robf@mcs213j.cs.umr.edu (Rob Fugina) (12/13/90)
My roommate just asked me to design a clock for him. It's for astronomical purposes. Relatively simple, really, but I need a little advice and some chip numbers. The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. I figured it would be easy enough to design an oscillator with a 555 to 100Hz, calibrated with a 15-turn pot and an oscilloscope (I'll take it to campus for one of my EE labs). From there, I can use 7400-series chips to do the logic. BCD counters for both digits of the hour, and for the ones digits of the minutes and seconds. What do they make as far as divide-by-6 counters? I would need these for the tens digits of the minutes and seconds. The outputs of the counters would go to 7-segment decoders. Next is how to set it. I figured I could, for each digit, put a momentary switch directly to the digit's clock line, which would advance that digit. Special attention would have to be taken to make sure they flip to 0 correctly. A spst switch could enable/disable these switches. B Power supply...how easy is it to make it AC line powered, with a battery backup? This I'm not so worried about, but I'd like advice on whether it would be better powered by AC line or by battery...i.e., how long do you think the battery would last? The whole thing would be made up of a 555, 6 counters, 6 BCD-to-7 seg decoders, and 6 LED digits. Oh, and I nearly forgot...the most complicated part will be using logic gates to reset 0 at the right time...I'm looking at two 24-input NAND gates or something similar. Advice, please! Rob Fugina robf@cs.umr.edu
gsteckel@vergil.East.Sun.COM (Geoff Steckel - Sun BOS Hardware) (12/13/90)
In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >My roommate just asked me to design a clock for him. It's for astronomical >purposes. Relatively simple, really, but I need a little advice and some >chip numbers. > >The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. >I figured it would be easy enough to design an oscillator with a 555 to >100Hz, calibrated with a 15-turn pot and an oscilloscope (I'll take it I recommend: use a 32768 Hz crystal and a National Semi crystal-to-1-Hz chip, then run it through 4000 series CMOS for the dividers. Result: a LOT more accurate than a 555. I think you would find the 555 completely unusable for timekeeping, unless you wanted to reset it twice daily. I don't have my parts manuals here or I'd recommend chips. You can get programmable divide-by-n chips and n-to-7 segment decoders in CMOS. You can then run the whole mess off 3 alkaline batteries. Be sure to connect EVERY LAST input to +V or ground! CMOS behaves VERY unpredictably with floating inputs. If you use a LCD display, it will run all the time with less than 1 mA drain. They are available from Digi-Key, among others. Just a quick sketch. Hope this helps. geoff steckel (gwes@wjh12.harvard.EDU) (...!husc6!wjh12!omnivore!gws) Disclaimer: I am not affiliated with Sun Microsystems, despite the From: line. This posting is entirely the author's responsibility.
chik@eecg.toronto.edu (Raymond Chik) (12/13/90)
In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >My roommate just asked me to design a clock for him. It's for astronomical >purposes. Relatively simple, really, but I need a little advice and some >chip numbers. > >The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. >I figured it would be easy enough to design an oscillator with a 555 to >100Hz, calibrated with a 15-turn pot and an oscilloscope (I'll take it >to campus for one of my EE labs). .......... In terms of precision of oscillation frequency, crystal oscillator is usually used in (real time) clocks.... I really doubt that if the 555 will generate an accurate enough clock for serving your purpose...... (I think you need a very accurate clock, is that right?) >to do the logic. BCD counters for both digits of the hour, and for the >ones digits of the minutes and seconds. What do they make as far as >divide-by-6 counters? I would need these for the tens digits of the minutes >and seconds. The outputs of the counters would go to 7-segment decoders. > >Power supply...how easy is it to make it AC line powered, with a battery >backup? This I'm not so worried about, but I'd like advice on whether >it would be better powered by AC line or by battery...i.e., how long do >you think the battery would last? > hm.... if you want to use AC, you will need an AC to DC converter. You could buy one or build one yourself which could be another fun part of your project..... But using battery is much more the direct and easy way. To estimate how long can a battery last, check out how much current (average) of all the components you will use divide that into the amount of charge the battery is containing ( specified in mAh, milli amp hour , for a AA size rechargeable NiCd, it;s about 500mAh.) >Oh, and I nearly forgot...the most complicated part will be using >logic gates to reset 0 at the right time...I'm looking at two 24-input >NAND gates or something similar. Advice, please! 24-input nand !? Sound scary! Does it really exist in the market? (I don't know, just that it's rediculus to do logic with a 24-input nand if your talking about CMOS on IC which suits this application because you need low power.) ************************************************************************** * May the force be with you!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!* * Raymond Y. V. Chik |_ \ _| * * VLSI Research Grp. || -------- || * * Dept. of Elec. Eng. _||o-+ | -+-+- +-o||_ * * U. of Toronto | | | +-+ | | * * |_ | | _|_ | _| * *Internet: chik@eecg.toronto.edu ||--+ | | | | +--|| * * chik@vrg.toronto.edu _|| / | | | ||_ * * 8-) >-( |-< %-> | | | * **************************************************************************
commgrp@silver.ucs.indiana.edu (BACS Data Communications Group) (12/13/90)
robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >My roommate just asked me to design a clock for him. It's for >astronomical purposes... >The clock has to count from 00:00:00 to 23:56:03, then reset >back to zero. I figured it would be easy enough to design an >oscillator with a 555 to 100Hz, calibrated with a 15-turn pot and >an oscilloscope ... >Power supply...how easy is it to make it AC line powered, with a >battery backup? This I'm not so worried about, but I'd like >advice on whether it would be better powered by AC line or by >battery...i.e., how long do you think the battery would last? >... There are plans for a digital sidereal clock in _Sky & Telescope_ magazine, July 1976, p. 59ff. The clock itself is a MM5314 clock chip (all divide-by-six and setting functions included; requires external driver transistors for LED readouts). The "sidereal rate generator" is a TTL circuit which counts 1461 pulses from the 60Hz power line, then inserts four additional pulses between the next 60-Hz pulses going to the MM5314's input. 4/1461 closely approximates the factor required to make the sidereal clock gain ~ 3m 56s per day (one day/year). Alternaively, one could use an oscillator with a special crystal which runs slightly faster than the rate for a normal clock; sidereal wristwatches have been advertized in _S&T_. A clock which counts power-line cycles may err by several seconds at any time, but the long-term stability is excellent. (A 555 oscillator is _not_ stable enough for a clock. National Semiconductor makes an IC which generates 60 Hz from a 3.579545 MHz color-tv crystal.) Better designs have been published subsequently, however, it's easier anymore to do sidereal time by running a program in your PC/Mac/whatever, which reads the internal clock and does a calculation. The _S&T_ article also tells how to set a sidereal clock; local sidereal time depends upon your longitude. The Earth rotates ~366 times per year, but since it goes around the sun once, one of those rotations is "unwound," resulting in only 365.xxxx solar days/year. The constellations appear to make one revolution per year. We base our normal affairs upon the solar day; the sun is at the same meridian at the same time every day. A sidereal ("star time") clock is synchronized to the actual rotation of the earth; it gains one solar day per year over ordinary time. Astronomers use sidereal time because the same stars are at the same places every day at the same sidereal time, thus simplifying calculations. On sidereal time, the constellations appear fixed while the sun goes around once per year. Running our mundane affairs on sideral time would be inconvenient because the sun would be out at "midnight" part of the year. -- Frank Reid reid@ucs.indiana.edu
ken@swbatl.sbc.com (Ken Gianino 5-9081) (12/13/90)
In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >My roommate just asked me to design a clock for him. It's for astronomical >purposes. Relatively simple, really, but I need a little advice and some >chip numbers. > >The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. >I figured it would be easy enough to design an oscillator with a 555 to >100Hz, calibrated with a 15-turn pot and an oscilloscope (I'll take it >to campus for one of my EE labs). From there, I can use 7400-series chips >to do the logic. BCD counters for both digits of the hour, and for the >ones digits of the minutes and seconds. What do they make as far as >divide-by-6 counters? I would need these for the tens digits of the minutes >and seconds. The outputs of the counters would go to 7-segment decoders. > >Next is how to set it. I figured I could, for each digit, put a momentary >switch directly to the digit's clock line, which would advance that digit. >Special attention would have to be taken to make sure they flip to 0 >correctly. A spst switch could enable/disable these switches. >B >Power supply...how easy is it to make it AC line powered, with a battery >backup? This I'm not so worried about, but I'd like advice on whether >it would be better powered by AC line or by battery...i.e., how long do >you think the battery would last? > >The whole thing would be made up of a 555, 6 counters, 6 BCD-to-7 seg >decoders, and 6 LED digits. > >Oh, and I nearly forgot...the most complicated part will be using >logic gates to reset 0 at the right time...I'm looking at two 24-input >NAND gates or something similar. Advice, please! > >Rob Fugina >robf@cs.umr.edu Why 23:56:03? Doesn't sidereal time use 24 hours and just lose 4 min/day? Boy, I've kinda forgotten. At any rate, Astronomy magazine had an article on how to convert a Radio Shack 24 Hr alarm clock (digital) to sidereal time. It just added a couple run-of-the-mill counter chips to change the 60 Hz count input the right amount. I don't have the date but try 1988 +1/-2 years. Ken
ressler@galileo.ifa.hawaii.edu (Mike "IR" Ressler) (12/13/90)
In article <1990Dec12.214854.19076@swbatl.sbc.com> ken@swbatl.sbc.com (Ken Gianino 5-9081) writes: >In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >>My roommate just asked me to design a clock for him. It's for astronomical >>purposes. Relatively simple, really, but I need a little advice and some >>chip numbers. >> >>The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. > > Why 23:56:03? Doesn't sidereal time use 24 hours and just lose 4 min/day? I'm glad someone caught this. Sidereal time is indeed at 24 thing, it's just that the hours are slightly shorter - 1.002737909 times shorter (23h 56m 4.09053s to the sidereal day). You just need to design a normal 24 hour clock that uses a clock frequency shifted up by this amount. Very vague recollections say that someone actually made 1.002738 MHz crystals, but I have no idea who. Good luck. -- Mike Ressler - Infrared Photon Jockey ressler@galileo.ifa.hawaii.edu If at first you don't succeed, get a bigger sledgehammer.
henry@zoo.toronto.edu (Henry Spencer) (12/13/90)
In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >My roommate just asked me to design a clock for him. It's for astronomical >purposes... If he'll settle for a clock that keeps true sidereal time :-), the March 1988 issue of Astronomy had an article (note, some corrections appeared on page 37 of the June 1988 issue) on converting a Radio Shuck digital alarm clock to keep sidereal time. -- "The average pointer, statistically, |Henry Spencer at U of Toronto Zoology points somewhere in X." -Hugh Redelmeier| henry@zoo.toronto.edu utzoo!henry
mcovingt@athena.cs.uga.edu ( Michael A. Covington) (12/13/90)
In article <10633@uhccux.uhcc.Hawaii.Edu> ressler@galileo.ifa.hawaii.edu (Mike "IR" Ressler) writes: >In article <1990Dec12.214854.19076@swbatl.sbc.com> ken@swbatl.sbc.com (Ken Gianino 5-9081) writes: >>In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: >>>The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. >> >> Why 23:56:03? Doesn't sidereal time use 24 hours and just lose 4 min/day? > >I'm glad someone caught this. Sidereal time is indeed at 24 thing, it's just >that the hours are slightly shorter - 1.002737909 times shorter (23h 56m >4.09053s to the sidereal day). You just need to design a normal 24 hour clock >that uses a clock frequency shifted up by this amount. Very vague >recollections say that someone actually made 1.002738 MHz crystals, but I have >no idea who. Good luck. You are right. Sidereal time (defined as the right ascension of the meridian) does go from 00:00 to 24:00 just like ordinary time; the units (day, hour, minute, second, etc.) are a wee bit shorter. There are many companies, such as JAN Crystals, that will custom-grind a crystal to *any* frequency for under $10. Look in the back of any electronics magazine.
yh@cs.Princeton.EDU (Yoichi Hamazaki) (12/14/90)
One of simplest(and may be cheapest) way to get clock working between 0:00:00 to 23:56:04 is modifing commercial digital alarm clock. 1) Buy digital alarm clock of your faverite. (AC powered or Battey, LED,FLD or LCD) 2) make two mono-stable multivibrator(One-shot) using CMOS or TTL. Pulse width of them are 4seconds and 0.09 seconds. 3) Find alarm signal from clock LSI and connect to 4sec. One-shot trigger. 4) Connect output of 4sec. one-shot to trigger of 0.09sec one-shot. 5) Connect output of 0.09sec one-shot to 'set to 0:00:00' switch input. (Usually, both of time-adjust switches) 6) Turn on clock, adjust time, and set alarm to '23:56' When clock becomes 23:56, 4sec one-shot makes pulse, 4 seconds later 'set to 0:00:00' signal(s) is fed to clock for 0.09 seconds. Then clock count time from 0:00:00 again. -- Yoichi HAMAZAKI Dept. computer science, Princeton Univ. (from Electrotechnical Lab. (ETL), Japan) E-mail: yh@princeton.EDU Ham: W2/JR1OQY (2m & 70cm)
dale@lamont.ldgo.columbia.edu (dale chayes) (12/14/90)
In article <1833@umriscc.isc.umr.edu>, robf@mcs213j.cs.umr.edu (Rob Fugina) writes: > My roommate just asked me to design a clock for him. It's for astronomical > purposes. Relatively simple, really, but I need a little advice and some > chip numbers. > > The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. > I figured it would be easy enough to design an oscillator with a 555 to A lot of discussion about counting and setting an stuff omitted ..... > > The whole thing would be made up of a 555, 6 counters, 6 BCD-to-7 seg > decoders, and 6 LED digits. > > Oh, and I nearly forgot...the most complicated part will be using > logic gates to reset 0 at the right time...I'm looking at two 24-input > NAND gates or something similar. Advice, please! For my time and effort, I would chop all the discrete parts and do it in a single chip microcontroller. There are lots that would work just fine. Use one you are familiar with (or pick one and learn.) Use the crystal as the time base and count either with one of the built in timer/counters, or write a software timing loop. (I'd go for the counter/timer to make live simple. Let it overflow once per second or 10 times per second. Keep track of time in software (then the end-of-sideral-day rollover is trivial. (If you do it in hardware, you into another bunch of '688s or other comparator.) You should be able to mux the 7-seg displays right off a pair of eight bit ports to get plenty of digits directly. If I was going to do it, I'd use an 8751 because I'm familiar with it but I might choose one of the 68HCxxxxx processors if I was starting from scratch. In any case, I would provide a battery backup if you friend is going to depend on the clock. Dale ============ -- Dale Chayes Lamont-Doherty Geological Observatory of Columbia University Route 9W, Palisades, N.Y. 10964 dale@lamont.ldgo.columbia.edu voice: (914) 359-2900 extension 434 fax: (914) 359-6817
cjp@megatek.UUCP (Christopher J. Pikus) (12/14/90)
From article <1833@umriscc.isc.umr.edu>, by robf@mcs213j.cs.umr.edu (Rob Fugina): > > Power supply...how easy is it to make it AC line powered, with a battery > backup? This I'm not so worried about, but I'd like advice on whether > it would be better powered by AC line or by battery...i.e., how long do > you think the battery would last? > > Rob Fugina > robf@cs.umr.edu It would be preferable for you to run this thing off AC if possible. It is far more accurate than a crystal. A typical crystal will be accurate to about 0.1% (1000 ppm) and will drift with temperature. The 60Hz that is derived from a wall socket is tied into a national power grid and is kept pretty close in sync. There is a master clock for the power grid and if it starts drifting away from the 60Hz, then the AC freq is sped up/slowed down to bring it back in line. short of an atomic clock the 60Hz AC is the most accurate you can get. (Of course you can spring for one of those Ovenized Temperature Controlled Crystal Oscillators (OTCXO) with accuracy to 10E-7 but they require 50V to keep them warm and cost several hundred dollars, the voltage would pretty much require you to use AC anyway). A sample circuit for the AC is as such: Diode |\ 74LS14 ? (schmidt trigger buffers) |\ | | \ +-----| >|-------| >o------------ 60 Hz square wave | |/ | | / | |/ | _________ 5V regulator 110VAC | | | 6.3VDC ____ ________ ________+_|~ + |------------+------| |------+---------- +5V }||{ | | | in| |out | AC line }||{ | bridge | |+ |____| |+ }||{ | rect. | Big --- |gnd --- }||{ | | Cap --- | --- }||{ | | | | | another Big Cap }||{ | | | | | and little caps ________}||{__________|~ - |------------+--------+---------+ ^ |_________| | | _|_ GND | = | . 6.3V transformer -- That is all ... Christopher J. Pikus, Megatek Corp. INTERNET: cjp@megatek.uucp San Diego, CA UUCP: uunet!megatek!cjp or ucsd!megatek!cjp
jvb7u@astsun.astro.Virginia.EDU (Jon Brinkmann) (12/15/90)
In article <3023@lamont.ldgo.columbia.edu> dale@lamont.ldgo.columbia.edu (dale chayes) writes: #In article <1833@umriscc.isc.umr.edu>, robf@mcs213j.cs.umr.edu (Rob Fugina) writes: #> My roommate just asked me to design a clock for him. It's for astronomical #> purposes. Relatively simple, really, but I need a little advice and some #> chip numbers. Why re-invent the wheel? There was an article a few years ago in either Astronomy or Sky and Telescope showing how to simply (a couple of parts) modify a $20 Radio Shack digital clock to make a sidereal clock. Sorry, but I don't remember what issue it was! Jon -- Jon Brinkmann Astronomy Department Internet: jvb7u@Virginia.EDU University of Virginia UUCP: ...!uunet!virginia!jvb7u P.O. Box 3818 SPAN/HEPnet: 6654::jvb7u Charlottesville, VA 22903-0818
robf@mcs213f.cs.umr.edu (Rob Fugina) (12/16/90)
In article <10633@uhccux.uhcc.Hawaii.Edu> ressler@galileo.ifa.hawaii.edu (Mike "IR" Ressler) writes: >>>The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. >> Why 23:56:03? Doesn't sidereal time use 24 hours and just lose 4 min/day? >I'm glad someone caught this. Sidereal time is indeed at 24 thing, it's just >that the hours are slightly shorter - 1.002737909 times shorter (23h 56m >4.09053s to the sidereal day). You just need to design a normal 24 hour clock >that uses a clock frequency shifted up by this amount. Very vague >recollections say that someone actually made 1.002738 MHz crystals, but I have >no idea who. Good luck. > Mike Ressler - Infrared Photon Jockey ressler@galileo.ifa.hawaii.edu Thank you for this information! I was only going on what my roommate said, and it's he who wants the clock. It's not often that the correct way is the easier way... So, if I found a 1.002738MHz xtal, I would divide it by 1000000 to have 1Hz...so what do I use to divide by a million? Six divide-by-tens? Rob
mcovingt@athena.cs.uga.edu ( Michael A. Covington) (12/17/90)
Simple solution: Get yourself an automotive clock that already has a crystal in it. (You'll need one that uses a chip that has a 24-hour mode; might be tricky). Or get any set of chips that makes a 60-Hz line-operated clock. Use an MM5369N, which is a special chip that takes a 3.58-MHz color TV crystal and divides it down to 60 Hz. Get a crystal that's a wee bit faster, and you're done. JAN Crystals advertises in all the electronics magazines and a custom crystal costs less than $10.
vail@tegra.COM (Johnathan Vail) (12/18/90)
In article <1833@umriscc.isc.umr.edu> robf@mcs213j.cs.umr.edu (Rob Fugina) writes: My roommate just asked me to design a clock for him. It's for astronomical purposes. Relatively simple, really, but I need a little advice and some chip numbers. The clock has to count from 00:00:00 to 23:56:03, then reset back to zero. I figured it would be easy enough to design an oscillator with a 555 to 100Hz, calibrated with a 15-turn pot and an oscilloscope (I'll take it Any RC circuit like a 555 will not give very good accuracy as a time base. Make it crystal controlled or timed off the line frequency. to campus for one of my EE labs). From there, I can use 7400-series chips to do the logic. BCD counters for both digits of the hour, and for the ones digits of the minutes and seconds. What do they make as far as divide-by-6 counters? I would need these for the tens digits of the minutes and seconds. The outputs of the counters would go to 7-segment decoders. If I was doing it I would use a crystal (with cap tweaker if accuracy is desired), 1 single chip micro, LED or LCD smart display chip. You could power it from the AC line and derive a very accurate (in the long run) 60 HZ timebase to control it. Next is how to set it. I figured I could, for each digit, put a momentary switch directly to the digit's clock line, which would advance that digit. Special attention would have to be taken to make sure they flip to 0 correctly. A spst switch could enable/disable these switches. With a micro you can switch between read and sidereal time, and easily add other features. Setting is as easy as a digital watch. Well, thats what I would do. I prefer for projects like this to make the hadware as simple as possible. At least use something other than a 555 to provide the timebase. jv "Fickt nicht mit der Raketemensch!" _____ | | Johnathan Vail | n1dxg@tegra.com |Tegra| (508) 663-7435 | N1DXG@448.625-(WorldNet) ----- jv@n1dxg.ampr.org {...sun!sunne ..uunet}!tegra!vail
f@Alliant.COM (Bill Freeman) (12/18/90)
I wonder how hard it would be to un-seal and polish the resonator in one of
my spare cheap LCD watches to move it up 0.274% ? I presume that virtually
all watches use 32768 Hz ceramic resonators (except for those real garbage
R-C ones that were around something like 5 years ago). Are there any watch
resonator experts out there who could pontificate for me?
-Bill
--
--
...!{decvax!linus,mit-eddie}!alliant!f Bill Freeman KE1G
alliant!f@eddie.mit.edu PP-SMELheskett@titan.tsd.arlut.utexas.edu (Donald Heskett) (12/19/90)
I'm not a "watch resonator expert" but I believe that polishing the resonator to alter the frequency of the timepiece is not viable. Unless I'm mistaken, the resonators are quartz crystals with electrodes plated onto parallel faces. The biggest problem, I'm guessing, would be restoring the electrode on the crystal face you ground down. Why make life difficult? Custom crystals are relatively cheap as are the components to divide the frequency down to near 60hz.
squishy@casbah.acns.nwu.edu (Shishin Yamada) (12/29/90)
"missing beats," that is, every nth oscillation, skip the clock's
increment. On a 24h:00m:00s normal day, there are 86,400 seconds. If the
clock counted by using the 60Hz line, this makes 5,184,000 beats. On a
23h:56m:00s sideral day (24 normal hours to us), there are 86,160 seconds.
This makes for 5,169,600 beats on a 60Hz line.
Therefore, if you skipped a beat every 14,400 beats of a 60Hz line,
your clock would then count sideral time. In this manner, you could have a
counter that went up to 14,400 beats, and then have it blank the next
(14,401th beat) of the 60Hz line. This correction would occur every 240
seconds (4 minutes). This introduces a very small error into your clock,
since the 240th second will be 59/60th of a normal second, but not much as
it corrects itself every four minutes. Pretty accurate, I think.
It's a weird idea, but then again, I just thought of it! Hope you like it!
If you'd like more of my weird ideas, please e-mail me on BitNet. I
am fastest to reply on that medium!
======================================================================
Shishin "SQUISH" Yamada | BitNet: squishy@casbah.acns.nwu.edu
Senior Undergraduate, EE | Compuserve: 76636,3254
Northwestern University | Prodigy: KTVB99A
=======================================================================
Newsgroups: sci.electronics
Subject: Re: Sidereal Time Clock
Summary:
Expires:
References: <1833@umriscc.isc.umr.edu> <3023@lamont.ldgo.columbia.edu> <4391@alliant.Alliant.COM>
Sender:
Followup-To:
Distribution: sci.electronics
Organization: Northwestern University
Keywords: squishy@casbah.acns.nwu.edu (Shishin Yamada) (12/30/90)
On a side-note, if you might try modifying the clock frequency by
Message-ID: <2369@casbah.acns.nwu.edu>
Date: 29 Dec 90 12:06:30 GMT
References: <1833@umriscc.isc.umr.edu> <3023@lamont.ldgo.columbia.edu> <4391@alliant.Alliant.COM>
Followup-To: 4391@alliant.Alliant.COM
Distribution: sci.electronics
Organization: Northwestern University
Lines: 33
"missing beats," that is, every nth oscillation, skip the clock's
increment. On a 24h:00m:00s normal day, there are 86,400 seconds. If the
clock counted by using the 60Hz line, this makes 5,184,000 beats. On a
23h:56m:00s sideral day (24 normal hours to us), there are 86,160 seconds.
This makes for 5,169,600 beats on a 60Hz line.
Therefore, if you skipped a beat every 14,400 beats of a 60Hz line,
your clock would then count sideral time. In this manner, you could have a
counter that went up to 14,400 beats, and then have it blank the next
(14,401th beat) of the 60Hz line. This correction would occur every 240
seconds (4 minutes). This introduces a very small error into your clock,
since the 240th second will be 59/60th of a normal second, but not much as
it corrects itself every four minutes. Pretty accurate, I think.
It's a weird idea, but then again, I just thought of it! Hope you like it!
If you'd like more of my weird ideas, please e-mail me on BitNet. I
am fastest to reply on that medium!
======================================================================
Shishin "SQUISH" Yamada | BitNet: squishy@casbah.acns.nwu.edu
Senior Undergraduate, EE | Compuserve: 76636,3254
Northwestern University | Prodigy: KTVB99A
=======================================================================
Newsgroups: sci.electronics
Subject: Re: Sidereal Time Clock
Summary:
Expires:
References: <1833@umriscc.isc.umr.edu> <3023@lamont.ldgo.columbia.edu> <4391@alliant.Alliant.COM>
Sender:
Followup-To:
Distribution: sci.electronics
Organization: Northwestern University
Keywords: dbell@cup.portal.com (David J Bell) (12/30/90)
By now, you've probably figured out that you only need to post once. You'll see the echo LATER, not immediately! You may also have figured out that skipping every n'th "beat" of a 60 Hz timebase will make your clock run SLOWER, not faster, as is required of a sidereal clock... Dave dbell@cup.portal.com
squishy@casbah.acns.nwu.edu (Shishin Yamada) (12/30/90)
Dave and others who noticed:
Sorry, I am new to the newsgroup and did not mean to mess up with an
echo. I also goofed in the fact that the clock runs SLOWER not FASTER. The
gist of the post should have been:
Using a digital clock that obtains it's timing pulses from the 60Hz
line, which is typically sent into a wave-shaped to produce a cleaner
square wave output of 60Hz. You would want to skip a pulse every 14,400
pulse. This clock would count a normal 240 seconds, then count 61/60th of a
second. This correction would thus make the clock count to 23:56:00 in a 24
hour day. The maximum error (assuming the AC line is OK) is 1/60th of a
second off actual sidereal time in 14,400 ticks. Or a relative error of 1
part in 14,400 which is equal to a relative error of 0.0069444% (percent).
I would think that error would be acceptable, considering it corrects
itself every four minutes. One possible counter for this would be two
divide-by-ten serial counters in series, then sequential logic to determine
if the result is equal to 144 decimal.
One possible binary clock circuit driven by an AC line is given in the
January 1991 issue of Popular Electronics. It could be modified with
appropriate logic circuits.
Once again, sorry to all for the screwed up info in my first (ever)
post. Best of luck with your project!
*************************************************************************
Shishin "SQUISH" Yamada * Undergrad Senior, Electrical Engineering
BitNet: Squishy@Casbah.ACNS.NWU.EDU * Northwestern University, Evanston IL
**************************************************************************wilner@motcid.UUCP (Corey S. Wilner) (01/02/91)
From what I understand, using the 60Hz line as a reference for timing is not a good practice if you want any accuracy. I have heard that cycles can be longer or shorter than 1/60th of a second and some cycles can be lost completely. Anyone care to comment on this with more substance other than my glaring generalities?! *********************************************** Corey S. Wilner | Give me a jingle: Motorola Cellular | ..!uunet!motcid!wilner 708-632-7206 | ***********************************************
squishy@casbah.acns.nwu.edu (Shishin Yamada) (01/03/91)
In article <5972@mint17.UUCP> wilner@motcid.UUCP (Corey S. Wilner) writes: >From what I understand, using the 60Hz line as a reference for timing is not >a good practice if you want any accuracy. I have heard that cycles can be >longer or shorter than 1/60th of a second and some cycles can be lost completely. >Anyone care to comment on this with more substance other than my glaring >generalities?! > Corey: So far, there have been many previous messages about the 60Hz timebase of the AC line "being fixed-in-phase" with the rest of the nation. I come from Illinois, and our state has the highest percentage of nuclear power use in the U.S. brought to us by Commonwealth Edison (so far so good :) ). Anyways, I used to hear the worst things from my high school electronics teacher about Florida's infamous 'brown-outs.' I do believe a lot of timing electronics does make use of the AC line (Most VCR's do), and I'm surprised because Japan (biggest manufacturer) has both 50/60Hz @ 100V rms. Their modern clocks now do not have a 50/60Hz switch. One case in point is that the battery back-up mentioned will NOT be viable with the 60Hz method. Lastly, I'd like to apologize to everyone for the unintended 'echos' of my original message. I'm sorry that happened, I do not know why it did, I only see it once on our site (although messed up), and I have revieved a lot of concerned e-mail about it. ---------------------------------------------------------------------- Shishin "Squish" Yamada * BitNet: squishy@casbah.acns.nwu.edu EE Class of '91 * Northwestern University ----------------------------------------------------------------------
squishy@casbah.acns.nwu.edu (Shishin Yamada) (01/03/91)
On a side-note, if you might try modifying the clock frequency by
Message-ID: <2369@casbah.acns.nwu.edu>
Date: 29 Dec 90 12:06:30 GMT
References: <1833@umriscc.isc.umr.edu> <3023@lamont.ldgo.columbia.edu> <4391@alliant.Alliant.COM>
Followup-To: 4391@alliant.Alliant.COM
Distribution: sci.electronics
Organization: Northwestern University
Lines: 35
"missing beats," that is, every nth oscillation, skip the clock's
increment. On a 24h:00m:00s normal day, there are 86,400 seconds. If the
clock counted by using the 60Hz line, this makes 5,184,000 beats. On a
23h:56m:00s sideral day (24 normal hours to us), there are 86,160 seconds.
This makes for 5,169,600 beats on a 60Hz line.
Therefore, if you skipped a beat every 14,400 beats of a 60Hz line,
your clock would then count sideral time. In this manner, you could have a
counter that went up to 14,400 beats, and then have it blank the next
(14,401th beat) of the 60Hz line. This correction would occur every 240
seconds (4 minutes). This introduces a very small error into your clock,
since the 240th second will be 59/60th of a normal second, but not much as
it corrects itself every four minutes. Pretty accurate, I think.
It's a weird idea, but then again, I just thought of it! Hope you like it!
If you'd like more of my weird ideas, please e-mail me on BitNet. I
am fastest to reply on that medium!
======================================================================
Shishin "SQUISH" Yamada | BitNet: squishy@casbah.acns.nwu.edu
Senior Undergraduate, EE | Compuserve: 76636,3254
Northwestern University | Prodigy: KTVB99A
=======================================================================
Newsgroups: sci.electronics
Subject: Re: Sidereal Time Clock
Summary:
Expires:
References: <1833@umriscc.isc.umr.edu> <3023@lamont.ldgo.columbia.edu> <4391@alliant.Alliant.COM>
Sender:
Followup-To:
Distribution: sci.electronics
Organization: Northwestern University
Keywords: rsd@sei.cmu.edu (Richard S D'Ippolito) (01/04/91)
In article <5972@mint17.UUCP> wilner@motcid.UUCP (Corey S. Wilner) writes: >From what I understand, using the 60Hz line as a reference for timing is not >a good practice if you want any accuracy. I have heard that cycles can be >longer or shorter than 1/60th of a second and some cycles can be lost completely. >Anyone care to comment on this with more substance other than my glaring >generalities?! Most, if not all, power companies are connected to a wide-area grid. Any generator on the grid which gets more than a few degrees out of phase with the others will cause huge reactive circulating currents which will result in lots of line losses. Any generator which gets 180 degrees out will be a sink and find itself the recipient of one heck of a lot of energy for a short, but spectacular, while! Utilities are very careful about their physical plant, and have extensive protection systems to prevent the above. Therefore, you will find that a whole grid may be ahead or behind "real" time by several cycles and that sections of the grid may be slightly ahead or behind that by very small fraction of a cycle. Usually, but not always, by the end of the day the difference is cleared by adjusting the frequency of the grid so that the correct number of cycles (60*60*60*24) have occurred. This is done according to allowed correction rates (how fast or slow they can run the generators without causing other problems). I did say "usually": During long periods of extra heavy loads, the grid generators slow down and the utility group may decide not to make up all of the lost cycles by the end of the day, or the amount to be made up can't be completed at the maximum correction rate. Very light loads (a major feeder goes out for a while) causes them to run fast, and again, the gain during the transient may not be fully compensated for. I've called my local electric utility and seem to remember that an error of a second or so, i.e., 100 cycles lost or gained out of 5M (60Hz +/- 0.001Hz) could occur in unusual circumstances, so if you need better than that, use a separate reference. Remember, the utility is charged with delivering energy, not providing a frequency standard. If you decide to call your utility, ask for someone in the engineering department and tell the engineer you want to know the normal cycle loss or gain at the end of the day and the maximum correction rates that they will apply. For short terms, only the correction rate would matter (for example, in a 1-2 Hr camera exposure). For longterm time keeping, forget it! Rich
henry@zoo.toronto.edu (Henry Spencer) (01/05/91)
In article <5972@mint17.UUCP> wilner@motcid.UUCP (Corey S. Wilner) writes: >From what I understand, using the 60Hz line as a reference for timing is not >a good practice if you want any accuracy. I have heard that cycles can be >longer or shorter than 1/60th of a second and some cycles can be lost completely. Power frequency varies a bit in the short term to suit the power companies' convenience, but over the long term it is *extremely* accurate, referenced to atomic clocks and very carefully kept spot-on. They even crank in the leap seconds. (One word of caution, however: this applies only if your power company is part of one of the big grids. Little local outfits with no outside ties are reportedly a bit sloppy at times. I'm not sure how many of those are left, mind you...) -- "The average pointer, statistically, |Henry Spencer at U of Toronto Zoology points somewhere in X." -Hugh Redelmeier| henry@zoo.toronto.edu utzoo!henry
rrw@naucse.cse.nau.edu (Robert Wier) (01/11/91)
In talking about synching power generators with a grid, about
a year ago, I lead an IEEE student group tour down to the Childs
hydro power plant (central Az). This is very interesting since
it was one of the first plants operating in Az (and the oldest still
on line). You go look at the equipment, and much of it is dated
1890 or 1905.
Anyway, it is a very small operation, about 500kw run from a small
resevoir fed by springs. If I remember correctly there is about
300' of head. When it was first brought online, they said it
supplied all the needs of Northern Arizona, and half of Phoenix!
I was suprised to learn when the plant was shut down that they didn't
disconnect from the grid (for example, to clean the pipes and
turbine valves). They just let the grid power the generators as
large electric motors. Said it was a whole lot easier than trying
to bring them up from a dead stop and re-synch them to the grid
(which had to be done by hand). They had the original Westinghouse
synch-meter which was about the size of a wall clock. It as
(was) calibrated in + and - degrees, with a big crank to make
speed adjustments. The tech on duty indicated that he was there
once when someone goofed and put the plant on line with a 3 degree
difference. He said it sounded like a cannon going off and
shook the whole building. Didn't do any damage though.
I found it suprising that it was still economical to operate a
plant of this size. But basically the capital expense was paid
off long ago, the equipment doesn't wear out very fast, and it only
takes a couple of guys to run it. Mainly they just have to keep
watch on the resevoir water level to make sure it doesn't get
too low.
It was nice for the students as well since all of the equipment
is open-frame, and you can see the rotors going 'round and so forth
since there aren't any covers. REALLY noisy, though.
- Bob Wier
-------------- insert favorite standard disclaimers here ----------
College of Engineering
Northern Arizona University / Flagstaff, Arizona
Internet: rrw@naucse.cse.nau.edu | BITNET: WIER@NAUVAX | WB5KXH
or uucp: ...arizona!naucse!rrwrsd@sei.cmu.edu (Richard S D'Ippolito) (01/12/91)
In article <3122@naucse.cse.nau.edu> Robert Wier writes: > I was suprised to learn when the plant was shut down that they didn't > disconnect from the grid (for example, to clean the pipes and > turbine valves). They just let the grid power the generators as > large electric motors. Said it was a whole lot easier than trying > to bring them up from a dead stop and re-synch them to the grid > (which had to be done by hand). There's another reason that _very_ large horizontally-mounted generators are never stopped -- if they sat there for a long time, the weight would cause a permanent flat spot in the shaft and bearings! Rich