Newsgroups: comp.robotics
Path: brunix!cat.cis.Brown.EDU!agate!howland.reston.ans.net!europa.eng.gtefsd.com!newsxfer.itd.umich.edu!gumby!wupost!csus.edu!netcom.com!nagle
From: nagle@netcom.com (John Nagle)
Subject: Re: controling servo motors
Message-ID: <nagleCqJ1z9.n0B@netcom.com>
Keywords: servo 6811
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <2s3ukv$h6e@lastactionhero.rs.itd.umich.edu>
Date: Sat, 28 May 1994 19:25:56 GMT
Lines: 29

ocg@cw-u02.umd.umich.edu (John Roldan) writes:
>Hi all,
>I bought a $20 servo motor at my local hobby store and am trying to 
>control it with a 6811.  I am sending a 50% duty cycle square wave down 
>the yellow wire and am supplying 5v on red and blac.  At various 
>frequencies all under 30 hz   all the motor does is move to the extreme 
>right and jitter or just jitter.   So my questions are:

     Your duty cycle is way too big.  Try 0.5ms to 1.5ms on, 13ms off.
The length of the ON pulse controls the servo, and really matters.
The OFF time isn't that critical.

>Also, is it possible to power these things with higher than 5v?  The 
>thing is resetting my cpu sometimes from the glitches it causes, and I 
>already have a noise isolated 9v area on my daughterboard.  Id hate to 
>have to ad another 5v regulator on the 9v side of my board.

      5.8V is the standard battery pack for these things, so 6V should be OK.
But you need more filtering.  Try a 6V regulator, just for the servo(s),
and then put a 500MFD 12V cap across the servo (red and black)
There's a big current draw from the servo just as you command it a new 
position and the motor starts up. The capacitor will provide some energy
storage to handle that startup surge.  If, after that, you still have
noise problems, put some smaller caps upstream of the regulator,
downstream of the regulator, and across the CPU's power.  The 68HC11
is good about tolerating noise, and coexists well with R/C servos, 
provided you buffer the servo's big startup current needs.

						John Nagle
