Special Stage Forums banner

1 - 18 of 18 Posts

·
I am not here anymore
Joined
·
2,798 Posts
Discussion Starter #1
I am looking for a good reference clock to see how much clock drift I am getting on some other devices. Specifically, I am looking for something that keeps accurate time over a day or three to tenths of a second?

How well do atomic clocks keep time? Are they accurate through the day or do they have to correct themselves?

alan
 

·
Lost....We're not lost
Joined
·
419 Posts
Specifically, I am looking for something that keeps accurate time over a day or three to tenths of a second?

How well do atomic clocks keep time? Are they accurate through the day or do they have to correct themselves?

alan
The 2 atomic clocks I use during events seem to keep time just fine (both are made by Casio). I rarely have to adjust them to rally time even after being away for months.

I have another atomic clock @ home (Oregon Scientific) that has problems keeping the signal. Sometimes it re-sets itself to different time zones when I least expect it. It is a major pain to get it back to the original time zone. :confused:
 

·
The Scorpion King
Joined
·
678 Posts
How about a GPS receiver? There are some clocks that are built into NTP servers that have GPS inputs also, but these get pretty pricey.

--
John
 

·
Registered
Joined
·
1,998 Posts
My experience with atomic clocks is that you get what you pay for. The $20 ones are accurate for awhile, and some of them will remain accurate for days, but long-term accuracy varies widely between examples. My Casio wrist watch - which cost significantly more - keeps accurate time for days, even when it can't reset itself at night.

I have found that either satellite-based time (GPS) or cell phone time will sometimes vary from WWV...although I haven't enough consistent results to say that with any certainty.

Best bet is a radio receiver set to 5, 10 or 15 MHz. Scare up one of the old Radio Shack Timecubes if possible.

Bruce
 

·
Registered
Joined
·
297 Posts
GPS time is theoretically better than WWV since it compensates for those few nanoseconds of signal transit, however- what gets displayed as "time of day" on your GPS unit is a different matter altogether.

The "atomic" clocks that sync to WWV every so often do drift and aren't quite ideal, but they are a good measure better than pretty much anything else in their price range.

For my dime, and I have several $20-30 radio-controlled clocks, but to check them -internet based "atomic clocks" like the AWYSE app on the i-phone or NISTIM on the PC do the job adequately for my needs.
 

·
I am not here anymore
Joined
·
2,798 Posts
Discussion Starter #7
Hmmm. Bruce's comments seem like rationale for me to get that Casio Pathfinder that I have been thinking about getting for the last eight months ;)

The devices that I am testing are my various iPhone/iTouch/iPads (with no internet access). There was was a bit under a second drift over 24 hours between the device that I have tested (my iPhone) and my cheapie Armitron watch. There was about a second of drift between that watch and the official time at Oregon.

alan
 

·
Registered
Joined
·
332 Posts
My experience is that computing devices (computers, cell phones, PDAs) have pretty crummy accuracy unless they're synced to something. I've seen them lose or gain seconds per day. Cell phones are normally time synced by the tower but that relies entirely on whether the tower's clock is correct. Some providers are much better than others. My Sprint phone was always dead on but my AT&T phone is often off by two or three minutes.

GPS time is derived from an atomic clock and should be pretty accurate; it's often used as a time reference for NTP primary servers. However, displaying the time can introduce errors. My old Garmin hand-held GPS seems pretty consistent, but my Navigon car GPS has a "hiccup" once every 10 seconds where the display gets delayed briefly, probably due to other stuff going on with the CPU.

The radio-based "atomic clocks" are using WWVB's signal, broadcast on 60 kHz in Ft. Collins, Colorado. The signal itself is quite accurate; the problem is the clocks usually only sync once or twice per day, so in between they can drift pretty significantly.

Probably the best way to get a time you know is accurate is to listen directly to WWV on one of its shortwave frequencies, like Bruce suggested. There's also CHU on 3.330, 7.850, and 14.670 MHz.
 

·
straight at T
Joined
·
2,472 Posts
My experience is that computing devices (computers, cell phones, PDAs) have pretty crummy accuracy unless they're synced to something. I've seen them lose or gain seconds per day. Cell phones are normally time synced by the tower but that relies entirely on whether the tower's clock is correct. Some providers are much better than others. My Sprint phone was always dead on but my AT&T phone is often off by two or three minutes.
Or you could be at Baie where the cell phones pick up the towers on the other side of the bay, in a different time zone!!! New Brunswick is Atlantic time, Quebec is Eastern.

It is also one of the few places I've seen where you can have five bars of signal and still not be able to make a call (towers are line-of-sight across the water so their signal is good but far enough away that the signal from the phone isn't stong enough).

Adrian
 

·
I am not here anymore
Joined
·
2,798 Posts
Discussion Starter #11
My experience is that computing devices (computers, cell phones, PDAs) have pretty crummy accuracy unless they're synced to something. I've seen them lose or gain seconds per day. Cell phones are normally time synced by the tower but that relies entirely on whether the tower's clock is correct. Some providers are much better than others. My Sprint phone was always dead on but my AT&T phone is often off by two or three minutes.
When I was working in CT in 2003 and would drive around New England sightseeing on weekends, I had to turn off the feature for my cell phone to resync time when a different time was detected. There were places (NH, ME) where it was resyncing every few miles. The time differences would be a couple of minutes. I haven't trusted cell phone time since.

I used to work on the time and time zone code in the Solaris operating system. I thought it was strange when we would get new feature requests to add leap second support. I would expect the class of users wanting leap second support would understand that the native time support in the operating system drifts enough to completely obscure adding a leap second.

I still haven't gotten a good reference clock to do complete testing, but my iPod touch is drifting less over a day than my cheapie Armitron watch and Robic stopwatch.

alan
 

·
pressing on tirelessly
Joined
·
2,225 Posts
It is also one of the few places I've seen where you can have five bars of signal and still not be able to make a call (towers are line-of-sight across the water so their signal is good but far enough away that the signal from the phone isn't stong enough).
It actually might not be a technical problem. We have a couple of spots at NNR where you can receive calls but not call out. There are some others where you can do voice but not data. I'm not sure of the details but I think it has to do with some inter-operator "roaming" contracts.
 

·
straight at T
Joined
·
2,472 Posts
It actually might not be a technical problem. We have a couple of spots at NNR where you can receive calls but not call out. There are some others where you can do voice but not data. I'm not sure of the details but I think it has to do with some inter-operator "roaming" contracts.
I'm certain that in this case it was that the low power of the phone wasn't enough to trip the cell tower at 15+ km. The phone in question was on its home network, not roaming. If you got a call, the phone would ring and then drop the call the moment you picked up.

Adrian
 

·
Registered
Joined
·
332 Posts
For many years people in some rural parts of the U.P. kept their "bag phones" even as everyone else moved to hand-helds, because the higher transmit power of the "bag phones" gave them better coverage.
 

·
Spectator Wrangler
Joined
·
830 Posts
For many years people in some rural parts of the U.P. kept their "bag phones" even as everyone else moved to hand-helds, because the higher transmit power of the "bag phones" gave them better coverage.
When Alltel (now Verizon) discontinued analogue service, "old" bag phones were dead. Marquette County ( largest in da UP) had to purchase some digital bag phones to get coverage throughout the county. 3 watts in a bag phone vs. 300mw in a handheld. Try to get a handheld phone with a real antenna or a plug for an external one.
 

·
NASA Rally Sport grassroots!!!
Joined
·
2,835 Posts
I think it would be informative to discuss the actual need for an accurate clock and the real effects of an clock involved in the rally (competitor or volunteer) being grossly sloppy. (Say, loosing 10 seconds a day)

Let's put it into three categories:
Great time: a second or less per day off
Good time: less than 4 seconds a day off
Meh time: more than 4 seconds a day off

I submit that there would be no difference in any of the final placings as a result of having either a good or great clock anywhere, without anyone doing anything.

With the interaction of reasonable stewards, I think the same is true of meh time.

Anders
 

·
Registered
Joined
·
1,998 Posts
Hmmm. Bruce's comments seem like rationale for me to get that Casio Pathfinder that I have been thinking about getting for the last eight months ;)
Another advantage of the Casio is that not only does it synch itself automatically once a day, but it can be made to synch any time manually. Takes about two minutes. I sometimes have to do this when it can't get WWV inside a motel room - like in the PNW.

Bruce
 
1 - 18 of 18 Posts
Top