Digital does not mean Accurate

[ceagle]

One of the weirdest quandaries of our technological day and age is that we have little clocks on just about everything... our computers, cel phones, vcrs, etc.... but, they STILL don't each keep the same time! ;P
They used to say if it was "quartz", then it would be, but most aren't. My pager runs fast, and my car clock runs slow, etc etc etc.... why can't we come up with a cheap, ubiquitous, non-quartz circuit that will keep accurate digital time? 8|

Comments

[kiswara.livejournal.com]

Digital time: [On] / Off.

[c-eagle.livejournal.com]

*eagle analogs your ears!*

[loveyouenola.livejournal.com]

My dad used to work at the Defence Science and Technology Organisation (Gov run research centre) here in Salisbury. He told the story of how they got a huge atomic clock delivered - worth millions - I don't know what was wrong with it, but after they'd finished working on it, the technician left it running overnight.

The next morning, dad was sitting reading the paper before work started and the techno came in, looked at the readout on the clock, checked his wristwatch and then rang the couriers "Yep, all fixed, ready to go!"

:D

[c-eagle.livejournal.com]

... and thenceforth he established the basetime for modern society! :D

[androgynism.livejournal.com]

I read somewhere that quartz clocks lose time if they're exposed to large temperature changes. They usually lose about 10 seconds per year.

[c-eagle.livejournal.com]

oooo... I s'pose that is plausible :9
Had not herded that though!

[androgynism.livejournal.com]

Just the heat from your body is enough to make quartz wrist watches lose twice as much time, because you take it off and put it on.

[c-eagle.livejournal.com]

I'da woulda thotted theyda compensated fo dat though... *pips, and scritches hayd*..

[orv.livejournal.com]

Yeah, they're affected by temperature, in both directions. For really precision applications quartz crystals are put in little temperature-controlled "ovens."

The accuracy also depends on how accurately the quartz crystal was cut to begin with. If the original "grind" was off the crystal may be stable, but stable on the wrong frequency. The more carefully the crystal is ground, the more expensive it is, so the crystals in cheap mass-produced items are likely to vary a lot.

Devices that plug into the wall are usually referenced to the power line instead. This can actually be better on a long-term basis than a quartz crystal, but will be less accurate over the short term (hours to days).

Increasingly, devices that have radios or network connectivity, like cell phones, are using clocks referenced to some central source, which in turn is probably referenced to an atomic clock somewhere -- either via NTP or GPS. This can give exceptional accuracy if it's done right.

[scruff-e-coyote.livejournal.com]

We do have circuits that keep accurate time - radios attached to devices (such as home weather instruments) that tune via short wave into the U.S. Department of Commerce's National Institute of Standards and Technology atomic clock in Fort Collins Colorado.

[julian-wilbury.livejournal.com]

Is that the same thing as the Atomic Navy Clock? That's my vice of choice.

I'm oldskool and wear a standard wristwatch. But the one I'm on right now slows down like woah. :( Which makes me sad, because it's a really, really cool Lion King watch.

(Hi coyotes are awesome.)

and he's a cool coyote!

[c-eagle.livejournal.com]

*warble* ... I don' even wear a wristwatch enny mo' .... time is on my pager, which I wear nearly alla time. And also on my cel phone. And computa skreeeeeen.
But I do have a shortwave radio which can tune into that nostalgic tick tock tick tock too!

[hastka.livejournal.com]

I'm about 99.44% certain that they have wristwatches now that sync with WWV et al. But certainly yes, cell phone time had darn well better be kept pretty accurate, as they certainly have the infrastructure to make people's lives better there.

Personally, I always preferred the mechanical workings of self-winding watches, even if it doesn't keep time quite so accurately. Which makes me wonder why nobody's come up with electronic watch movement with 1/10 second sweep, or something. Although my last couple of watches have been largely clear, heh.

[c-eagle.livejournal.com]

wellllll... heh!... I admit that I know of nothing that comes closer to accuracy in the time department as that... but I'm talking about a 'self-contained' circuit/system... :9

[artfulreggie.livejournal.com]

Vote for Reggie in 08. A chicken in every pot and some cesium-133 in every watch! ;)

I think the key here is to just enable clocks to sync with a more accurate and official clock via a mesh network. ;)

[tenaya-owlcat.livejournal.com]

I swear, the time at my house, the time the busses run by, the time by the clock chimes on campus, and the time in my office are all different! O.o

[aldi.livejournal.com]

The only devices in our home that keep time accurately are my two Sony Hi-Fi Beta SL-HF860D. No surprise: back in 1988, they cost me $950 each! Sony now makes plastic caca. No, worse: they hire Samsung to make their caca for them.

[rcoony.livejournal.com]

Generally I go by the clock on my satellite receiver, and set all my clocks slightly ahead of that, so I don't miss any shows. Otherwise, I just set all my clocks slightly ahead of the most "official" time I can find.

I like nice old mechanical-type clocks, myself. They are just more fun. I have a wind-up wristwatch, and if I could get all pendulum clocks powered by weights (none of these fake grandfather clocks powered by electricity and with a pendulum only for show), I'd do that. Well, I'd get a couple of normal clocks too so I'd be able to set all the mechanical ones to the proper time.

I'm thinking of getting a couple of 24-hour clocks so that I can easily remember what time it is in places like Australia and whatnot when I'm chatting online. Just the other night I came across the perfect one. Costs a little more than I wanted to spend, but it's too awesome not to get.

[jedd-marten.livejournal.com]

Just manufacturing shortcuts. A properly designed and calibrated quartz oscillator should drift only a few seconds per year.

[hastka.livejournal.com]

On a related peeve: I'm getting pretty darn irritated with people thinking digital video is somehow superior. I have digital cable, and the MPEG artifacts evident in almost EVERY frigging show that involves motion just bug the everliving heck out of me.

[kaysho.livejournal.com]

Because digital has increased our sense of precision, and therefore our standards for what we consider "accurate". A clock that drifted a few minutes a week that you would periodically resynchronise with the "time and temperature" telephone recording used to be nice and accurate. After all, the minute hand had to be moved only a fraction of an inch!

Now we perceive a huge inaccuracy when our clock says it's 2:50:28 and the right time is 2:51:07. :)

Besides, you're familiar with the "faster better cheaper" triangle. An oscillator that's cheap enough to put into a pager or the like without making it more expensive than people will want to buy (for what's honestly an ancillary function of the device, anyway) isn't going to be better. :)

[cabcat.livejournal.com]

Mine all seem to run fine, just my old Seiko analogue clock runs fast because the adjustment is off.