DPRG mailing list archives: [DPRG] Does anyone have any experience with "servo savers"?

Thu Dec 13 12:33:38 CST 2007

Interesting idea!  Jon is right about the necessity of measuring
spring deflection.  One other thing to consider is the bandwidth of
these actuators.  The slew rate of the servo, combined with the speed
of the control loop that calculates your new servo position based on
the spring deflection, will limit the bandwidth. Is this something as
a fun experiment, or do you have an application in mind?

-Sanjay

> Date: Wed, 12 Dec 2007 13:29:19 -0600
> From: "Ed Paradis" <legomaniac at gmail.com>
> Subject: [DPRG] Does anyone have any experience with "servo savers"?
> To: DPRG <dprglist at dprg.org>
> Message-ID:
>         <1d18bd830712121129m5c264671i546729590002b7c4 at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Hi everyone,
>
> I was looking up servos for yet-another-robot project, and I saw some
> items called "servo savers".  They seem to be springs that dampen
> shocks to your servos.
>
> I am interested in building series elastic actuators, and if these
> things work like I think they do, they're a really simple way of
> implementing a series elastic motor.
>
> The idea in a series elastic motor (simply)  is that your motor drives
> a spring, which then drives your load, be it wheels or levers.
>
> >From what I can find online, thats exactly what these servo savers
> could do.  They seem to be more to protect your servo if a crash tries
> to back drive your servo by compressing a spring.  The energy of the
> shock goes into compressing the spring instead of cracking a gear in
> your servo.
>
> With a properly rated spring (probably one with a much lower spring
> constant than typical for these things), you could have a great series
> elastic motor that is available ''off the shelf''.
>
> Has anyone seen one of these in person?  Does this seem possible?
>
> Ed
>
>