Gyro?

azmax100

Hi all.

I am in the learning stage of some electronic parts.

Can anybody tell me in the easiest way to understand what is gyro and what is the funtion.

How is this thing can be implement in robotic and what the purpouse.

I hope i'm not asking·so much. Sure somebody out there have some experience with Gyro.

If u dont mind letting me know what are your project that use gyro. Just wabt to have some knowledge and idea.

Thanks in advance.

James Long

Azmax (you didn't give a real name)

There are different type of gyro.

Mechanical- is a rotating balanced mass which is suspended in a 2 axis movable frame. When up to speed the gyro will keep it's relative position to where it was started. If you turn and till your robot (or other unit being built) the gyro will stay in the same orientation as it was. The frame has sensors on it to tell you how much you moved in each axis. Most of the time this is calculated out in degrees. Like you turned 30 degrees to the left and tilted 5 degrees forward.

Mechanical gyro are extremely expensive and are heavy. They are very complex electro/mechanical devices.



Electronic gyro- is a very precisly machined (micro machined) beam that looks similar to a balance (the ones on attourneys logos) These micromachined parts can flex with motion, and give a electrical signal from the bending moment. They are accurate, but only while in motion. There are different electronic gyro with multiple axis.

Accelerometer- is similar to an electronic gyro. It senses acceleration forces placed upon it. It doesn't how ever actually measure rotational forces.



All of the above are used to calculate the relative movement of a robot. This is for the robot to know balance, location, orientation, etc.

The user decides on how the gyro(or accelerometer) will be used. There are many different uses...

I personally have a design with gyros to give me orientation for a automated flight control. The gyro and accelerometers tell me when the plane moves and I can calculate the attitude of flight (whether is it upside down or right side up, etc)

I don't know if all of this helps. It is·a long post and may be more indepth than you wanted.

It would also help you to do some research on the internet...there is plenty to read out there.

James L


Post Edited (James Long) : 4/26/2007 2:07:26 PM GMT

Dave Hein

James,

Can you provide more information on your gyro design? I am interested in building a control system for a model rocket to keep it flying straight up. I know that one of the problems with electronic gyros is drift. Electronic gyros measure rotational acceleration. Angular position is obtained by integrating the output of the gyro twice. Even a small DC error in acceleration will quickly integrate to a large positional error. I am interested in what you have learned in working with electronic gyros.

Dave

James Long

Dave,

I was in Tripoli (actually my wife ran HQ for about two years) and I know about rockets in depth.

I haven't found a real good way to use electronic gyros in rockets. There is a reason for this that I'll explain.

Accelerometers are used to re-enforce the attitude of the craft and to compare with the gyro's output, and to read position relative to ground. Because a rocket accelerates higher than other vehicles the accelerometer will not be effective in any axis other than the one that is accelerating. (there will be virtually no sideways acceleration)

The gyro would have to be pretty sensitive to see the rotation of the rocket which is usually pretty slow (degrees per second). Most Rockets turn from 10 to 35 degrees through their flight profile(high speed/high acceleration). I'm using the typical flight....there are times in Kansas that rockets will turn a true 90 degrees while under power....but I wouldn't be launching then. Plus you do not want to correct in this condition....it would be unlikely you ever got it back.....it would be a long way down range if the flight were to go vertical........the parachute would only take it farther away.

If the flight takes 1 min to the maximum angle of 35 degrees(while under power/high speed coast)·from vertical that is only·0.58 degrees per second. This would have to be·a very sensitive gyro chip.

Now if we were trying to correct for rod whip (rocket whipping off the rod at launch) the gyro would probably do a great job.

But correcting for weather vaining.....I'm not sure you could get a clean enough signal to do the job. It would be based on the design of the noise canceling circuit and the software written for the angular movement calculation.

I'm basing this all on high power rocketry and do not know what specific application you are looking for.

I think you could make the flight profile much better....but I do not think you would ever get it perfectly vertical. A sun seeker at high noon would be better. (George Gassaway)

James L

Post Edited (James Long) : 4/26/2007 4:59:51 PM GMT

Dave Hein

James,

Thanks for the info. I have two projects in mind -- one is use control fins to correct for weather-cocking and the other is to build a rocket that uses vectored thrust to stay vertical. I am familiar with the sun-seeker rocket. I have thought about using several photo-sensors to detect the position of the sun. I also have considered magnetic field sensors and horizon detection.

Another possiblity would be to build a mechanical gyro. However, this seems fairly complicated and it would probably take some time to get it to work correctly.

Dave

James Long

Dave,

I don't know of a good way to do it....I've not experiemented with magneto type sensors. Spark Fun sells a few (www.sparkfun.com) and that seems to be a good way to know the horizon.....but like I said...I'm ignorant about those sensors.

The sun seeker works good as long as the sun is high and the wind cooperates....but it has flaws too.

the FMA copilot would be another system I would consider....although pretty heavy and bulky for a rocket....seems like a pretty good idea.(you mentioned horizon detection)

I like the vectored thrust idea.....pretty hard to implement.....so keep me informed of your progress. I'm a rocketeer at heart, but have a new business to keep me busy.

Sorry I couldn't provide a better idea.....or a solution.

James L

Dave Hein

There is a web page on "Building a Balancing Scooter" that shows a good application for gyros.· It has several good links at the bottom of the page on balancing robots.· The one on nBot is very informative.· There is also another one on LegWay, which is a balancing robot built from the LEGO robot components.

The web page is located at http://www.tlb.org/scooter.html

There are a number of posts about balancing robots on the Parallax forum.· Just search for the word "balancing".

Brooke's Sensors Page has lots of good information about gyros and other kinds of sensors.· His page is located at
http://www.pacificsites.com/~brooke/Sensors.shtml

He has a list of good links on MEMS, Quartz rate sensors, mechanical gyros, fiber optic and others.· I found the fiber optic gyros (FOG) to be very interesting.· They have very low drift, but they are still very expensive.· Fiber optic gyros or ring laser gyros could be integrated into a chip in a few years, and the cost will drop dramatically.

For now, it looks like the rate sensors like those used on model helicopters will work for me.· They can drift a degree every few seconds, but they only need to work for a few seconds in my application.

Dave



Post Edited (Dave Hein) : 4/29/2007 2:19:16 AM GMT