View Full Version : Help! Need help in building electronic gyro
zed1234
Sep 23, 2008, 01:50 PM
Hello friends,
I need a electronic gyro for project of flying ball. Please help I need the circuit diagram, parts list,and other data or where I can get plans and/or information on making a
homemade Gyro?Thanks in advance.
fredhs
Sep 24, 2008, 03:39 AM
Hi,
although not totally impossible, building the actual gyro from scratch is not really within the realm of a hobbyist.
For many applications, you will need a gyro combined with an accelerometer. Gyros and accelerometers are commercially availabe at reasonable cost. Here is a well documented project: http://www.dprg.org/projects/2003-01a/
Regards,
Fred
Brandano
Sep 24, 2008, 05:49 AM
could two linear accelerometers be used as a gyro by comparing differences in acceleration and integrating the deducted turn rate?
arocholl
Sep 24, 2008, 07:13 AM
could two linear accelerometers be used as a gyro by comparing differences in acceleration and integrating the deducted turn rate?
No. Google for "Kalman filter" to understand it better. Accelerometer are impacted by *any* acceleration whereas gyro aren't, they only measure turn rate. Both elements are required for model control.
arocholl
Sep 24, 2008, 08:33 AM
Hello friends,
I need a electronic gyro for project of flying ball. Please help I need the circuit diagram, parts list,and other data or where I can get plans and/or information on making a
homemade Gyro?Thanks in advance.
http://cgi.ebay.com/Angular-Rate-Sensor-XV-3500CB-Gyroscope-Prototype-PCB_W0QQitemZ230292430684QQihZ013QQcategoryZ36327Q QcmdZViewItemQQ_trksidZp1713.m153.l1262
Brandano
Sep 24, 2008, 10:54 AM
arocholl, I understand that, I was only considering if it was possible to use two linear accelerometers rigidly connected and sensing along parallel and spaced vectors. any linear acceleration will ideally register on both accelerometers with the same entity, while a rotational component perpendicular to the sensing direction will measure as a difference between the two recorded accelerations. As such you can detect instant angular accelerations, however this won't give you a rate of rotation. To get that you need to integrate accelerations over time, but I believe this too is technically feasible. Could this be the method used by MEMS gyroscopes? as opposed to the piezoelectric gyroscopes?
Blue Sky
Sep 26, 2008, 07:07 PM
arocholl, I understand that, I was only considering if it was possible to use two linear accelerometers rigidly connected and sensing along parallel and spaced vectors. any linear acceleration will ideally register on both accelerometers with the same entity, while a rotational component perpendicular to the sensing direction will measure as a difference between the two recorded accelerations. As such you can detect instant angular accelerations, however this won't give you a rate of rotation. To get that you need to integrate accelerations over time, but I believe this too is technically feasible. Could this be the method used by MEMS gyroscopes? as opposed to the piezoelectric gyroscopes?
I've wondered about doing that myself.
In my own thought experiment I'd locate one 3 axis accelerometer at the CG and
then put an accelerometer in the tail and one in a wingtip.
You would take the difference between the outlying and center accelerometers to give you the rate of rotational acceleration about the pitch and roll axis respectively.
You could also derive for yaw. You would need to integrate the readings to get the position.
A possible advantage would be gravity as an absolute reference for the sensors. Yes, I understand that the accelerometer can't tell the difference between gravitational acceleration and acceleration do to motion, but that ignores the fact that we have more than just a single isolated reading with no other frame of reference. In the real world you have a history of measurements, flight dynamics for the plane, expected response to commands, other sensor measurements including things like change in airspeed, change in altitude, GPS and whatever extra bells and whistles you might have. So I don't see why you couldn't do it.
On the down side, it would be more complex and gyros generally have a faster response. In that sense, accelerometers and gyros nicely complement each other.
Be aware this is all armchair stuff for me and my opinion should probably not be given the same weight as others with more direct experience.
-Dave
teamdavey
Sep 26, 2008, 07:13 PM
"Could this be the method used by MEMS gyroscopes? as opposed to the piezoelectric gyroscopes?"
No, they all sence the effect of coriolis forces on a mass in rotational motion. Both piezo and MEMs utilize vebrating elements - one a bending beam type the other a rotationally vibrating mass. Google MEMs rate sensors for a full description complete with math.
mjsas
Sep 27, 2008, 03:34 PM
The WII controller has accelerometers in it and they can be hacked for some uses.
The EK2 -0702A mixing receiver has two circuit boards in it, the first is a standard receiver and the second has a solid state gyro, micro, BEC, and esc's. You can find the Esky 4 in 1 receivers for less then the cost of a gyro.
Also see http://www.geology.smu.edu/~dpa-www/robo/nbot/ for information on using gyros and accelerometers.
rich smith
Sep 27, 2008, 05:45 PM
Blue Arrow Nano Gyro 2gram $26
http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=4320
You cannot build one for this price.
Hello friends,
I need a electronic gyro for project of flying ball. Please help I need the circuit diagram, parts list,and other data or where I can get plans and/or information on making a
homemade Gyro?Thanks in advance.
rc-conk
Sep 28, 2008, 06:01 AM
try using ones in a ps3 controller they have motion sensors?
rich smith
Sep 28, 2008, 01:46 PM
The WII uses 3-axis ADXL330 and PS3 2-axis ADXL203 accelerometers. As mentioned above do not function as gyro. Also they cost a lot and take many hours of work to play with. (many hours of play to work with?) :)
try using ones in a ps3 controller they have motion sensors?
vBulletin® Copyright ©2000-2009, Jelsoft Enterprises Ltd.