FY 2019 budget requests shutdown of NIST time stations

The GPS 1 pps pulse has tons of jitter. It's long-term stable, so it is used to discipline a quartz or rubidium oscillator. The slower the loop and the better the oscillator, the better the result.

I suspect that the loop is a time zoom edge comparison, software driven, not some sort of classic PLL. You can do a lot of computing in a second.

Does anyone make wi-fi based clocks? That's pretty obvious.

googles...

Yes.

That doesn't explain why the rate is 8 to 14 times higher than the USA. However, since the German study was done using patient surveys with only 190,000 reports out of 18.6 million patient (visits?): 190,000 / 18.6 million = 1% instead of using hospital release records, and probably extrapolated by a factor of 100x, I would not expect much accuracy in the German figures.

Between 2002 and about 2009, I survived a series of expensive medical problems that involved the services of the medical profession. I rarely caught the doctor making a mistake. At best, it was a failure to do a sanity check on the diagnosis or test results. However, I experienced a substantial number of mistakes on the part of hospital staff, office staff, pharmacy errors, lab tests, radiology evaluations, and of course, billing. I expected the overworked doctors to be the prime cause of medical errors, but it appears that the problem is elsewhere.

Incidentally, since 2002, I have had to switch various doctors several times. It wasn't because I was dissatisfied with their work, but rather because they chose to abandon private practice and become parts of a medical group. Three of my doctors all went to PAMF: When I asked, it was primarily to eliminate managing their office, doing their own billing, but most important, protecting them against malpractice suits by utilizing the much larger resources of the PAMF legal staff.

Most of my old smartphones are spending their retirement days in offices and homes acting as media players. When not playing music, they display the time, which is synchronized to us.pool.ntp.org via wi-fi.

Most clocks that need to be synchronized to a central server, such as school clocks, user either wi-fi or power line networking to an NTP or SNTP server: etc...

Here's some that use BlueGoof or GPS -> 72MHz broadcasts:

GPS clocks are more interesting and cool: Might as well repurpose that old oscilloscope.

Arsenic was used starting in 1909; see Salvarsan.

The oldest person in the UK has just turned 112. When she was born the female life expectancy was 54 (men 48).

OTOH, my grandmother lived to 96 and her mother, my great grandmother (who was a significant influence on my mother), also lived to 96. I ought to do some genealogy, but she was probably born in the late 30s. 1830s that is.

I suspected that. Can anyone describe to be the actual signal chain used to derive an external timing pulse or frequency from the GPS signal? As in, how many layers of conversion that result in jitter or phase noise?

I'd like to apply a cheap GPS module to generate a 10MHz reference, but I don't want to simply believe the mfr's specs, but understand they hoops they jumped through to get there.

Clifford Heath

Iimpossible to transmit with any sensible bandwidth and transmitter hardware, which is why things like WWVB work out so well.

per second source.

The processing chain is not the source of the jitter per se. The GPS measu rement in typical devices is my nature periodic. The signals are received and the delays from all the satellites are then known which allows the loca tion of the user to be calculated which feeds back into the time calculatio n. This is typically repeated once per second. The error of any such calc ulation is based on a variety of factors which are not known exactly and th erefore can not be corrected. As those factors change the location and tim e will jitter around.

The carrier is typically not observable because the signal is well into the noise. Without much more expensive techniques the jitter in the pps signa l has to be averaged out by filters after the GPS receiver. In rather more expensive units I'm told they directly observe the carrier and obtain a hi ghly accurate time signal without all the jitter. Distance can be resolved to the cm level and so time is also resolved to much smaller resolution.

Rick C.

] per second source.

surement in typical devices is my nature periodic. The signals are receive d and the delays from all the satellites are then known which allows the lo cation of the user to be calculated which feeds back into the time calculat ion. This is typically repeated once per second. The error of any such ca lculation is based on a variety of factors which are not known exactly and therefore can not be corrected. As those factors change the location and t ime will jitter around.

he noise. Without much more expensive techniques the jitter in the pps sig nal has to be averaged out by filters after the GPS receiver. In rather mo re expensive units I'm told they directly observe the carrier and obtain a highly accurate time signal without all the jitter. Distance can be resolv ed to the cm level and so time is also resolved to much smaller resolution.

Ahh, Thanks Rick, Lasse, mystery explained.

I thought that GPS required carrier phase measurement, but only in some was a signal (more-or-less) directly derived from it made available.

Anyone have a pointer to cheap carrier-phase GPS modules?

Clifford Heath.

ail.com:

e:

s] per second source.

measurement in typical devices is my nature periodic. The signals are rece ived and the delays from all the satellites are then known which allows the location of the user to be calculated which feeds back into the time calcu lation. This is typically repeated once per second. The error of any such calculation is based on a variety of factors which are not known exactly a nd therefore can not be corrected. As those factors change the location an d time will jitter around.

o the noise. Without much more expensive techniques the jitter in the pps signal has to be averaged out by filters after the GPS receiver. In rather more expensive units I'm told they directly observe the carrier and obtain a highly accurate time signal without all the jitter. Distance can be res olved to the cm level and so time is also resolved to much smaller resoluti on.

I don't think "cheap" and "carrier phase" go together. The first thing is to be able to actually *see* the carrier. The GPS devices in low cost gear don't actually see the carrier because it is in the noise. They essential ly use the code modulation as a filter to pull the signal out of the noise, but at the chip level (chip being one of many pseudo random bits making up a data bit) rather than at the carrier level. I want to say the carrier c an be something like -140 dBm. That's hard to pick up. I guess they can a pply similar techniques at the carrier level to pull it from the noise, but I haven't seen a description of this. The units I've seen that work at th e carrier level have much larger antennas mounted on a tripod. I expect th e antenna alone costs more than a low cost GPS receiver.

I want to try a WWVB receiver based on this averaging technique performed i n an FPGA with no analog front end and a 1 bit ADC. The samples would be v ery noisy, but with 60,000 cycles per bit a lot of averaging can pull the s ignal from pretty far down in the mud.

Rick C.

ood for your health' is just bull.

shape.

ns by Big Brother

tudent would.

eural net configuration.

eople,

day, the nonsense,

ors and faulty diagnosis,

ckly!

' them.

The average life expectancy was less than half what is is now, so we live m ore than twice as long, but most of the excess deaths happened in childhoo d, and if you lived long enough to have a mature understanding of the quest ion (which John Larkin clearly hasn't) you'd could expect to live to about two thirds of current life expectancy (if male), but slightly less than hal f (if female - childbirth was frequently fatal).

shows that while you were most likely to die at about 40 - if female - or 5

The risk of early death was highest when young - and they don't seem to hav e bothered documenting anybody who died younger than 20 - and declined with increasing age.

That's called a GPSDO or GPS diciplined oscillator. There are units, kits, and components on eBay: It differs from a rubidium oscillator in that a GPSDO usually uses a TCXO or VCTCXO in some manner of phase or frequency locked loop, while a rubidium standard generates its own reference, does not require a GPS receiver, but does have a loop of some sort to generate the 5 or

Most standards have 10MHz and 1pps outputs. Google for GPSDO and you should find plenty of information:

If you want to go the rubidium route, there are numerous secondary standards on eBay for about twice what a TCXO type of GPSDO would cost.

You may need some kind of distribution amplifier if you're driving a pile of test equipment external clock inputs: An external outdoor GPS antenna is also a good idea. Some of my test equipment takes a 1 MHz or a 5 MHz external clock input, so you may need to build a clock divider.

More:

A rubidium is not a formal primary standard, but a GPS disciplined source is. But a rub is good for a few PPB, good enough for most people.

It's surprisingly simple:

Yep. Rubidium oscillators will age and drift making them a secondary standard. I've read about combination GPSDO and Rubidium frequency standards which I assume somewhat solves the aging problem.

The Efratom FRS series of rubidium oscillators are now about 25 years old. Watch out for sick or dying rubidium lamps on eBay.

More on rubidium:

"Rubidium frequency standard Extreme Teardown"

"FE-5680A Rubidium Standard Teardown" (Dave Jones)

"Rubidium Frequency Standard FE-5680A" (Dave Jones) (22:38)

"Stability and Noise Performance of Various Rubidium Standards"

More on GPSDO:

"Notes on Trimble Thunderbolt performance and modifications"

Jeff Liebermann wrote

I would not let go of my old analog scope. Digtal sampling scopes lie, are evil ;-) Those work only part of the time, very short parts of the time in fact :-)

Most of that Giggle Hertz radio based stuff can be sone with a rtl-sdr stick, and some frequency converters I build.

Also I do not see teh value of that ScopeClock, way too many parts for a simple display without even video capability, 35$ plus shipping!

This is all ye need:

then get one of these with free shipping:

You need a real analog scope with Y input of course, not such a fake digital thing. And now you can use it as door video too.

Jeff Liebermann wrote

I payed _half_ the price of a GPS 10 MHz reference, 71$99 free shipping to be precise, on ebay for my super good working FE-5680A Rubidium Atomic Frequency Standard 10MHz in 2012.. Seems word got around an now they can ask what they want. Mine is not for sale ;-) mm let's see if it hits 1000$...?? And even then, why, it is a cool thing to have.

Actually JL was the one who drew my attention to that, and made me buy the ebay thing. Good investment , and very useful to calibrate things. So thank you JL!

I haven't used an analog scope in a decade or two. All the traces are green!

Our many digital scopes are light, run cool, and so far have been absolutely reliable. All the analog scopes eventually died.

The only kinda-analog scope that I use is my Tek 11802 sampler, good to 50 GHz. Only the CRT display is analog, and it's a magnetic-deflection raster scan. I'll miss it when it finally dies.

You can buy a new one from SRS for ~$1,500

George H.