Looking for a BASIC Stamp expert with timekeeping experience in New York City
dtalvacchio
Posts: 36
Hello everybody--
For some time now I have been working on an art piece that uses a BASIC Stamp to operate a digital clock with a twist. The twist is that each tick of the seconds digit does not occur at one-second intervals. Instead, the ticks are programmed to occur randomly at intervals of roughly 0.7, 0.8, 1.2, and 1.4 seconds. Because these options are symmetric about 1 second, the clock wants to hover around the "actual" time . . . it can be said to move in and out of ordinary time, while still functioning as a more-or-less practical clock.
I have built the clock and done all the programming, and it works fairly well. The problem is that it is using the stamp's built-in timekeeping function, which is not quite accurate enough for the long-term precision required here. The clock has been running a little fast, not because of the randomizing component, but simply because everything is a little fast (I think). I did some research and came across the attached article, which solves just this problem by using a crystal for the timekeeping. I have built the circuit but not done the programming for the crystal, because I must have become convinced that it wasn't going to work in my case, with the randomizing bit. As I built the clock a couple of years ago, I am a little out of touch with all of the programming and circuitry knowledge. The issue now is that artwork was just accepted into a Biennial which begins in mid-June. I desperately need an expert who happens to live in or near New York City (my studio is in Queens) who can assess the situation, based on what I already have, and help to make the clock as accurate as possible by the end of May. I would be able to pay the consultant for his or her time. I'm assuming this would be fairly easy to solve for a BASIC stamp expert with any experience in crystal-based timekeeping such as shown in the attached article.
Please let me know right away if this is something that you might be able to help with . . . and/or please let any New York City area folks know about this consultant opportunity if you happen to know anybody. Thank you much.
Best--
Dominick T.
For some time now I have been working on an art piece that uses a BASIC Stamp to operate a digital clock with a twist. The twist is that each tick of the seconds digit does not occur at one-second intervals. Instead, the ticks are programmed to occur randomly at intervals of roughly 0.7, 0.8, 1.2, and 1.4 seconds. Because these options are symmetric about 1 second, the clock wants to hover around the "actual" time . . . it can be said to move in and out of ordinary time, while still functioning as a more-or-less practical clock.
I have built the clock and done all the programming, and it works fairly well. The problem is that it is using the stamp's built-in timekeeping function, which is not quite accurate enough for the long-term precision required here. The clock has been running a little fast, not because of the randomizing component, but simply because everything is a little fast (I think). I did some research and came across the attached article, which solves just this problem by using a crystal for the timekeeping. I have built the circuit but not done the programming for the crystal, because I must have become convinced that it wasn't going to work in my case, with the randomizing bit. As I built the clock a couple of years ago, I am a little out of touch with all of the programming and circuitry knowledge. The issue now is that artwork was just accepted into a Biennial which begins in mid-June. I desperately need an expert who happens to live in or near New York City (my studio is in Queens) who can assess the situation, based on what I already have, and help to make the clock as accurate as possible by the end of May. I would be able to pay the consultant for his or her time. I'm assuming this would be fairly easy to solve for a BASIC stamp expert with any experience in crystal-based timekeeping such as shown in the attached article.
Please let me know right away if this is something that you might be able to help with . . . and/or please let any New York City area folks know about this consultant opportunity if you happen to know anybody. Thank you much.
Best--
Dominick T.
Comments
Do remember that the random function in Stamp Basic uses a 16 bit counter for its "seed" value. If you generate a new "random" number every 1/8 second, the sequence of numbers will repeat every 65536 / 8 seconds = 8192 seconds = roughly every 2 1/4 hours. Not many people would notice, but you'd need a different source of random numbers or you'd have to stick with the same random value for several clock "ticks" or even several seconds so the pattern doesn't repeat for perhaps a day. Note that it wouldn't repeat at the same time each day (because it would be every 65536 "ticks"), so people probably wouldn't notice.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Stephen
Use a random number range to get the accurate time from a time keeping chip. This way every time the random number is within a "range" it would correct the time to be accurate.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Dave W.