Used to work with a real 'chump' who decided he needed an Accutron... same as the first one in the video[1]. It was Los Angeles mid 70's; he & his wife found some _especially_ slimy high brow jewelers who correctly pegged the guy as a hopeless moron. It was one of those ritzy titzy outfits with lots of statues, private viewing rooms, extended pinky fingers, fountains, complementary wine and centrally located tuxedo adorned pianist. The whole deal. They sold him the watch and fed him rations of BS about how only they, or other official Accutron dealers could set or adjust the time on the watch. They got like $20USD to do it too! A lot of money back then. The dealer also didn't provide a manual, warranty or other documentation. For some years our hero was in for regular 'adjustments', and couldn't have been happier! I remember asking him what the 'bail' thing on the case back was, and he replied 'the dealer said to never mess with that'! :) The dealer also sold him numerous megabuck cleanings, battery swaps & who knows what else; I remember there were a bunch of them & each seemingly took weeks... the thing cost him fortunes! It was an amazingly long time before he wised up... said moron claimed there were always other suckers cue'd up at the dealer for the same BS. [1] Seems I remember hearing that those old Accutrons used no longer available Mercury batteries... could be wrong.
Excellent presentation! Clear, concise, methodical, step-by-step, easy to understand explanation of the subject. You should be a professor...thank you for the video!
Enjoyable (long!) video. One thing I didn't see mentioned, although I might have missed, is GPS leap seconds. The GPS time is currently offset from UTC by 18s (IIRC).
Hi Martin, great video; keep up the good work! I was in the timekeeping business back in RSA many years ago and we used to synchronise our lab clocks using an analogue TV receiver and a pickup coil to extract the horizontal sweep frequency which the SABC kept in sync with an atomic clock maintained by the CSIR. Also look up "John Harrison" and "longitude" that's a fascinating story about timekeeping which had a profound effect (a giant leap, actually) in accelerating maritime navigation and hence trade and commerce in the 18th century.
Thanks Martin! I love that shirt. That is so weird - I was just thinking about my Accutron's last night before I went to sleep. I'll get back onto my 2182's within the next few months.
Thanks for another wonderful video! I did something similar and bought a used Spectracom 9283 for my lab. I did this because I wanted low phase noise for my RF signal generators. However for folks that can't / don't want to spend that kind of money should take a look at Leo Bodnar's precision GPS reference clock. It's only uses a TCXO but after playing around with mine for a few weeks, I don't think you can beat it for the money. It provides a 3 volt CMOS output and I have yet to find a piece of test equipment that won't work with it. For a distribution amplifier look for a used Stanford Research FS710 or an FS730. The FS730 will cost more but it provides much lower phase noise. These things are great too and will provide a sine wave output regardless of the input signal type.
I've just placed an order for the Leo Bodnar's mini precision GPS reference clock. It will be interesting to compare the performance against the cheaper eBay device I have. Reading the specs it seems to cover the cons I show in part 2. When GPS signal is lost, the cheaper device drifts within a very short period of time. The Leo Bodnar device seems to be more intelligent and maintain the reference frequency like the Symmetricom TrueTime XL-DC.
I see the RB modules on ebay though I'm wondering if there's a way to guess the quality before buying? I would hate to spend $500(USD) only to find it's junk.
great tutorial ! temperature compensated seaworthy pendulum clocks where invented 1762 (the "H4") by genius John Harrison. This story is told in the book "Longitude" by Dava Sobel.
If you are interested in learning more about the story behind connections between trading firms and the NYSE, which Martin references in this excellent video, I highly recommend Michael Lewis's book "Flash Boys."
Я не мог догадаться, поэтому пришлось перевести английский текст, чтобы понять в чём же дело. Кварцевый генератора синхронизируется по сверхточным спутниковым часам или спутниковыми сигналам от GPS!
Very bad video, containing mostly smattering, and also wrong information in several parts. Only a few examples: @14:45: high time resolution (ns, ps, fs) in scopes has nothing to do with high accuracy, and does not need highly precise reference oscillators 19:20 - 20:50: confusing arguments when using different oscillator in ATEs. What IS the possible problem, in fact? Same as before; in most used cases, a highly stable / accurate time base is NOT needed. 46:30 1Mhz signals gives more digits resolution than 10MHz? What? If using reciprocal counters, the resolution is mostly independent from the input frequency, depends only on the Gate Time, the internal reference frequency (the higher, the better), and the time interpolation technique used. Very probably you confuse that with the digital rounding error for e.g. 9 digits resolution, when the result is displayed as either 9.999 999 99 or 10.000 000 0 MHz. That display depends on the ratio between the reference and the input frequency, i.e. if it's a bit higher or a bit lower than 1 (for 10MHz) or 10 (for 1MHz). Highly resolved and accurate frequency comparisons should not be done by direct frequency measurements anyhow. Better use T.I. methods and statistics.
Used to work with a real 'chump' who decided he needed an Accutron... same as the first one in the video[1]. It was Los Angeles mid 70's; he & his wife found some _especially_ slimy high brow jewelers who correctly pegged the guy as a hopeless moron. It was one of those ritzy titzy outfits with lots of statues, private viewing rooms, extended pinky fingers, fountains, complementary wine and centrally located tuxedo adorned pianist. The whole deal. They sold him the watch and fed him rations of BS about how only they, or other official Accutron dealers could set or adjust the time on the watch. They got like $20USD to do it too! A lot of money back then. The dealer also didn't provide a manual, warranty or other documentation. For some years our hero was in for regular 'adjustments', and couldn't have been happier! I remember asking him what the 'bail' thing on the case back was, and he replied 'the dealer said to never mess with that'! :) The dealer also sold him numerous megabuck cleanings, battery swaps & who knows what else; I remember there were a bunch of them & each seemingly took weeks... the thing cost him fortunes! It was an amazingly long time before he wised up... said moron claimed there were always other suckers cue'd up at the dealer for the same BS.
[1] Seems I remember hearing that those old Accutrons used no longer available Mercury batteries... could be wrong.
Excellent presentation! Clear, concise, methodical, step-by-step, easy to understand explanation of the subject. You should be a professor...thank you for the video!
Enjoyable (long!) video.
One thing I didn't see mentioned, although I might have missed, is GPS leap seconds. The GPS time is currently offset from UTC by 18s (IIRC).
Hi Martin, great video; keep up the good work! I was in the timekeeping business back in RSA many years ago and we used to synchronise our lab clocks using an analogue TV receiver and a pickup coil to extract the horizontal sweep frequency which the SABC kept in sync with an atomic clock maintained by the CSIR. Also look up "John Harrison" and "longitude" that's a fascinating story about timekeeping which had a profound effect (a giant leap, actually) in accelerating maritime navigation and hence trade and commerce in the 18th century.
Thanks very much for your post, very fascinating.
Interesting - thx for that - reminds me a bit of my old machine shop teacher's favourite saying "You need a lathe to make a lathe"..
Great video, packed with very interesting info. When someone really knows what he is talking about, it becomes fun to watch.
Very happy to see you back Martin. Interesting!
It's a long time to see you again , Happy to see you.
Thanks Martin! I love that shirt. That is so weird - I was just thinking about my Accutron's last night before I went to sleep. I'll get back onto my 2182's within the next few months.
Thanks Robert.
Very interesting topic, great video!
Thanks for another wonderful video! I did something similar and bought a used Spectracom 9283 for my lab. I did this because I wanted low phase noise for my RF signal generators. However for folks that can't / don't want to spend that kind of money should take a look at Leo Bodnar's precision GPS reference clock. It's only uses a TCXO but after playing around with mine for a few weeks, I don't think you can beat it for the money. It provides a 3 volt CMOS output and I have yet to find a piece of test equipment that won't work with it. For a distribution amplifier look for a used Stanford Research FS710 or an FS730. The FS730 will cost more but it provides much lower phase noise. These things are great too and will provide a sine wave output regardless of the input signal type.
Thanks for the post and handy info Roy.
I've just placed an order for the Leo Bodnar's mini precision GPS reference clock. It will be interesting to compare the performance against the cheaper eBay device I have. Reading the specs it seems to cover the cons I show in part 2. When GPS signal is lost, the cheaper device drifts within a very short period of time. The Leo Bodnar device seems to be more intelligent and maintain the reference frequency like the Symmetricom TrueTime XL-DC.
I see the RB modules on ebay though I'm wondering if there's a way to guess the quality before buying? I would hate to spend $500(USD) only to find it's junk.
A Rubidium atomic clock isn't that expensive, today they are 150~200 dollar. I bought 5 when the prices where low (40$/pc), best investment ever.
If only all educators could be this good.
great tutorial !
temperature compensated seaworthy pendulum clocks where invented 1762 (the "H4") by genius John Harrison.
This story is told in the book "Longitude" by Dava Sobel.
Can't wait for Part 2
If you are interested in learning more about the story behind connections between trading firms and the NYSE, which Martin references in this excellent video, I highly recommend Michael Lewis's book "Flash Boys."
Я не мог догадаться, поэтому пришлось перевести английский текст, чтобы понять в чём же дело. Кварцевый генератора синхронизируется по сверхточным спутниковым часам или спутниковыми сигналам от GPS!
what if you tune the wrist watch with a tuning fork? hehe
Useful video
you should do some live chats and you can make some donations from the super chat
very good
you need a roomier bench
I know...it's a little tight! ;-) I should have a bigger space by the end of the year.
mine is made of an 8' by 4' sheet of 3/4" ply, standing on legs of 3" by 2", cheap and easy to make, but you need a good-sized room to put it in.
Bulova is properly pronounced: bull of a . (bull ov a) Ron W4BIN
#1
subtitles, plz
Very bad video, containing mostly smattering, and also wrong information in several parts.
Only a few examples:
@14:45: high time resolution (ns, ps, fs) in scopes has nothing to do with high accuracy, and does not need highly precise reference oscillators
19:20 - 20:50: confusing arguments when using different oscillator in ATEs. What IS the possible problem, in fact? Same as before; in most used cases, a highly stable / accurate time base is NOT needed.
46:30 1Mhz signals gives more digits resolution than 10MHz? What?
If using reciprocal counters, the resolution is mostly independent from the input frequency, depends only on the Gate Time, the internal reference frequency (the higher, the better), and the time interpolation technique used.
Very probably you confuse that with the digital rounding error for e.g. 9 digits resolution, when the result is displayed as either 9.999 999 99 or 10.000 000 0 MHz. That display depends on the ratio between the reference and the input frequency, i.e. if it's a bit higher or a bit lower than 1 (for 10MHz) or 10 (for 1MHz).
Highly resolved and accurate frequency comparisons should not be done by direct frequency measurements anyhow.
Better use T.I. methods and statistics.
*TUNED not "tunned". *DISCIPLINED not "disaplined"