View Single Post
  #20   Report Post  
Posted to rec.crafts.metalworking
Tim Wescott Tim Wescott is offline
external usenet poster
 
Posts: 1,620
Default Rotation detection?

On Tue, 07 Jul 2009 13:32:50 -0700, Bill Noble wrote:

"Don Foreman" wrote in message
...
On Mon, 6 Jul 2009 03:05:21 -0400, "Buerste" wrote:

Imagine a simple fractional hp DC gearmotor with a torque limiter or
torque
limiting clutch that powers a small machine at 60 or so RPM. Lets say
the
machine jams and the torque limiting device kicks out. How can I
cheaply, easily sense this condition and turn off the motor?

This is a personal project and when I get it running it will amaze and
astound!

Several have suggested sensing drive and driven parts but left the
signal processing details as an exercise.

I'm in dialup-land just now so it'd be tedious to download a datasheet,
but I'm pretty sure the CD4046 CMOS phaselocked loop would work well
here. It would produce an output signal if the frequency of (digital)
inputs differ, as they would here. I'd suggest something like Allegro
hall-effect geartooth sensors with integral magnets because they're so
easy to apply. They could easily sense screwheads on the drive and
driven parts. The CD4046 costs less than a buck and will accept
supply voltages from 3 to 18 volts DC.

You'll find a datasheet at www.ti.com Search for CD4046B.

Ping me if you're interested in this approach. I'll be back in town
tomorrow.


if you are going to build your own timer, I would strongly recommend the
555 - set the timer to 2 seconds, hit the reset with the reed switch I
recommended, and wire the output to a 2n2222 in open collector mode to
drive a relay (or if a small relay, drive directly, just don't forget
the suppression diode) - minimal parts count, very cheap, easy to hook
up - just follow the schematic in the application handbook


Had I bothered jumping in on this thread earlier this is pretty much what
I would have recommended. National called it a "pulse omission detector"
in their data sheets, although I don't know if Fairchild will have as
rich a selection of circuits as the 1980 National Linear Databook.

--
www.wescottdesign.com