Reading a gyroscope's orientation in a coil?
dave.harper
orientation, and needed some advice in determining if one method in
particular is sound. (It has been years since I had a class in
mechatronics and/or generators)
Assume a freely rotating gyroscope given an initial rotational velocity
and then allowed to coast (unpowered):
1. If I were to place a permanent magnet on the rotating component
(aka the rotor) so that it's North-South orientation is perpendicular
to the axis of rotation, wouldn't the magnitude of the voltage induced
in the coil depend on the rotor's tilt within the coil? That is,
wouldn't the coil AC voltage be at a maximum when the magnet was
rotating perpendicular to the coil's axis and a minimum when the axis
are aligned? Assume I can account for the effect the slowing speed of
the rotor has on coil voltage.
2. If this theory is sound, what would be the best way to digitally
quantify the magnitude of the AC voltage? The simplest way I can think
of is to rectify it, use smoothing cap, and read it via an ADC. I'm
open to other, more sensitive (and easily accomplished) methods.
If I actually play around with this idea, it will be done on a personal
budget and on my spare time, so please keep that in mind before
recommending extremely expensive or complex ideas.
Thanks in advance for any help!
Dave
John Fields
wrote:
>I've been evaluating different methods to sense a gyroscope's
>orientation, and needed some advice in determining if one method in
>particular is sound. (It has been years since I had a class in
>mechatronics and/or generators)
>
>Assume a freely rotating gyroscope given an initial rotational velocity
>and then allowed to coast (unpowered):
>
>1. If I were to place a permanent magnet on the rotating component
>(aka the rotor) so that it's North-South orientation is perpendicular
>to the axis of rotation, wouldn't the magnitude of the voltage induced
>in the coil depend on the rotor's tilt within the coil? That is,
>wouldn't the coil AC voltage be at a maximum when the magnet was
>rotating perpendicular to the coil's axis and a minimum when the axis
>are aligned?
---
Yes
---
> Assume I can account for the effect the slowing speed of
>the rotor has on coil voltage.
---
OK
---
>2. If this theory is sound, what would be the best way to digitally
>quantify the magnitude of the AC voltage? The simplest way I can think
>of is to rectify it, use smoothing cap, and read it via an ADC. I'm
>open to other, more sensitive (and easily accomplished) methods.
---
Don't rectify it, don't smooth it, just read it with an ADC.
---
>If I actually play around with this idea, it will be done on a personal
>budget and on my spare time, so please keep that in mind before
>recommending extremely expensive or complex ideas.
--
John Fields
Professional Circuit Designer
Rich Grise
Well, it has to have an axis of some kind, that means some kind of axle,
with bearings at the end, and since you say "freely rotating", that
means it's on a gimbal of some kind. Just put positional sensors at the
gimbal's bearings, and maybe slip rings if it has to go >= 360°.
I think sensing a PM by one coil would be more of a hassle than it's
worth, but the obvious answer is, do it and see what happens! :-)
Have Fun!
Rich
CWatters
"dave.harper" <dave....@gmail.com> wrote in message
news:1138906129.7...@z14g2000cwz.googlegroups.com...
> I've been evaluating different methods to sense a gyroscope's
> orientation, and needed some advice in determining if one method in
> particular is sound. (It has been years since I had a class in
> mechatronics and/or generators)
Not sure if this is of interest but...
It's possible to buy very very small Piezoelectric vibrating Gyroscopes for
use in model helicopters and the like. I have one installed in a 1/5th scale
Nieuport 11 to counter a the yaw motion when the tail comes up on take off.
Some are about the size of a large sugar lump and only weigh a few grams.
It's should be possible to obtain the sensors on their own.
Examples...
http://www.mh.ttu.ee/risto/rc/electronics/radio/gyro.htm
http://www.gyroscope.com/d.asp?product=PIEZO
http://www.helimax-rc.com/accys/hmxm1010-rcrm.html
http://www.siliconsensing.com/
Comparison...of different technology.. (bit basic)..
http://www.spp.co.jp/sssj/sindoue.html
In general the larger and more expensive units have less drift and noise.
Don Foreman
wrote:
It'll work -- but metal in the region can cause eddy-current drag on
the rotor.
>Dave
dave.harper
CWatters wrote:
> "dave.harper" <dave....@gmail.com> wrote in message
> news:1138906129.7...@z14g2000cwz.googlegroups.com...
> > I've been evaluating different methods to sense a gyroscope's
> > orientation, and needed some advice in determining if one method in
> > particular is sound. (It has been years since I had a class in
> > mechatronics and/or generators)
>
> Not sure if this is of interest but...
>
> It's possible to buy very very small Piezoelectric vibrating Gyroscopes for
> use in model helicopters and the like. I have one installed in a 1/5th scale
> Nieuport 11 to counter a the yaw motion when the tail comes up on take off.
> Some are about the size of a large sugar lump and only weigh a few grams.
> It's should be possible to obtain the sensors on their own.
These are rate sensors, correct? If I wanted an accurate orientation
reading, rate sensors would work so long as I gave it an initial
orientation and integrated them (along with some tensor math) in order
to determine my position. However, aren't these sensors somewhat
noisy? I believe the estimated and actual orientation would diverge
(drift) due to noise integration if I used them for more than a couple
seconds.
Am I correct in this statement?
Dave
CWatters
>
> These are rate sensors, correct? If I wanted an accurate orientation
> reading, rate sensors would work so long as I gave it an initial
> orientation and integrated them (along with some tensor math) in order
> to determine my position. However, aren't these sensors somewhat
> noisy? I believe the estimated and actual orientation would diverge
> (drift) due to noise integration if I used them for more than a couple
> seconds.
>
> Am I correct in this statement?
>
I believe you are correct yes.
dave madden
and accelerometers. I haven't used those chips, but I did have a
small PC board fabricated through their service. Otherwise, no
connection.
http://www.sparkfun.com/commerce/categories.php?cPath=23_85
dhm
David Harmon
"dave.harper" <dave....@gmail.com> wrote,
>1. If I were to place a permanent magnet on the rotating component
>(aka the rotor) so that it's North-South orientation is perpendicular
>to the axis of rotation, wouldn't the magnitude of the voltage induced
>in the coil depend on the rotor's tilt within the coil?
Yes. But gyroscopes usually go to great lengths to eliminate
friction and other effects on the rotor, and if you put such a
magnet on the rotor the induced current in every nearby conductive
part will
1. suck rotational energy out that would have to be replaced at a
much greater rate than otherwise and
2. affect the alignment of the gyro and reduce accuracy.
dave.harper
I'm really only looking to have it measure orientation for, say, a
minute (not long term). And I'd like to have it drift less than, say,
5 deg/min. I'm making the assumption that eddies and other EM
interactions shouldn't affect it significantly during that short time
period.
However, since you've brought it up and if you don't mind me asking,
how significant do you think the eddies would be? I'm looking at using
aluminum rings (or possibly polycarbonate if aluminum is a problem).
Secondly, how much "drag" would be induced on the rotor by the current
induced in the coil? Ideally, the current in the coil would be near
zero (just enough to read via an ADC). Shouldn't that minimize the
"drag" on the rotor from the coil?
Thanks in advance for any extra help,
Dave
Sjouke Burry
Seconds for the cheap ones, 10's of minutes for
the very expensive ones.
Also at very high rate of turn some of them
start skipping.
We used a 3 axis magnetic compass together with
GPS to make a guidence system for blind people.
(Sory ,just experimental,not on the market).
That way, you have a consistent(but inaccurate)
knowledge about current orientation.
Jimbo
It doesn't sound like you have the option of changning the gyroscope
itself, but If you do I suggest looking into fiber optics based
gyroscope. There are NO MOVING PARTS and are highly accurate. Of
course just like any technology there are pros and cons so do your
homework. Here are some companies to get you started:
Photonic Systems, Inc. in Billercia, MA
Corrsys-Datron Sensorsystems, Inc. Southfield, MI
Elektron Systems, Inc. St. Petersburg, FL
Phasebridge, Inc. Pasadena, CA
GOOD LUCK!
Jimbo