Thursday, December 3, 2009

Note on Register Mechanical Design

Register controlled by Futaba S3003 servo
I've just been told by one of DZ users that his design uses about 90° of servo motion to move from totally open to totally closed position. My kneejerk reaction was -

Design your registers so 180° of servo travel is necessary to move the damper from fully closed to fully open position.
Primary reason for this is that the torque required from the servo is proportionally less, hence, less stress, longer usable life and quieter operation - servos are becoming to growl as the load goes up, and as their age goes up.

There are several things you need to keep in mind when you're designing products using servos.

Servo range of motion

All servos, being analog devices, are different. Normally, servos are supposed to work with PWM control signal being between 200μsec and 2000μsec, but there are instances that can go as low as 160μsec and as high as 2300μsec.

Servo controller signal range

Not all servo controllers are created equal, either. Some of them support advanced range limiting and calibration options, some of them are incapable of producing anything other than a predefined range of signals with no calibration options whatsoever - the only calibration option would be to design the device in such a way that the servo travel is artificially limited, and later use software controlled range limiting to make things work properly.

Hardware vs. Software

DZ3 will support both range limit and calibration within coming weeks (DZ2 supported it already, and it's just a matter of porting it over), so your hardware (both servos and controllers) is the only limitation for your design.

On the picture: existing register designed to utilize 180° of servo travel.


  1. Although I completely agree that one should try to build 180degree of travel device, sometimes it's simply not possible. The example I mentioned may not be actuall 90 degrees, but it's close and I had to do so due to physical restrains - it's a small register (2x12") And there is simply no place for the arm to swing 180 degrees without hitting something inside the dust. Probably if I put a camera in the duct and looked at things from inside I could modify it to have a bigger travel, but it works as it is rignt now.

  2. Would it be possible to post some more detail on the construction of the servo mounting bracket in the picture? Hard to tell from the angle in the picture exactly what the bracket looks like. More pictures would be good, a dimensioned drawing would be even better!

  3. Will do later today or over weekend.

    For now, it may not be the worst idea to go get familiar with SketchUp, for that's what I'm using for drawings - I'll create one and post the link to it in a separate message here.