Subject:
|
Re: 360 deg. swiveling electrical connection
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 3 Feb 1999 23:00:56 GMT
|
Reply-To:
|
bryan.beatty@autodesk.STOPSPAMcom
|
Viewed:
|
1121 times
|
| |
|
|
Ben Kimball wrote:
> Has anyone figured out a good way to maintain an electrical connection
> between two parts separated by a 360 deg. rotating joint? Wire would
> just twist up eventually and pop loose or tear.
Daniel Miller wrote:
>
> On Wed, 3 Feb 1999, Tim McSweeney wrote:
>
> > The other options is to use a wire but not allow the turret to turn more
> > than 180degrees either side of center, if you are at 179 and wish to go
> > to -179 you have to go all the way back around rather than just nipping
> > across those 2degrees. It's slower but a lot easier to implement.
>
> Give it enough wire and there's no reason you couldn't nip over those two
> degrees. On the theory of the more range the better, you could let the
> wire twist some, and get maybe +/-360 or more. Just have it re-center
> after every attack and you should be OK.
Well, if you use one of the long leads and set up the mechanics right,
your turret could probably spin 'round a dozen times with no ill
effects. Suppose your rotating assembly is a turret that turns both
left and right in response to various environmental stimuli. Its
rotation angle therefore does somewhat of a random walk over time. In
this case, what you're worried about is what happens if it happens to
rotate in the same direction many times in a row.
If you have an angle sensor that keeps track of the turret's rotation,
then you program your robot so that if the angle sensor is reading more
than one complete rotation of the turret since starting the program, it
takes advantage of "slack time" (if you can spare it) to quickly
"de-rotate" the turret some integer number of rotations towards its
original setting (as many de-rotations as you can spare the time for).
The above approach wouldn't work too well if your rotating assembly is
something like a wheel, which spins constantly, rapidly, and
unidirectionally, since there wouldn't be much opportunity for
backtracking. But for turret-like applications, I bet this would work
fairly well.
Just a thought...
|
|
Message is in Reply To:
 | | RE: 360 deg. swiveling electrical connection
|
| (...) Give it enough wire and there's no reason you couldn't nip over those two degrees. On the theory of the more range the better, you could let the wire twist some, and get maybe +/-360 or more. Just have it re-center after every attack and you (...) (26 years ago, 3-Feb-99, to lugnet.robotics)
|
11 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
