Shop OBEX P1 Docs P2 Docs Learn Events
memsic sensor and rotation — Parallax Forums

memsic sensor and rotation

ArchiverArchiver Posts: 46,084
edited 2003-09-14 22:36 in General Discussion
Hello,
I'm trying to use the Memsic accelerometer to measure rotation, so I can
know the direction my robot is facing on a flat surface. The basic stamp
test programs that come with parallax version of the sensor work very well
for x and y tilt, but using the rotation test program doesn't yield any
change when I turn the circuit around. That's not entirely true, values
change as I turn it head over heals (pitch), or side over heals (roll), but
not as I turn it in a circle on a flat surface (yaw), which is what I want.
The sensor is perpendicular to the plane as the instructions require. Still,
no results. Does anyone have any advice on this matter? Perhaps I
misunderstood which type of rotation the sensor measures. Please note I am
using BS2sx, and have adjusted all the values in the test programs
accordingly...of course there's always the chance that I overlooked one.
Thanks,
Nick

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2003-09-14 21:22
    Nick,

    The Memsic 2125 measures change in gravity. When mounted flat
    (parallel to earth) it senses pitch and roll, but only up to plus or
    minus 60 degrees. Their rotation example shows how to overcome this
    limit by mounting the sensor perpendicular to the earth and using the
    two axis' outputs to achieve a full 360 value in either pitch or
    roll. By turning your circuit around its y axis you never change
    either plane with respect to the earth. It sounds like you need
    something more like a digital compass perhaps? I don't know any
    references for these offhand but I'm sure other folks here could help.

    Trent

    --- In basicstamps@yahoogroups.com, nick stedman <nickstedman@s...>
    wrote:
    > Hello,
    > I'm trying to use the Memsic accelerometer to measure rotation, so
    I can
    > know the direction my robot is facing on a flat surface. The basic
    stamp
    > test programs that come with parallax version of the sensor work
    very well
    > for x and y tilt, but using the rotation test program doesn't yield
    any
    > change when I turn the circuit around. That's not entirely true,
    values
    > change as I turn it head over heals (pitch), or side over heals
    (roll), but
    > not as I turn it in a circle on a flat surface (yaw), which is what
    I want.
    > The sensor is perpendicular to the plane as the instructions
    require. Still,
    > no results. Does anyone have any advice on this matter? Perhaps I
    > misunderstood which type of rotation the sensor measures. Please
    note I am
    > using BS2sx, and have adjusted all the values in the test programs
    > accordingly...of course there's always the chance that I overlooked
    one.
    > Thanks,
    > Nick
  • ArchiverArchiver Posts: 46,084
    edited 2003-09-14 22:36
    The memsic measures tilt with respect only in the plane of gravity (up
    and down, to those not on Earth). So if your Memsic is horizontal, it
    will measure inertial acceleration, but not tilt or rotation. I hope
    that helps.

    On Sunday, September 14, 2003, at 03:56 PM, nick stedman wrote:

    > Hello,
    > I'm trying to use the Memsic accelerometer to measure rotation, so I
    > can
    > know the direction my robot is facing on a flat surface. The basic
    > stamp
    > test programs that come with parallax version of the sensor work very
    > well
    > for x and y tilt, but using the rotation test program doesn't yield any
    > change when I turn the circuit around. That's not entirely true, values
    > change as I turn it head over heals (pitch), or side over heals
    > (roll), but
    > not as I turn it in a circle on a flat surface (yaw), which is what I
    > want.
    > The sensor is perpendicular to the plane as the instructions require.
    > Still,
    > no results. Does anyone have any advice on this matter? Perhaps I
    > misunderstood which type of rotation the sensor measures. Please note
    > I am
    > using BS2sx, and have adjusted all the values in the test programs
    > accordingly...of course there's always the chance that I overlooked
    > one.
    > Thanks,
    > Nick
    >
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject
    > and Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to
    > http://docs.yahoo.com/info/terms/
    >
    >
Sign In or Register to comment.