocity increases, the net deflection required (should?) decrease, somewhat mitigating the increased torque requirements.

Many, many, many other non-discrete and inter-dependent/inter-related considerations.
Hence the use of extensive simulation, modeling, and actual flight testing IRL...

 

[jderimig]

I was more interested in the torque ratings for the servos, than the G ratings. How much torque does it take to rotate and hold a fin in the airstream when you are going Mach 1?

On the other hand the faster you go the less deflection that is needed to provide a lateral (or angular) correction force. So I propose that the lateral correction force (canard lift) versus torque is independent of speed.

 

[OverTheTop]

If you place the shaft of the servo around the center of lift on the canard (fairly easily calculable) then the torque requirements (and following that the power requirements) greatly reduce. You don't want it directly on the center of lift axis or the backlash in the servo will create chattering issues.

 

[Kelly]

If you place the shaft of the servo around the center of lift then the torque requirements greatly reduce.

Interesting. It seems both surprising (that it doesn't take much torque to turn a canard in a high velocity airstream) and yet, in hindsight, kind of obvious as well. Thanks for pointing that out!

 

[OverTheTop]

The center of lift does waft around a little depending on the planform of the fins, fin profile and air velocity. I chose to use trapezoidal double-diamond on my setup for this very reason. Relatively easy to calculate and allow for shifting of center of lift, especially in the supersonic region.

You will notice the hub is not centered on the fin.

 

[OverTheTop]

Unlike the R/C stabilizer arrangement that doesnt require a receiver's output, the heli pizzo heading hold gyro requires a continuous pulse wave modulated input from an onboard R/C receiver in order to power up the servo and to send centering and correction signals. Once the receiver on the rocket receives the initial handshake signal from the R/C transmitter the canards center and the transmitter can be shut off. From this point on, as long as power remains continuously applied the receiver will output PWM control signals to the pizzo gyro and the rocket will "hold heading". I've bench tested this several times, but haven't flown it yet. I'd be interested in test flying your control board sometime in the spring, PM me and we can discuss off forum. Tnx, Bil.

Alternatively use a receiver that has a "failsafe" mode and will output fixed PWM widths when the input from the transmitter is lost. This works well and is how my Mk 1 Vertical Trajectory System worked. It means a momentary power glitch will not likely affect the system greatly.
https://forum.ausrocketry.com/viewtopic.php?f=56&t=5324
It was interesting to experiment with that system but cheap hardware and good open-source firmware (Ardupilot) have seen me change up my hardware to something much more capable. Now I have just got to get to fly it.
https://ardupilot.org/

 

[thzero]

Did you do a write-up with the ArduPilot version too?

I'm reading through your long writeup.... I need more new skills/knowledge I think!

 

[OverTheTop]

https://forum.ausrocketry.com/viewtopic.php?f=56&t=6632
I should document this one a bit better than what it is on the forum currently.

 

[Spaceman Here]

Hello everyone,

I would like to introduce you to a gyroscopic stabilization system which I made.The system is designed to prevent the rocket from rolling.
The idea is to use it with rockets that have a built-in camera or other equipment that does not withstand the rolling of the rocket during its flight.
The system uses a MEMS gyroscope and digital circuits.RC servo motors were used for this test model.
The electronic system can be customized for all types of rockets.

I would love to hear your opinion on this stabilization system

The video shows a static test of the system.

Good idea! But by reducing spin, won't you lose spin stabilization? Or will your system duplicate that straight as an arrow effect that the spinning provides? Also, if it uses SF, will the rocket run a higher risk of uneven burning? Thanks

 

[Titan II]

The OP has not been here for over two years.

BTW....we do not typically use spin stabilization.