Philip, I think this sounds pretty reasonable, here's my 2-cents worth.
I think calibration should be part of delivering the clock. You can imagine a designer developing elaborate models for the timebase and you don't want to stifle creativity here. I think calibration is part of the delivered clock and not done by the judges. The clock should be settable by a layperson, but I think you should let people go wild calibrating the clock. The two big enemies the designer fights will likely be environmental and aging. I think you've covered a good environmental specification, but aging should be incorporated into the contest. For example, a temp compensated xtal oscillator my be calibrated to well less than 0.1 ppm and look really fantastic for a few days, but the crystal might age 2-3 ppm over the first year of operation. I imagine running the contest over a few months. At some submission date the clocks are collected at your office, set to the current time, and left to run for a few months (or however much time you have). At the end of 3-months all of the clocks are measured and a winner is picked. Maybe you can publish intermediate results for dramatic effect. Tom Van Baak has a nice story on his website about how he got interested in precision timing, http://www.leapsecond.com/. He wanted a clock that would be accurate to better than 1-second over a year so that he could appreciate and adjust the clock during a leapsecond. I don't know how, but I think that it might be possible to build a clock for this contest that meets that criteria. If so, I think the entries need a feature for measuring accuracy more precisely. I suggest requiring that the entires have a BNC connector that outputs a TTL level pulse once per second, the rising edge of the signal marks the beginning of a second (this is usually just called a PPS signal). The accuracy of the clock can be done visually to 1s resolution, finer measurement can be done with comparison against PPS from a GPS receiver. jeff [EMAIL PROTECTED] wrote: > A number of people have asked for more details on IEEE Spectrum's digital > clock competition, so we've formulated the following list. Throughout, the > idea is to build a clock that an ordinary person would want to use, in an > ordinary home. That's why we want a display that can be read with ease > from across a room. > > > Operating environment and other specs for IEEE Spectrum's Digital > Clock Competition: > > --between 10 and 50 degrees C > > --between 0 and 100 percent relative humidity > --with seven-segment LED display, no smaller than 0.56 inches > --no limit on power > --calibration should be within the grasp of a layman > --lacking an oscilloscope here in the office, we will check > accuracy against a WWVB or GPS signal (other suggestions--even volunteers > to help in the judging--are welcome) > --parts to be available from any of the big distributors > (RadioShack, Mouser, DigiKey, Maplin, etc.) or, in sufficient quantities > (100s, say) from a surplus store > > > > > Philip E. Ross > Senior editor > IEEE Spectrum Magazine > 212 419 7562 > http://www.spectrum.ieee.org > _______________________________________________ > 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.
