what frequency to drive servos at?

So I had always understood that RC servos were supposed to be driven at 50hz. This is what all my servo controls drive them at. I've always thought
this seemed a bit slow - but so far it's never been a problem. But so today I just got a bunch of HS-81MGs that are for a project I'm working on - and they had black red and yellow wires. Just to be sure that yellow was signal (I'm used to black red and white) I googled it, and stumbled on this page: http://wolfstone.halloweenhost.com/TechBase/svoint_RCServos.html So yellow is signal as I guessed - but that's not the big discovery. This is:
"The rate at which these pulses are sent isn't terribly important - only the width of the pulse. Some typical rates are 400 Hz (2.5 mSec pulse spacing) and 50 Hz (20 mSec pulse spacing)."
So they are saying that servos can be driven 8 times as fast as the standard 50hz. This is an awesome discovery for me as I always like to speed up things... So - my question is this: Is this information correct? Can all servos be driven at whatever speed you feel like as long as you can get the proper pulse width? So I think 400hz is about the maximum speed there as I've had some servos that took a maximum pulse width of about 2.35 ms.
Thanks!
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Good Morning!

Normally, servos are driven at about 50 Hz with 1ms pulse left margin and 2ms right margin (abd of course 1,5ms middle position).
So called "Digital servos" can be driven at up to 200Hz. Maybe some servos operate at 400 Hz, too. The problem: Servomotors ar only getting power at pulse, so if You doube the pulse-Rate, You also doubling the Power into the motors, and the Heat generated in it. Digiservos are constructed to dissipate heat from pulses up to 200 Hz, standard servos may burn at even 100Hz. Try it, monitor the Motortemp at full load, and find the point where the Servos are just NOT getting tooo hot....
Voila' ;-)
Michael Buchholz.
P.S.: Sorry for my bad english, I'm german....
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Michael Noone wrote:

It's *generally* correct, but it depends on the servo. If you do a Google search for the newsgroup archives you'll find a discussion of this a few months back. The upshot is that though 400 Hz may be a theoretical high end for a servo frame rate, and the type of error correction scheme used in the servo electronics may lower that number, even substantially. A popular type of GWS servo was found to have an upper bound of about 125 Hz, for example. Higher rates caused the servo to lock up.
The reason to increase frame rate is for increased torque, but again there is an upper bound at which the motor will not deliver more torque, even at higher frame rates. Each servo needs to be individually tested.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.