Hi All you would still need is a way to regulate the light intensity.
Bob On Jan 9, 2010, at 7:58 PM, J. Forster wrote: > Consider this: > > A torsion pendulum with blackened vanes on the perimeter suspended from a > fiber. Part way up the fiber is an optical shutter and a small fixed > magnet so the shutter is stable in two positions with some hysteresis. One > position lets light fall on the vanes, one does not. > > This would make an optically pumped pendulum w/ no electronics. > > -John > > ============== > > > > >> Hi >> >> If you abandoned the non-elecronic side of the requirement, you could hit >> it with a pulsed LED and probably get phase data off of a couple of photo >> detectors. >> >> Crazy stuff ... >> >> Bob >> >> >> On Jan 9, 2010, at 6:21 PM, J. Forster wrote: >> >>> Maybe you could "pump" the pendulum optically, using a beam of light, >>> like >>> those glass bulb "radiometers" they sell that spin on a sunny window >>> ledge. >>> >>> -John >>> >>> ============= >>> >>> >>> >>>> Like a magnetically coupled escapement >>>> Sent via BlackBerry by AT&T >>>> >>>> -----Original Message----- >>>> From: Bob Camp <[email protected]> >>>> Date: Sat, 9 Jan 2010 12:36:11 >>>> To: Discussion of precise time and frequency >>>> measurement<[email protected]> >>>> Subject: Re: [time-nuts] Non electrical time-nuttery >>>> >>>> Hi >>>> >>>> How about a rotary pendulum on a quartz fiber spring with some kind of >>>> trick magnets to drive it / read it out? Put the pendulum and spring >>>> inside an evacuated glass envelope to get around the vacuum pump issue. >>>> The enclosure could be pretty small. >>>> >>>> Drive the magnets with a second external clock, and feedback compensate >>>> it. Let the external clock do all the readout via a very normal gear >>>> and >>>> pointers system. The trick would be getting the feedback loop to work >>>> purely mechanically with enough gain to "unload" the master pendulum. >>>> >>>> Bob >>>> >>>> >>>> On Jan 9, 2010, at 2:07 PM, Lux, Jim (337C) wrote: >>>> >>>>> OK.. So we're moving back in electrical technology.... >>>>> But what about mechanical? Could modern technology get a substantial >>>>> (>order of magnitude) improvement over 19th century chronometers >>>>> (either >>>>> pendulum or balance wheel or whatever). I know there's some really >>>>> good >>>>> quartz fiber torsional spring schemes, but I think they still need >>>>> electrical means to keep them moving and to read it out. >>>>> >>>>> So how good can one do with a mechanical, hydraulic, (or chemical, I >>>>> suppose) system? Let's assume it has to have a "direct" readout that >>>>> is >>>>> human readable by a causal bystander. (this starts to sound like the >>>>> 10,000 year clock or whatever it is..) >>>>> _______________________________________________ >>>>> time-nuts mailing list -- [email protected] >>>>> To unsubscribe, go to >>>>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>>>> and follow the instructions there. >>>>> >>>> >>>> >>>> _______________________________________________ >>>> time-nuts mailing list -- [email protected] >>>> To unsubscribe, go to >>>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>>> and follow the instructions there. >>>> >>>> _______________________________________________ >>>> time-nuts mailing list -- [email protected] >>>> To unsubscribe, go to >>>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>>> and follow the instructions there. >>>> >>>> >>> >>> >>> >>> _______________________________________________ >>> time-nuts mailing list -- [email protected] >>> To unsubscribe, go to >>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>> and follow the instructions there. >>> >> >> >> _______________________________________________ >> time-nuts mailing list -- [email protected] >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> >> > > > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
