home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!uvaarpa!darwin.sura.net!bogus.sura.net!howland.reston.ans.net!paladin.american.edu!auvm!VAXF.COLORADO.EDU!POWERS_W
- Return-Path: <@VMD.CSO.UIUC.EDU:POWERS_W%FLC@VAXF.Colorado.EDU>
- X-Envelope-To: CSG-L@vmd.cso.uiuc.edu
- X-Vms-To: @CSG
- Mime-Version: 1.0
- Content-Transfer-Encoding: 7BIT
- Message-ID: <01GTXWSROZZ60000D9@VAXF.COLORADO.EDU>
- Newsgroups: bit.listserv.csg-l
- Date: Mon, 25 Jan 1993 16:07:21 -0700
- Sender: "Control Systems Group Network (CSGnet)" <CSG-L@UIUCVMD.BITNET>
- From: "William T. Powers" <POWERS_W%FLC@VAXF.COLORADO.EDU>
- Subject: Monkeying around with heads
- Lines: 34
-
- [From Bill Powers (930125.1600)]
-
- Avery Andrews (930125.0800) --
-
- >But, for both groups, applying a constant load cause the head
- >to fail to attain its final position, (though it almost gets
- >there), attaining the final position when the load is removed.
-
- This sounds like a proportional control system. The loop gain
- could be calculated if we knew some numbers. At least we could
- get the error sensitivity: newton-meters of torque per radian of
- error. Considering that the head normally doesn't experience any
- torque disturbances beyond the small amount due to offset of the
- axis of rotation from the center of gravity, this control system
- doesn't need very much static error sensitivity. It's not really
- sufficient to say that the head "fails" to attain its final
- position. The amount by which it fails, and under what magnitude
- of disturbance, makes a lot of difference. Also, the dynamics of
- control are very important: how fast does the head accelerate and
- decelerate? I'll bet the angular accelerations are pretty large
- during a sudden movement. And how's the stability? Does the head
- oscillate back and forth around the final position a few times,
- or is it a strongly damped movement, approaching the final
- position asymptotically?
-
- The problem I always have with articles like this is that the
- authors aren't thinking in terms of control theory, so they don't
- record the kind of data you'd need to evaluate the control
- system.
-
-
- Best,
-
- Bill P.
-