WiFi Clock

When I entered my name and couldn't remember my PW I asked for a code which it sent me and so I was able to log on.

What do you mean by wi-fi clock? We have 4 radio controlled clocks ie in theory they should all show the correct time & update themselves without user intervention.

My "Atomic" clocks are somewhat fussy, see my opening comments!

Mine really is an "atomic clock". Slightly cumbersome, but accurate to something like 30ms per year after 10 years. Its a Symmetricom LPRO rubidium oscillator in a small box with a mains power supply. It probably only gets turned on for a few days per year however for checking the calibration of frequency counters. The counters it was originally bought for checking were in a huge building at a location far from windows and with a steel roof a long way up. It was impossible to receive GPS signals there. It is the place where Raspberry Pis are made in S. Wales.

Aging: Monthly (after 1 month): <5E-11/month

John

This one is 100uV at up to 1000km. That should be enough to do it, and the transmitting antenna is omnidirectional. That is 60KHz (a little less than double the quartz crystal in your digital watch or ATX computer RTC).

The design inside a clock, consists of a piece of Russian Ferrite (other suppliers do not make that particular flavour of ferrite rod). There is a tuned circuit, an LC fitted to the antenna. And that's enough of an antenna for a 100uV signal. The integrated circuit fitted in the clock, has to acquire that 100uV signal.

The scheme in North America is similar, except the signal strength does not guarantee good reception at all times of the day. A Radio clock in North America should be able to achieve a time fix once a day (perhaps in the evening hours). Other times of the day, the signal might be a little weak. Propagation may also differ between summer and winter. The UK scheme is good all day.

If you don't have the piece of ferrite, you can use an ordinary "aircore coil of wire", but the amount of wire required, is about the same size as a foot diameter wall clock with a big face. Not exactly convenient. (There is a UK web site which contained pictures of a working wire coil.) But if for some reason you lacked the piece of Russian ferrite, it's still possible to wind your own coil antenna (and fit it with a capacitor to tune for max signal at the intended frequency). I don't know how the Q of those two schemes compare (the sharpness of the frequency response). That ferrite makes a big difference.

Really, your atomic clock should have worked. You are not in North America where our signal does not meet the minimum at all times of the day. The LF signal penetrates buildings no problem at all. It is not a 2.4GHz or 5GHz Wifi signal, it is only 60KHz and a relatively low frequency for RF. You would think that harmonics of line frequency would potentially interfere.

Paul

If you have wifi the 60kHZ clock is simply irrelevant.

Boot up a PI ZERO W, and it will acquire 'internet time' and 'know' about time zones Couple it up to a 5 digit display of some kind - or indeed some stepper motors.. ..and you have a better than a few ms clock.

Now moved to Anthorn in Cumbria some time ago and as its further away from places like London the signal isn't quite as good as it once was...

The trouble with such things is power. If you want a wall clock you have to provide a power supply to keep it powered all the time, which means running a wire. If it had a battery it would run down in a few days. I don't think you want to add the clock to the ever lengthening list of things you need to keep charged.

I expect you could do something similar with something like an ESP32 that goes to sleep for 59 seconds of every minute, waking up just to update the display and then going back to sleep again, syncing with the internet perhaps once a day. But unless somebody has already implemented that it's not exactly a ten minute programming exercise.

I expect the commercial 'wifi clocks' do roughly the same thing.

Theo

Surprised MK still sell these, only one I remember seeing was above the mantel in my parents' 1968 house.

Dad made an octagonal clock from a sheet of plywood with matchsticks as roman numerals.

Those would actually be quite neat for wall mounted things that need power, like TVs and smart widgets. The downside is they tend to be quite ephemeral and not worth chiselling up the wall for. But maybe in these days of dry lining everything it wouldn't be implausible to fish a wire behind the dot and dab.

Theo

I have a 12" Acctim radio controlled wall clock in the kitchen. Runs on 1 x AA battery, changed annually, worked perfectly form day 1.

My quartz clocks are not radio controlled and run on a battery. But since they need resetting every 6 months anyway, I am not overly bothered about their accuracy

From what I can see, nobody answered your question.

Did you get a WiFi clock?

If so, which one?

And if not, could you give a link to the ones you found?

Thanks

I did!

Oddities: You can't immediately set time zone 0 that seems to be taken as a null entry so set it to another then to 0. Sometimes when it flips over to a new hour it shows a random time then shows the correct time. It sets its brightness either automatically or to a specified level, the one I have in the lounge has its sensor covered so I set that manually, the bedroom one is set up to automatic. Service was impeccable, the original instructions (on their website) didn't seem right. I emailed the seller (address on website) and the instructions were completely update in 24 hours.

I ended up buying a second one so have one in the lounge and one in the bedroom.

Does the gentle warmth from power constantly passing through the electronics affect the temperature reading?

I have two 'made in china' clocks with inbuilt temperature measurement. Both are in enclosures without any visible ventilation slots. One is in a small enclosure and the other in a larger enclosure similar to the photo linked in the original post. Both are 5V USB powered. Both read consistently 1.5C higher than a (digital) thermometer claimed to have an accuracy of +/- 0.3 C. The accurate digital device is a low power device running for 2 years from a coin cell battery (also with wi-fi for setting and reporting).

I suspect that readings given on the two clocks are influenced by the temperatures generated by the electronics. Both have fairly high intensity LED displays rather than LCD. But, most cheap digital thermometers don't claim an accuracy better than +/- 1C

I left the temperature sensor tucked inside the clock so don't use it for temperature.

You can choose the currency (along with the country/language settings).

I built some wifi capable thermostats. I mounted the temperature sensors on the same PCB as a Pi Pico W.

Compared with an old room stat with a display, they read high. How high?

Easy enough to fix in software....