Gyro-controlled Rocket

Dave Hein

I built a gyro-controlled rocket using an SX18 and a 2 dimensional rate gyro from SparkFun.· The rocket is named Quad Pod II, and it uses a gimballed motor mount instead of fins to control the direction of flight.· It is intended to fly at a low speed and altitude.

I've had successful flights on Estes E9 and Apogee F10 motors.· Videos of two of the flights are at
http://www.youtube.com/watch?v=XiVpxxK_PxA·and
http://www.youtube.com/watch?v=wQk2z1nT3cU

I have more information on the rocket at
http://home.swbell.net/davehein/ModX/quadpod_ii.html

A picture of the rocket, a schematic and the source code are attached below.
Dave

Zoot

That is very cool. So without the gyro direction control on the engine, the platform gets really unstable???

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When the going gets weird, the weird turn pro. -- HST

1uffakind.com/robots/povBitMapBuilder.php
1uffakind.com/robots/resistorLadder.php

Dave Hein

Zoot said...
That is very cool. So without the gyro direction control on the engine, the platform gets really unstable???

My first attempt at the Quad Pod did not have active control.· It consisted of a motor mount tube glued to four PVC pipes.· I initially thought that it would be stable, but the test flight proved otherwise.· After the first launches I looked at the force vectors and realized that there was no correcting force to keep the rocket vertical.· One of the early flights is posted at
http://www.youtube.com/watch?v=AkwYps7gdZQ·.

The webpage for the original Quad Pod is at http://home.swbell.net/davehein/ModX/quadpod.html·.

Dave

Zoot

Wow, what a difference the stabilization makes. Nice!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When the going gets weird, the weird turn pro. -- HST

1uffakind.com/robots/povBitMapBuilder.php
1uffakind.com/robots/resistorLadder.php

Joss

I want one.

Brian_B

Good job Dave, now to get it to land on the moon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Thank's Brian



·"Imagination is more important than knowledge..." ·· Albert Einstein

www.PropelX.com

Dave Hein

Thanks, Brian.· I have to work out a few more details, but I'm sure it will be able to land on the moon.· However, I'll have to check with Home Depot to see if they have any space-rated plywood and PVC pipe.

Dave

Brian_B

Dave,
I see that Armadillo Areospace is getting $500K for their test lander, you could get big money for the Qaud Pod.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Thank's Brian



·"Imagination is more important than knowledge..." ·· Albert Einstein

www.PropelX.com

sylvie369

Wow.

Next May, over Memorial Day weekend, WOOSH (NAR #558) will host the National Sport Launch at Bong Recreation Area in southern Wisconsin. Any chance you could make plans to come, and fly that thing for us there?

Dave Hein

sylvie369 said...

Next May, over Memorial Day weekend, WOOSH (NAR #558) will host the National Sport Launch at Bong Recreation Area in southern Wisconsin. Any chance you could make plans to come, and fly that thing for us there?

I might be able to make it to the NSL.· I'll put it on my calendar.
Dave

sylvie369

Dave Hein said...

sylvie369 said...·

Next May, over Memorial Day weekend, WOOSH (NAR #558) will host the National Sport Launch at Bong Recreation Area in southern Wisconsin. Any chance you could make plans to come, and fly that thing for us there?

I might be able to make it to the NSL.· I'll put it on my calendar.
Dave

Great! It'll definitely be a hit.

Timothy D. Swieter

Cool work Dave!

I was watching Armadillo Aerospace videos recently. I see your rocket is ~sort of~ similar. When I was watching the AA videos I trying to think how I could build something smaller to test a gimbal motor mount. It looks like you have done what I was only starting to wonder about. Neat project, very neat.

Question:· Why a gyro and not an accelerometer?· With an accelerometer couldn't you still calculate the vertival access and position the servos?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Timothy D. Swieter, E.I.

www.brilldea.com·- Prop Blade, LED Painter, RGB LEDs, uOLED-IOC, eProto fo SunSPOT, BitScope
www.sxmicro.com - a blog·exploring the SX micro
www.tdswieter.com

Post Edited (Timothy D. Swieter) : 8/16/2008 2:48:47 PM GMT

Seth

I have found that you can get mixed signals with accelerometer. It can be hard to tell if the accelerometer is tilted or if it is accelerating.
If you know that you will only accelerate or tilt, it can work great; But rockets have both acceleration and tilt.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The pessimist is never disappointed.

Dave Hein

Timothy D. Swieter said...
Question:· Why a gyro and not an accelerometer?· With an accelerometer couldn't you still calculate the vertival access and position the servos?

A rate gyro senses angular rotation independent of acceleration.· An accelerometer senses acceleration along a specific axis.· If it is mounted at the center of mass of an object it will not detect rotation, but only acceleration.· If it is mounted away from the center of mass it will detect both linear acceleration and acceleration due to rotation.· I believe two accelerometers would be needed for each dimension to be able to seperate linear acceleration from rotational forces.
Dave

jazzed

Also, MEMS accelerometers are very sensitive to vibration making them almost worthless for guidance control except when paired with a gyro and data processed with a kalman filter.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
--Steve

PJMonty

Dave,

You can power the SX-18 from 3.3 volts and remove one of the regulators from the project. Since you're using linear regulators, they're wasting power in the form of heat. The only catch is you'll need to use the new USB SX-Key to program the SX-18 at the lower voltage. The USB SX-Key is powered from the USB port, not the project.

Thanks,
PeterM

CCraig

Dave,

I really like this project. Very well done.

PJMonty's idea could mean you could use a smaller lighter battery also, but I don't think you're fighting a weight problem.

Again, well done. When you start making a jet pack, let me know.

Chris

Dave Hein

I'm not too worried about weight and power consumption right now.· The rocket has to weigh about 2.5 pounds so it hovers, and doesn't go real high.· The 9-volt battery for the processor last quite a while, but the servos do run down the other battery fairly quickly.· I could run the servos directly off of 6 volts and avoid the regulator.· Four AAA batteries weigh slightly more than one 9-volt battery, and would probably run 1.5 to 2 times as long without the regulator.· The 9-volt batteries work OK for now.

I have run an SX18 at 3.3 volts in a previous project.· This was a camera controller that strobed the shutter switch on an Aiptek camera.· The camera ran at 3.3 volts internally, so I tapped off the power from the camera to run the SX18.· I have a serial SX key, so I had to use the device that Parallax sold to isolate the power from the SX key.· I think its call a key ring, but I may be wrong about the name.· This works OK.· However, the USB key sounds like a better solution.

I do plan to build a more conventional rocket in the future that would fly to 500 feet or so.· I can easily trim down the gimballed motor mount to fit into a 5.5 inch body tube.· The batteries and controller board would be moved up toward the nose cone, and I would run two cables down to the servos.· The goal would be to get the weight down as low as possible, so I would probably go with a single battery.· I might use switching regulators to extend the battery life.

Dave