Clock Losing Time....

A microwave oven cooks things, so it doesn't need to be a clock too. Unplug it, plug it back in, and then don't set the time.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

You'd have to take a peek inside to know for sure. It's relatively trivial to code a ToD clock that can *frequency* (not phase) lock to the AC mains and autodetect 50 vs 60 Hz. But, that doesn't mean the folks designing the controls actually *cared* enough to do that! You'd be amazed (scared!) at how cheap the controls are in those things :-/

I remember every electric clock on every range I ever saw had stopped working long before.

I don't believe 60Hz is reference grade. The thing alerady has an xtal; they should just count that down.

--
Les Cargill

For consumer stuff, "Westinghouse" is now just a sticker that gets put on whatever was cheap in China that week.

RC oscillator, ceramic resonator. You might be able to improve the situation a little by swapping parts, but it'll be a "try it and see" thing.

Some microcontrollers are available with internal 32 kHz oscillators that are sort of OK.

A lot of recent microwaves have the ability to turn off the time-of-day clock display altogether - the display stays blank unless you are cooking something, in which case it displays the normal countdown. Maybe do that and get a synchronous-AC-motor wall clock from the thrift store for $3. Or, get a nifty battery clock that sets itself to WWV, if there is good reception at that location.

Matt Roberds

Actually, over the course of a day (or so), it is *held* to be a reference. Short term accuracy (counts of 60 Hz cycles on the mains) can vary (and NEED to vary to accommodate load shapes) but these are adjusted periodically to ensure the total number of cycles (barring interruptions) is effectively

60.00Hz.

There was plans to ease this constraint "recently" (i.e., the last several years?) but I don't know what became of that.

I.e., if you have access to the line frequency (rectified AC off a transformer), it is relatively easy to implement a *frequency* locked loop and eliminate even those short term variations (that a *phase* locked loop would experience).

And, given even the crudest of accuracies on your timebase (stability is more important than accuracy) can *infer* whether your local LFC is running at 50Hz (also precisely controlled in places like Europe) or 60 Hz.

Of course, *your* clock may show a difference from what some COTS clock displays (esp anything that runs on a sync motor) but you'll know yours is "right"! :>

[And, of course, remember: even a broken clock is right twice a day! Which may be *more* than a WORKING clock!]

Thanks all, I think I had it figured out correctly.

I'll see if I can shut the clock display off, there's a perfectly fine ten dollar battery operated wall clock in the same room which only gets reset when DST changes.

If I can it'll stop my obsessive compulsive self from getting annoyed when I see the microwave's clock is "wrong".

Jeff

--
Jeffry Wisnia 
(W1BSV + Brass Rat '57 EE) 
The speed of light is 1.8*10^12 furlongs per fortnight.

It was implemented a couple of years ago. Since then clocks synched to the power line are less accurate. I have been kind of watching my clock radio and it gains time at a relatively slow rate. It takes about 3 months for it to gain 30 seconds, but that isn't a steady gain. It jinks around some from day to day, but overall it gains. So the power line frequency in my area is a little above 60 Hz. Different areas may have different frequencies.

Bill

If you want to be more "anal" (no insult intended) you can get one of the "atomic" wall clocks, that syncs with NIST.

Then you don't even have to worry about DST.

George H.

Buy GE next time.

Gummit mucks with all sorts of things. You'd think they'd have required all "new clocks (and devices that display the time)" to be "self setting"! I wonder what fraction of the GDP is wasted setting clocks? (the clocks in modern thermostats seem to be perpetually incorrect! I noticed this at a car dealership last week AND doctor's office. Why display the time if you're going to rely on someone to ensure it is correct?)

We have *one* such clock in the house. As we *don't* observe DST, here, it is wrong half of the year. I have to tell it we're living a few hundred miles further west (next timezone) to get the correct time. Obviously, it needed a "observes DST" switch and they opted not to spend the pennies for that! :-/

[Living WITHOUT DST is incredibly refreshing! However, it turns dealing with the rest of the country into a guessing game: "Are they on OUR TIME, now??"]

Well, they've always been less accurate /in the short term/ (there was a lot of error that was allowed "instantaneously")

By comparing with WWV/CHU?

Yikes! I'd have considered a second a week/month to be a lot!

Hmmm... I'll have to recheck the short (and long) term accuracies of the RTC's I've been using. As I've relied on the LFC for long term accuracy in my implementations (using the RTC just to carry me through those times when devices have been "powered off"), a new strategy may be to let the RTC be the final authority! :-/

Thanks for the update!

Can't find a way to turn clock display off (without unplugging the microwave). Nada on that in the instruction manual.

Time to focus on more important things now.

Jeff

--
Jeffry Wisnia 
(W1BSV + Brass Rat '57 EE) 
The speed of light is 1.8*10^12 furlongs per fortnight.

Lol. If you want to check the accuracy of a clock compare it to the time on your cell phone. It may not be as precise as WWV but a lot easier to access and apparently unlike the power line in the US, the long term accuracy is the same as the short term accuracy.

I used to think my hand held GPS had the best accuracy, but then I found this model has a bug where it can display the wrong time with an arbitrary offset and you wouldn't know it unless it was large enough to be obvious! This is clearly the result of poor consumer grade design with no mechanism to upgrade the firmware. A cell phone, in contrast, would not have a problem like this for long as so many people would complain about it that an upgrade would be made available.

RTCs from 32 kHz crystals gain and lose time depending on the effort exerted in design. Many, many years ago I had a watch with a trim cap inside. But due to the erratic temperature it was exposed to I could never correct it really. Now they just use a fixed capacitance and let it go. Even at constant temperature the drift will depend on the particular device with no trim available.

Other than cell phones the RCCs should be accurate, but most don't provide an indication of the current lock status so you never know. There is a lot of man made interference, so even if it is working this week, next week it can lose lock and start drifting.

--

Rick

I didn't state that quite properly.

Unplugging the microwave turns the clock display off, and turns the microwave into a small closet.

Plugging it back in starts the clock at 12:00 and if not set manually it just counts up from there each minute.

Jeff

--
Jeffry Wisnia 
(W1BSV + Brass Rat '57 EE) 
The speed of light is 1.8*10^12 furlongs per fortnight.

That depends to some extent. I have one radio controlled wall clock that automatically corrects for DST, except that it doesn't read the time change off of the radio signal. It is programmed according to the old dates. So it is off by an hour in between the time that the clock changes and the time that the time changes. I reset it by setting it one time zone off during that time. I don't know what people in the Eastern or Pacific Time Zones do.

Bill

FLLs aren't phase coherent--any offset voltage causes a frequency offset, which would be much worse than a PLL.

Frequency is just the time derivative of phase, so there's no magic in a FLL. Since the frequency has to vary in order for there to be phase excursions, a FLL can't have better stability than a PLL, and in fact is always worse due to offsets and so forth.

They used to be good for keeping your FM receiver tuned to the station, though.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

I have an oven like that--I just set the time wrong by 5 or 6 hours, so nobody gets confused.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Locking to the AC mains *phase* means you track it's changes. You don't want to track it because it is varying -- often considerably over the short term. Instead, you want to track it's average frequency over a very large number of cycles (e.g., 24+ hours worth) -- taking into account that cycles may drop out in that process due to irregularities in the line/distribution system.

Some instruments use capacitive coupling, namely a pickup wire somehow stuck alongside the hot AC line wire, to nab the AC line trigger. That could easily be flakey, especially if the box isn't properly earth grounded.

You can't expect a lot for $60 or whatever.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com