On Monday, 7 March 2022 16:08:59 EST Chris Albertson wrote: > On Mon, Mar 7, 2022 at 12:04 PM gene heskett <[email protected]> wrote: > > Technically, we can do it but you'll need a couple of dump trucks > > full of gold to finance one neasurement. We can't yet buy a > > calibrated answer in 10 milliseconds for a $500 bill. Someday? > > Maybe, but it may take a new method to be invented. > > I have a system that does FAR better. I tried to build a clock that > would give accurate absolute time measurements with the goal of it > being inexpensive. It is easy to do this if you have boatloads of > money but what if you limit the budget to about $50? Speaking as a long retired broadcast engineer, I spent the last 18 years of my working like as the CE at WDTV, originally licensed to Weston WV. The transmitter was an elderly GE made in the middle '50's of the last century, and had glass ovens with AT cut crystals for frequency. But after nearly 35 years the heater contacts gradualy failed, so the rock got cold and we went out of tolerance a long ways, which for a tv on channel 5 was +- 1k, with the aural speced at +- 1k of 4.500000 mhz above the visual. GE had no rocks that were any better, they had been gathering dust on the shelf for 40 years.
And here tells the story. I called up Vectron who made stable frequency sources, and they made me a pair of little 2x2x1 boxes. I got them in, put them in pcb boxes with a power supply that ran off the old 12 vac heater power, putting an 8 pin connector on the bottom of the box so it plugged directly into the crystals socket. Plugged them in, trimmed the frequency to within 5hz, expecting to have to chase them as they aged in, but I didn't have to, they simply did not age. You have to have a type approved frequency monitor, and the first thing I did when it went way off, was check the counter against WWV, then found the transmitter was still within 5hz. Sent the monitor back to Arlo Belair who fixed it. I sent that monitor back to the maker 3 more times in the next 15 years, but never again touched the trimmers on those Vectron src's. It was still within 5hz when it was turned off forever at midnight, June 30th 2008 to sign on with digital few seconds later. That level of precision cost us $750 a copy in 1988. > So with a $50 budget, I can measure seconds of time, referenced to a > world standard to 10 decimal places. 13 is possible if the budget > were larger. > > How? First, you need a really good crystal oscillator. On this > budget you have to buy them used on eBay. These will typically be > double ovenized and inside some insulated temperature controled and > sealed. box. They are removed from microwave telecommunications > equipment or laboratory grade frequency meters. These were made by > companies like HP but are now obsolete. These crystals will have a > fine frequency adjustment on them that was intended to be set by a > technician using a laboratory calibration standard. > > Here is the key, Connect the oscillator output to an analog pin of a > microcontroller, like an Arduino. The computer counts the cycles and > with an analog pin, can measure phase. The computer also is > connected to the 1Hz output of a GPS. GPS output is VERY good in the > long term. It is very stable over months and years but each pulse > might have up to 50 nanoseconds jitter (for more money you get 5ns > jitter) > > The software measures the phase of the crystal and the GPS and makes > very tiny adjustments to the crystal to keep it phase-locked to GPS. > We can easily measure phase to 1,024 parts per cycle. After a few > days, the system is locked. > > You might ask how I know I have 10 digits precision, don't ask. It > means you need several better clocks and a way to cross-reference > them. I bought a couple rubidium clocks as even these are now > obsolete for cell tower frequency control, and you can buy them from > Chinese companies that salvage old cell towers for $100 per unit. You > slave there to a better GPS and then filter with a Cystal > > General Relativty predicts that tiime runs slower if gravity is > stronger. The best clocks now can measure this effect if you move them > from a desk to a tall shelf because gravity is weakers on the shelf > then the desk. This is the curent state of the art for portible > clocks that can be lifted by one person. Hint: We can locate objects > from gravitational > attraction's effect on time. Massive objects are getting harder to > hide. because mass warps space-time. SO, we can do 13-digits on > a high-end hobby budget -- government labs do better. > > > > Does anyone here know of some brands/ models to check out? > > > > > > Google has pointed me at some Acuity products but I wondered if > > > there > > > were other options besides that. > > > > > > _______________________________________________ > > > Emc-users mailing list > > > [email protected] > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > Cheers, Gene Heskett. > > -- > > > > "There are four boxes to be used in defense of liberty: > > soap, ballot, jury, and ammo. Please use in that order." > > > > -Ed Howdershelt (Author, 1940) > > If we desire respect for the law, we must first make the law > > respectable.> > > - Louis D. Brandeis > > > > _______________________________________________ > > Emc-users mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/emc-users > > -- > > Chris Albertson > Redondo Beach, California > > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users > . Cheers, Gene Heskett. -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author, 1940) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
