Nothing's ever simple

[unixronin]

So, I exchanged yesterday's vision-protection-required blue-LED Walmart-house-brand alarm clock for a Westclox model with a red LED display.  That, one would hope, would be the end of alarm-clock issues.  Alas, it is not to be.

To start with, the instruction sheet says, "To set the alarm time, press and hold the ALARM SET button and..."

Uhhhh..... pardon me if I'm missing something, but ... this clock doesn't appear to have an 'ALARM SET' button.  There are five visible controls on the clock: an alarm on/off switch, and four buttons labelled TIME, HOUR, MINUTE, and ALARM zZz.  (This latter button is described on the instruction sheet separately from the ALARM SET button.)  No visible ALARM SET button exists.  How on earth is the alarm set on this clock with no ALARM SET button?

(My only speculation is that the alarm snooze and alarm set buttons are, in fact, one and the same button, and the instruction sheet is just badly written.  I'll test this theory in a few minutes.)

Then, a little further down the instruction sheet, it talks about battery backup.  About two thirds of the way down this section, the instruction sheet says, "If you have a long power failure, the clock uses an internal clock which is less accurate than line frequency."  (Emphasis mine.)

My GOD ...... it's 2005, people.  Quartz oscillators cost pennies.  You can buy PENS with built-in quartz clocks accurate to a few seconds a year.  You can buy alarm clocks for only a few dollars more than this one that synchronize themselves by radio to the NIST atomic clock, and are accurate to thousandths of a second per year.  And Westclox is relying on LINE FREQUENCY for the time source for this alarm clock?!?  Line frequency isn't even guaranteed to be fixed at 60Hz!  It's constantly fluctuating as load on the electrical power grid changes -- 60Hz is just the average!  This is why old-style line-synchronous electromechanical clocks were accurate to, at best, a few minutes per week.

Hastur on a flaming diesel-powered pogo stick, Westclox, catch up to at least the 1990s, even if you can't join the 21st Century along with the rest of us!

---

Update:

It does, after experiment, prove to be the case that the 'ALARM SET' and 'ALARM zZz' buttons clearly described on the instruction sheet as separate and distinct controls labelled as just noted are, in fact, one and the same button, labelled 'ALARM zZz'.

Comments

[juuro.livejournal.com]

However, the integral of the difference from 60 Hz at least used to be guaranteed to converge to zero. That is, the 60 Hz as an average is really an average so that if it was a bit fast earlier, it will be a bit slow later to compensate for that in the long run. Such guarantees cannot be given for stand-alone oscillators.

Are those cheapies, the pen clocks and such, really accurate to a few seconds a year? It has been long since I last looked into them, and then they were minutes a month.

[unixronin.livejournal.com]

Well, I could be mistaken, not having researched it in detail -- but I was under the impression commodity quartz "movements" were now typically on the order of 15-30 seconds a year, while good ones were closer to 3-5 seconds per year.

Your point about the average of 60Hz line frequency is taken, but ... it just seems like a bad idea to me to be using as a time reference something that was never designed for that purpose, and is nowhere guaranteed to be accurate for that purpose, when better alternatives exist.

Frankly, I'd have bought one of the NIST-synchronizing clocks, if they'd had one with a red LED display rather than LCD or electroluminescent. As noted, they're only a couple of bucks more. The travel alarm I already have is NIST-synchronized -- but it has an LCD display with an EL backlight.

[tww1fa]

Quartz movements can be very precise - if kept in carefully controlled temperatures. However, they can fluctuate wildly depending on temperature and the better movements require temperature compensation. Even so, Swiss chronometer certification for a high end quartz watch is +/- 0.2 seconds/day - which means that you could lose/gain up to 6 seconds per month. Few quartz movements are certified to that precision, and even fewer are properly calibrated even for fixed temperature.
In the long run, your power-grid based clock will probably be more precise than a quartz based one of equal cost.

[unixronin.livejournal.com]

Those monthly numbers are a little higher than I've seen clamied, but .... well, I know how many lies marketroids tell about cars and computers. Why should watch mechanisms be any different?

Still, I guess the power grid must have gotten a LOT more tightly controlled than I'm used to it ever having been (any time that I actually knew the tolerances), just to be even competitive witha quartz movement. Go figure.

And as noted, if there'd been one available with a red LED display, I'd have bought a NIST-self-synchronizer anyway.

[technoshaman.livejournal.com]

Actually, the specs on line power specify that the 60Hz has to be quite accurate. The voltage does fluctuate quite a bit, but the Hz... no.

That must be a petrodiesel pogo stick; biodiesel wouldn't flame up like that outside the engine. Smelly.

[unixronin.livejournal.com]

The voltage does fluctuate quite a bit, but the Hz... no.

Well, it doesn't fluctuate by much, and not as much as it used to ... but even a 0.5Hz drop for four hours of heavy load puts your clock two minutes slow. And last I knew, it was permitted to go as low as 57Hz when the generating system is under heavy load. Has that changed?

That must be a petrodiesel pogo stick; biodiesel wouldn't flame up like that outside the engine. Smelly.

Didn't you know? The flame adds an extra 2D6 damage. Hastur is one bad-ass mofo', don't fuck with him when he's got his pogo-stick. :)

[technoshaman.livejournal.com]

Yeah, yeah. Just wait'll the bureaucrats at the EPA get to him.

[olafthunderfoot.livejournal.com]

~pelts you with d30s~

[ilcylic.livejournal.com]

Hastur on a flaming diesel-powered pogo stick

HAAHAHAHAHAHAHA!

That's hilarious.

-Ogre

[jilara.livejournal.com]

I bet I know what happened to the instructions. They switched the design of the clock, due to availability of components, after the instructions (based on the spec) had already been written and sent to the printer. (Gee, you'd think I used to write hardware manuals or something.)

[unixronin.livejournal.com]

Who, you? Naaaaaaaaaaah....... :)

[radarrider.livejournal.com]

One possibility is that it doesn't refer to the frequency of the alternating current, but the time signal that is sent along the power lines in at least some areas. My bedside alarm clock, for example, sets itself to that time signal as soon as I plug it in.

[unixronin.livejournal.com]

Nope, it's not an auto-setting one. I'd have bought a self-setting one if they'd had one with a red LED display instead of LCD or EL.