Subject:
|
Re: FIber-optic rotation sensor
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 22 Dec 1999 21:08:27 GMT
|
Viewed:
|
902 times
|
| |
|
|
In lugnet.robotics, Iain Hendry writes:
> I'm wondering, what is the resultant "resolution" (ie, counts per rotation)
> of the fibre-optic device? What sort of raw values does it return, as you
> rotate the axle?
>
> Iain
Try this little experiment: Configure one of the RCX inputs as a light sensor
type (to make it a powered input) and set the data mode to RAW. This lets you
see the results on both the RCX LCD and the fiber-optic strands. Turn the
axle until the first strand lights up - the RAW reading goes to
(approximately) 530. If you turn it some more, the light goes off and the RAW
reading shoots up to 1015. It goes back to 530 again when the next strand
lights up. So now you can program the RCX to recognize whether the fiber-
optic strand is on or off (2 states). Multiply that by 8 strand positions and
you've got 16 counts per rotation, or a resolution of 22.5 degrees.
Incidentally, this is exactly the same as the rotation sensor. (Maybe the
internal mechanisms are similar?) So for most applications, the FOS will work
just as fine as the rotation sensor. Unfortunately, though, it can't handle
rotation speeds as high (at least in my programs - maybe someone has had
better luck with legOS), so if you want fast rotation recognition you'll need
a rotation sensor.
Ian
|
|
Message is in Reply To:
 | | Re: FIber-optic rotation sensor
|
| Mario Ferrari wrote in message ... (...) I'm wondering, what is the resultant "resolution" (ie, counts per rotation) of the fibre-optic device? What sort of raw values does it return, as you rotate the axle? Iain -- tokama (URL) UIN 5362826 AIM (...) (25 years ago, 22-Dec-99, to lugnet.robotics)
|
6 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
