Re: [time-nuts] Information on the Danjon Astrolab
In article [EMAIL PROTECTED], Brooke Clarke [EMAIL PROTECTED] writes Hi: After seeing a reference to Splitting the Second by Tony Jones on Tom's web page I read it and found that the Photographic Zenith Telescope was replaced by the Danjon Astrolabe. But Google provides almost no information about it. A patent search has turned up a number of optical systems, mostly for spacecraft, that use star separation for navigation, and if used at a know location on the earth could be used for timing. For these see my Stellar Time Keeping web page: http://www.pacificsites.com/~brooke/StellarTime.shtml Does anyone know how a Danjon Astrolabe works? IIRC, the Danjon astrolabe uses a mercury pool, and a semi-reflecting prism. The optical telescope is aligned horizontally, and the prism provides 2 light paths, one aimed upward at some angle (normally 45 or 60 degrees) and the other downwards. The mercury pool reflects the downward beam upwards. As a result, when you look through the telescope, you see double images. As the skies go through their apparent diurnal rotation, objects will pass through the nominal viewing angle, and the image doubling will go away. The observer presses a button at the magic moment to record the time. Advantages - can use more stars, not just those at the zenith. can observe more stars in a night, leading to improved accuracy in time determination. Disadvantage - requires manual operation, with resultant errors due to personal equation - the eye-hand delay. Accuracy of the Danjon astrolabe and the PZT were considered similar, but the automated capabilities of the PZT won out - I think. Source for this - the book Geodesy by Bomford. Sorry, no ISBN or publisher (I'm typing from memory) Have Fun, Brooke Clarke -- w/Java http://www.PRC68.com w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml http://www.precisionclock.com ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts -- Geoff Powell ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
In article [EMAIL PROTECTED], Geoff Powell [EMAIL PROTECTED] writes Does anyone know how a Danjon Astrolabe works? snip Source for this - the book Geodesy by Bomford. Sorry, no ISBN or publisher (I'm typing from memory) Geodesy Bomford, Guy Clarendon Press, Oxford 2nd Edition (1965) (Out of Print) 3rd Edition (1971) ISBN 0198519192 (Out of Print) 4th edition, from 1980. ISBN 0-19-851946-X Or you can try this Biblioquest search http://www.biblioz.com/main.php?action=5u=24b729e881416a4d4feb2da5a9d34 ed3author=Bomfordtitle=Geodesy It is considered one of the primary references for geodetic surveying, albeit somewhat dated. -- Geoff Powell ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
In message [EMAIL PROTECTED], Geoff Powell writes: IIRC, the Danjon astrolabe uses a mercury pool, and a semi-reflecting prism. Hmm, but doesn't the local value of horizontal affect the precision also ? You would really need to know the direction of local gravity for this to be any good... -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 [EMAIL PROTECTED] | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
From: Poul-Henning Kamp [EMAIL PROTECTED] Subject: Re: [time-nuts] Information on the Danjon Astrolab Date: Tue, 08 Aug 2006 07:59:00 + Message-ID: [EMAIL PROTECTED] In message [EMAIL PROTECTED], Geoff Powell writes: IIRC, the Danjon astrolabe uses a mercury pool, and a semi-reflecting prism. Hmm, but doesn't the local value of horizontal affect the precision also ? You would really need to know the direction of local gravity for this to be any good... Down? :-) Wouln't a number of measurements and comparision with other methods and measurements help to reveal such deviations? Then again, question is if the deviation is so large that it has a real impact. Cheers, Magnus ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
In message [EMAIL PROTECTED], Magnus Danielson wri tes: You would really need to know the direction of local gravity for this to be any good... Down? :-) There's a big difference between down enough that cows don't easily tip over and passes through the center of the earth. I know the Carlsberg Meridian had to take it into account when it was moved down to the Canary Islands. Wouln't a number of measurements and comparision with other methods and measurements help to reveal such deviations? Then again, question is if the deviation is so large that it has a real impact. Obviously, it all depends how precise you want it. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 [EMAIL PROTECTED] | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
On Tue, 08 Aug 2006 08:33:32 +, Poul-Henning Kamp [EMAIL PROTECTED] wrote: There's a big difference between down enough that cows don't easily tip over and passes through the center of the earth. LOL ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
On Tue, 08 Aug 2006 01:44:45 -0700, Rex [EMAIL PROTECTED] wrote: On Tue, 08 Aug 2006 08:33:32 +, Poul-Henning Kamp [EMAIL PROTECTED] wrote: There's a big difference between down enough that cows don't easily tip over and passes through the center of the earth. LOL Oh, you reminded me... A friend of mine once told me a long elaborate story about a mountainous section of the world where they had two breeds of cattle -- The clockwise and the counter-clockwise. It depended on which legs had evolved longer to let them stand upright and eat the grass on the steep mountain sides. Obviously, cross breeding between the two breeds was impossible in nature, and if man got involved with his special equipment to help them breed, you never knew what to expect. You could get a normal counter-clockwise or a clockwise or you could get a new breed of flatland cow or you could get a cow that could only stand upright on a hill so steep that they would just slide to the bottom. (A wall hugger?) Sorry. Hope my off topic posting doesn't start a trend here. ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
From: Poul-Henning Kamp [EMAIL PROTECTED] Subject: Re: [time-nuts] Information on the Danjon Astrolab Date: Tue, 08 Aug 2006 08:33:32 + Message-ID: [EMAIL PROTECTED] In message [EMAIL PROTECTED], Magnus Danielson wri tes: You would really need to know the direction of local gravity for this to be any good... Down? :-) There's a big difference between down enough that cows don't easily tip over and passes through the center of the earth. Hence the smiley. I know the Carlsberg Meridian had to take it into account when it was moved down to the Canary Islands. Ofcourse they would have to, they moved down on the longitude on an ellipsoid and then ofcourse they have a different local gravity and the gradient of that will differ (and hence the angle of the gravity will differ slightly from that of the normal on the ellipsoid). Ever looked at a gravity map of the earth? Wouln't a number of measurements and comparision with other methods and measurements help to reveal such deviations? Then again, question is if the deviation is so large that it has a real impact. Obviously, it all depends how precise you want it. Indeed. Cheers, Magnus ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
Hi Geoff: Thanks for the reference to Geodesy. Do you know if the first edition has the Danjon information? I ask because the second editions are rather pricey and the 3rd and 4th are not available at all. Could someone on this list make a copy of the Danjon section? The PZT used a large pool of Mercury to define Up. The source of many errors was considered in the design and as far as I know were all eliminated. The PZT worked by exposing a glass plate at 4 known times. I think it was reversed for two of the exposures and then after development read on what amounted to a coordinate measuring machine. Then a number of corrections could be made depending on the geometric relation between the 4 points. The Dent Meridian Instrument or as he called it the Dipleidscope uses a couple of mirrors behind a clear glass and when the two reflections coincide the star is on the meridian. The manual gives a number of ways of aligning it, one of which depends on already knowing the time. See: http://www.pacificsites.com/~brooke/Dent.shtml But it's intended for visual use, not automated. Have Fun, Brooke Clarke Geoff Powell wrote: In article [EMAIL PROTECTED], Geoff Powell [EMAIL PROTECTED] writes Does anyone know how a Danjon Astrolabe works? snip Source for this - the book Geodesy by Bomford. Sorry, no ISBN or publisher (I'm typing from memory) Geodesy Bomford, Guy Clarendon Press, Oxford 2nd Edition (1965) (Out of Print) 3rd Edition (1971) ISBN 0198519192 (Out of Print) 4th edition, from 1980. ISBN 0-19-851946-X Or you can try this Biblioquest search http://www.biblioz.com/main.php?action=5u=24b729e881416a4d4feb2da5a9d34 ed3author=Bomfordtitle=Geodesy It is considered one of the primary references for geodetic surveying, albeit somewhat dated. -- w/Java http://www.PRC68.com w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml http://www.precisionclock.com ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
[time-nuts] GPS SPD field
Hi, Reference a Rockwell Jupiter GPS receiver. Does anyone know if the decimal point in the NMEA $GPRMC 'SPD' ASCII field moves around? At rest it reports 0.000 but it is not obvious whether this format is maintained on the move or expands/contracts to suit. Does it keep the same precision with the most significant digits expanding or not? TIA - Joe ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
In article [EMAIL PROTECTED], Brooke Clarke [EMAIL PROTECTED] writes Hi Geoff: Thanks for the reference to Geodesy. Do you know if the first edition has the Danjon information? I ask because the second editions are rather pricey and the 3rd and 4th are not available at all. Could someone on this list make a copy of the Danjon section? I'm quoting from memory about the workings of the Danjon Astrolabe - the book was borrowed from my local library, and I don't remember which edition it was. Nor can I quote chapter and verse, I fear - although I will attempt to re-borrow the book, and copy the relevant section. Later this week, I hope. AFAIR, the discussion was of the workings of various instruments for determination of time from the stars, together with ways to mitigate the systematic errors in each instrument. Personal equation figured large in all this, which is why the PZT was preferred. -- Geoff Powell ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] GPS SPD field
Joe, The decimal should stay in the same place. Places to the left of the decimal will naturally grow as needed to handle the calculated value. Most all NMEA data fields will adjust to handle the data, but different manufacturers handle the details differently. Some will default to all null fields when first turned on: $GPRMC,,*XX, while some will fill the fields with zeroes: $GPRMC,00.00,V,.,N *XX. Makes life a little more interesting if you are writing a parsing routine. Randy -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Joe McElvenney Sent: Tuesday, August 08, 2006 11:03 AM To: time-nuts@febo.com Subject: [time-nuts] GPS SPD field Hi, Reference a Rockwell Jupiter GPS receiver. Does anyone know if the decimal point in the NMEA $GPRMC 'SPD' ASCII field moves around? At rest it reports 0.000 but it is not obvious whether this format is maintained on the move or expands/contracts to suit. Does it keep the same precision with the most significant digits expanding or not? TIA - Joe ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
Guys, First off, what I am about to do is ask a REALLY STUPID question, but more and more of the GPS stuff I do is drifting towards the precision timing end of things, so I thought I should ask. I have been seeing a lot of traffic concerning making 10MHz frequency dividers using PIC's. While they provide an elegant solution to providing an accurate 1PPS from a precision source, I have to ask if there is a reason for going this route? I am just using three HCT40103 down counters hooked to a DS4000 to get what I think is a very stable 1PPS. Am I missing something? I realize 40103's are as old as dirt (I guess I am showing my 4000 series CMOS days), but the HCT series have plenty of bandwidth. Please be gentle Randy ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] GPS SPD field
Hi Randy, I have seen at least one GPS (Panasonic) that sends 5 digitis after the decimal point for position in NMEA, and at least one commercial navigation program for PDA's that cannot handle 5 digits! They hard-coded the parsing routine to be stupid enough to always expect 4 digits, and it crashes when they receive 5 digits of precision... bye, Said ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
Hi Randy, I guess they like the intellectual challenge of tweaking PIC bits :) There is what I think may be a better way to generate the 1PPS: you can use an Arm micro such as the Philips LPC2102 series to generate a 1PPS output with 16.7ns settability: these micros have counter/timers with 32 bits of resolution, they can run at 60MHz (6x the input of 10MHz using an internal PLL), and can generate a pulse output (counter Match) anywhere in the second with 16.7ns resolution. They can even generate multiple outputs, say 1PPS, 10PPS, 100PPS etc all in paralell. On top of that, they can generate a software-selectable pulse-width, and can synchronize themselves to a GPS 1PPS input. Best of all, they are $2 in volume. For cheap eval boards, see: _http://www.sparkfun.com/commerce/product_info.php?products_id=667_ (http://www.sparkfun.com/commerce/product_info.php?products_id=667) That board can generate 5 independent pulse-outputs for $30. bye, Said ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
Let me add this photo - I found in a box my first attempt to make a PPS divider as well as an early breadboard prototype of the much simpler, PIC-based, divider. http://www.leapsecond.com/pages/ppsdiv/ver1.jpg /tvb ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] GPS SPD field
Said, Yeah, I meant to mention that. You really just need to parse on the commas and then figure out what was in between them. I have seen 2, 3, 4, and 5 place precision on varying receivers. I have to laugh at the marketing weenie who thought 5 place precision would help him sell standard, non-differential positioning receivers. Actually, advertising 5 place precision probably DID sell units to the technically challenged. Randy -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Tuesday, August 08, 2006 1:34 PM To: time-nuts@febo.com Subject: Re: [time-nuts] GPS SPD field Hi Randy, I have seen at least one GPS (Panasonic) that sends 5 digitis after the decimal point for position in NMEA, and at least one commercial navigation program for PDA's that cannot handle 5 digits! They hard-coded the parsing routine to be stupid enough to always expect 4 digits, and it crashes when they receive 5 digits of precision... bye, Said ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
Randy Warner said the following on 08/08/2006 03:23 PM: I have been seeing a lot of traffic concerning making 10MHz frequency dividers using PIC's. While they provide an elegant solution to providing an accurate 1PPS from a precision source, I have to ask if there is a reason for going this route? I am just using three HCT40103 down counters hooked to a DS4000 to get what I think is a very stable 1PPS. Am I missing something? I realize 40103's are as old as dirt (I guess I am showing my 4000 series CMOS days), but the HCT series have plenty of bandwidth. Hi Randy -- I think the concern in using the older discrete devices is their potential for jitter in general, and temperature sensitivity on top of that. But I've never done any experiments on just how big a problem those issues are. John ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
From: Tom Van Baak [EMAIL PROTECTED] Subject: Re: [time-nuts] Frequency Dividers Date: Tue, 8 Aug 2006 14:16:24 -0700 Message-ID: [EMAIL PROTECTED] Let me add this photo - I found in a box my first attempt to make a PPS divider as well as an early breadboard prototype of the much simpler, PIC-based, divider. http://www.leapsecond.com/pages/ppsdiv/ver1.jpg However, I would not use any of them for any real work, since the MTBF of those breadboards isn't longterm friendly. ;O) A small CPLD such as XC9536 would also do it in one chip, you program in friendly VHDL and know it will do the right thing. ;O) Cheers, Magnus ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
John, I totally understand the concerns about stability. In the experiments I have run the jitter is down in the ps range. Temp would be a whole other animal. I have just been running these in the lab. Oh well, it's all fun. Randy -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of John Ackermann N8UR Sent: Tuesday, August 08, 2006 2:31 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Frequency Dividers Randy Warner said the following on 08/08/2006 03:23 PM: I have been seeing a lot of traffic concerning making 10MHz frequency dividers using PIC's. While they provide an elegant solution to providing an accurate 1PPS from a precision source, I have to ask if there is a reason for going this route? I am just using three HCT40103 down counters hooked to a DS4000 to get what I think is a very stable 1PPS. Am I missing something? I realize 40103's are as old as dirt (I guess I am showing my 4000 series CMOS days), but the HCT series have plenty of bandwidth. Hi Randy -- I think the concern in using the older discrete devices is their potential for jitter in general, and temperature sensitivity on top of that. But I've never done any experiments on just how big a problem those issues are. John ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] GPS SPD field
On Tue, August 8, 2006 22:33, [EMAIL PROTECTED] said: Hi Randy, I have seen at least one GPS (Panasonic) that sends 5 digitis after the decimal point for position in NMEA, and at least one commercial navigation program for PDA's that cannot handle 5 digits! Have seen 6 digits as well. (Javad) I agree, that hard coding message lengths in NMEA messages are bad for portability. -- Björn ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Ugly Frequency Dividers
Well, since I am not versed in anything but Fortran or Basic, I would have to go the old fashioned way. - Mike Mike B. Feher, N4FS 89 Arnold Blvd. Howell, NJ, 07731 732-886-5960 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Randy Warner Sent: Tuesday, August 08, 2006 5:38 PM To: Tom Van Baak; Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Ugly Frequency Dividers Tom, Attached is my entry in the butt-ugly PPS divider contest. A whole lot of 74HC192's and 74HC74's. I think this is circa 1999 or so. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tom Van Baak Sent: Tuesday, August 08, 2006 2:16 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Frequency Dividers Let me add this photo - I found in a box my first attempt to make a PPS divider as well as an early breadboard prototype of the much simpler, PIC-based, divider. http://www.leapsecond.com/pages/ppsdiv/ver1.jpg /tvb ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.405 / Virus Database: 268.10.7/411 - Release Date: 8/7/2006 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.405 / Virus Database: 268.10.7/411 - Release Date: 8/7/2006 ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Ugly Frequency Dividers
That remembers me the ugliest (or most exotic?) frequency divider i ever saw: It was part of a Marconi (damn, i don't remember the model number, but it was an OLD model), more precisely the 600 MHz divide by ten prescaler. The input divider was based on two tunnel diodes that acted as a div. by two divider followed by the really most bizarre divide by five unit i ever saw: Fifteen discrete NPN transistors arranged in a star (or pentagon (Hell Echelon!! ) ) topology with the input placed at the center of the star. The 15 transistor were in a symmetrical loop of 5 three transistor units working in a closed loop. Years later, i've made searches to find the schematic of this prescaler, but without the model number, this is quasi-impossible. If one of this forum's members has this schematic, i would be pleased to ask for a copy! Damn! That strange star all-transistor divider could count up to 300 MHz after all!!! Have a good day gang and keep this forum running!! 73 de Normand Martel VE2UM Montreal, Quebec, Canada --- Randy Warner [EMAIL PROTECTED] wrote: Tom, Attached is my entry in the butt-ugly PPS divider contest. A whole lot of 74HC192's and 74HC74's. I think this is circa 1999 or so. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tom Van Baak Sent: Tuesday, August 08, 2006 2:16 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Frequency Dividers Let me add this photo - I found in a box my first attempt to make a PPS divider as well as an early breadboard prototype of the much simpler, PIC-based, divider. http://www.leapsecond.com/pages/ppsdiv/ver1.jpg /tvb ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Information on the Danjon Astrolab
Rasputin Novgorod wrote: ofcourse they have a different local gravity and the Wait a minute. The center of gravity ~is~ the center of gravity. You can't have two, or multiple centers. The strength of gravity varies from place to place, but that doesn't change the direction. If I'm wrong; please explain... You are wrong. Here's my attempt at an explanation. The direction of gravity -- the direction of the plumb line -- is everywhere normal to the local gravitational equipotential surface. (The particular gravitational equipotential surface that corresponds to mean sea level is called the geoid.) The geoid is is not exactly an ellipsoid of revolution, although such as ellipsoid (such as the WGS-84 ellipsoid, for example) is used to define the location of the graticule - the grid of meridians of longitude and parallels of latitude that we use for a horizontal datum. For many surveying purposes, a different reference surface, the geoid, is used instead of the ellipsoid. The direction of the plumb line (the down direction) is normal to the local gravitational equipotential surface -- that is, normal to the geoid if you're location is at sea level. The normal to the ellipsoid andthe normal to the geoid do not point directly at the earth's center of mass (the geocentre). They might point at the geocentre if the earth did not rotate on its axis, but the earth's rotation, and the centrifugal force that we observe as we rotate with it, causes the geoid to fit more closely to an ellipsoid than to a perfect sphere. The angular difference between the normal to the ellipsoid and the plumb line (the normal to the geoid) is called the deflection of the vertical, and is typically a few seconds of arc. (The deflection of the vertical was discovered during the 19th century by British surveyors during the great survey of India. The same survey organization, also discovered that Mount Everest was so tall, and named it after their leader.) If the earth was a perfect sphere (which it could only be if it didn't rotate on its axis), and was perfectly uniform in its mass density, then all plumb lines would pass through the geocentre. But the earth does rotate, and it has mountains and seas with different densities of the underlying rocks, so the geoid is not spherical, and the plumb lines do not generally pass through the geocentre. -- James Maynard Salem, Oregon, USA ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
In a message dated 8/8/2006 14:29:38 Pacific Daylight Time, [EMAIL PROTECTED] writes: Hi Randy, the Micro's allow you to do other stuff, like discipline the OCXO as well :) BTW: I took a closer look at the Sparkfun board, the chip has a bunch more Match outputs than I thought. The Philips parts have great counters/PWM units (driven by an internal PLL) for experiments.. The MSP430 should do the trick as well. bye, Said Said, Thanks for the clue on the Philips part. Sounds like a nice solution. I just got a couple of sample parts from Maxim that I am going to play with (DS4000, MAX5121, and MAX6629). If they can do what Maxim claims it might be a fun experiment. I have one of those little TI MSP430 dev kits that I am going to try to use to run the show. We'll see. Randy ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] GPS SPD field
In a message dated 8/8/2006 14:23:31 Pacific Daylight Time, [EMAIL PROTECTED] writes: Hi Randy, if I remember right, the fourth digit is about 18cm in resolution? Yeah, let's get 1.8cm resolution on the fifth digit! bye, Said Said, Yeah, I meant to mention that. You really just need to parse on the commas and then figure out what was in between them. I have seen 2, 3, 4, and 5 place precision on varying receivers. I have to laugh at the marketing weenie who thought 5 place precision would help him sell standard, non-differential positioning receivers. Actually, advertising 5 place precision probably DID sell units to the technically challenged. ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] GPS SPD field
Said, Yeah, that sounds about right. Whatever it is, it's WAAAY down in the noise, but it sure LOOKS good on a data sheet. I had a similar problem several years ago where we were in danger of losing a sale because the competing receiver (Raytheon Marine) was so much more stable according to the customer. He was absolutely right. Of course, if you put the unit in a dragster and ran a 6 second 1/4 mile the GPS still thought it was looking at the Christmas Tree lights at the starting line when you were finished and putting the dragster back in the trailer... It was so heavily filtered it almost behaved like it was in position hold mode. Randy __ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Tuesday, August 08, 2006 5:04 PM To: time-nuts@febo.com Subject: Re: [time-nuts] GPS SPD field In a message dated 8/8/2006 14:23:31 Pacific Daylight Time, [EMAIL PROTECTED] writes: Hi Randy, if I remember right, the fourth digit is about 18cm in resolution? Yeah, let's get 1.8cm resolution on the fifth digit! bye, Said Said, Yeah, I meant to mention that. You really just need to parse on the commas and then figure out what was in between them. I have seen 2, 3, 4, and 5 place precision on varying receivers. I have to laugh at the marketing weenie who thought 5 place precision would help him sell standard, non-differential positioning receivers. Actually, advertising 5 place precision probably DID sell units to the technically challenged. ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
I understand that the micros work fine, it just seemed like you would need a second one to discipline the OCXO because of the timing constraints on the divider. Correct. But at $2+/- each it's easy to use one uC for a divider and another uC for the GPSDO algorithm. Yet another for an IRIG option; perhaps another for NTP. I think I remember RickH saying that his CNS II clocks use up to 5 microprocessors. /tvb ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
OK, I give up. I guess I'm just an old SSI guy. Sigh. Randy __ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Tom Van Baak Sent: Tuesday, August 08, 2006 5:39 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Frequency Dividers I understand that the micros work fine, it just seemed like you would need a second one to discipline the OCXO because of the timing constraints on the divider. Correct. But at $2+/- each it's easy to use one uC for a divider and another uC for the GPSDO algorithm. Yet another for an IRIG option; perhaps another for NTP. I think I remember RickH saying that his CNS II clocks use up to 5 microprocessors. /tvb ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
Tom Van Baak wrote: I understand that the micros work fine, it just seemed like you would need a second one to discipline the OCXO because of the timing constraints on the divider. Correct. But at $2+/- each it's easy to use one uC for a divider and another uC for the GPSDO algorithm. Yet another for an IRIG option; perhaps another for NTP. I think I remember RickH saying that his CNS II clocks use up to 5 microprocessors. Actually, you don't always need to. Many uCs will provide timer inputs that you can wiggle in various ways to get the behavior you want in an interrupt fashion, where the number of cycles it takes to get the interrupt is predictable. You can then schedule the other functions (like poking the osc.) immediately afterwords, and know that they'll have enough time to complete. Hard to do with divide by 10; easy to do with divide by 1000 if you just want 10Khz on down output from your divider. If you want to be more clever, you can get some of them to act as dividers in hardware for the divide by 10 or 100. Gets a little more tricky programming the timer/counter on the uC, but it's doable. I've got some starter C/asm code for the atmel chips if anyone wants it as a starting point. (It's not complete; I got distracted trying to debug a broken LCD display...) -Dave ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
I have been seeing a lot of traffic concerning making 10MHz frequency dividers using PIC's. While they provide an elegant solution to providing an accurate 1PPS from a precision source, I have to ask if there is a reason for going this route? I am just using three HCT40103 down counters hooked to a DS4000 to get what I think is a very stable 1PPS. Am I missing something? I realize 40103's are as old as dirt (I guess I am showing my 4000 series CMOS days), but the HCT series have plenty of bandwidth. The answer probably depends upon the context. For low volume, design time and/or time to market is probably more important than small differences in cost. That means you normally use the tools you have and/or are familiar with. For high volume, parts cost is important. Old, mature, parts are often inexpensive. TI says $0.59 for the CD74HCT40103, qty 1K. Strange, Digikey says $0.36. That seems high to me for a jelly-bean type part so I'd guess that one is old enough to be past mature. (TI says they got it from Harris.) Digikey says $0.96 for the PIC12F629, Q100. So even ignoring the board space, it looks cheaper. (That's ignoring the NRE for the software and the cost of programming each chip.) The main reason I'd choose one over the other for a hobby project is the mood I'm in and/or unspecified design options. I happen to get a kick out of writing small, clean chunks of code, so I would probably go that way all other things being equal. The other main advantage of software is that design changes are often easier. (That's assuming the programmer is still around.) I understand that the micros work fine, it just seemed like you would need a second one to discipline the OCXO because of the timing constraints on the divider. It's often possible to do two things with one CPU. It's much more interesting if they are both timing critical, especially if you are counting cycles to get very tight control. A second CPU may be cheaper/simpler, but sometimes it's fun writing that sort of code. For something as simple as a clock divider, I'd probably try to push all that work into the timer hardware. -- The suespammers.org mail server is located in California. So are all my other mailboxes. Please do not send unsolicited bulk e-mail or unsolicited commercial e-mail to my suespammers.org address or any of my other addresses. These are my opinions, not necessarily my employer's. I hate spam. ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Re: [time-nuts] Frequency Dividers
I have used a number of pll controlled microcontrollers, and I would not recommend using one of those in a timing application such as those discussed here. These PLLs are generally not very clean spectrally (it's actually a good thing for EMI, some chips have purposeful spread spectrum clocks) and may have lots of jitter. However, most of the chips with PLL will let you disable the PLL and run from a crystal or an external oscillator. Alternately, it you use the timer instead of software loops, you can run the core from the PLL as long as the timer itself is not driven from the PLL. I use the Silabs C8051F133 in several projects and it will run with up to a 100 MHz clock (with many instructions running in one clock cycle) from the PLL or I believe 50 MHz with an external oscillator. And if you needed a 16 x 16 MAC engine for that counter, it has that too :-) http://www.silabs.com/public/documents/tpub_doc/dsheet/Microcontrollers/Precision_Mixed-Signal/en/C8051F12x-13x.pdf That's not your father's 8051!!! For timing applications, it has 5 general purpose 16 bit timers and a 6 channel 16 bit Programmable Counter Array, so by using one of the 16 bit timer as a prescaler for the PCA, you can have create up to 10 low-jitter timer outputs with 6 of them having up to 32 bits capacity, with minimum software overhead, so the CPU is mostly available for anything else you might want to do with it.. Pretty neat, uh? The other day, a rep for a well known semiconductor/microcontroller company that shall remain nameless was showing me their latest ad for an 8051 running at 50 MHz with a 4 clock core and in big letters: FASTEST 8051 AVAILABLE. I pointed him to the Silabs web site and left him there... Didier KO4BB PS: only problem for the hobbyist, it only comes in a surface mount 64 or 100 pins TQFP (cheap development boards are available, with JTAG programmer, prototyping area and serial interface). I am not associated with Silabs, however I am a very satisfied customer and I can recommend their products (hardware and software), they are topnotch and Silabs customer service is excellent. I routinely use them in products that operate way outside the generous -40 to +85 C temperature range without any problem. [EMAIL PROTECTED] wrote: In a message dated 8/8/2006 14:29:38 Pacific Daylight Time, [EMAIL PROTECTED] writes: Hi Randy, the Micro's allow you to do other stuff, like discipline the OCXO as well :) BTW: I took a closer look at the Sparkfun board, the chip has a bunch more Match outputs than I thought. The Philips parts have great counters/PWM units (driven by an internal PLL) for experiments.. The MSP430 should do the trick as well. bye, Said Said, Thanks for the clue on the Philips part. Sounds like a nice solution. I just got a couple of sample parts from Maxim that I am going to play with (DS4000, MAX5121, and MAX6629). If they can do what Maxim claims it might be a fun experiment. I have one of those little TI MSP430 dev kits that I am going to try to use to run the show. We'll see. Randy ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts ___ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
