Radio controlled alarm clock query.

[Tractorfan]

Hi,
I've got one of those electronic radio/alarm clocks that projects the time onto the ceiling. It is radio controlled and sets the time automatically from a station I know not where.
When we have a power outage, it resets itself upon restoration of supply.

But.

It takes several hours of showing the wrong time before it finally sorts itself out.
I'm just wondering why? Our church clock is always correct, but I assume that is thanks to a battery back up. Similarly, the car clock is always right.
Cheers, Pete.

[turretslug]

I've noticed that they can take a while to catch up- I assume that it looks for one of the LF time signals around Europe, I wonder if there is now so much RFI garbage in this region of the spectrum that it takes the gubbins a while to decide that it can discern a valid signal. They often seem to have a teensy-weensy fixed-orientation ferrite stick aerial in them that might suffice in a clean environment, but sub-100kHz is pretty grubby. I assume that the car clock may be set from either FM RDS or DAB and stays fairly closely on time with ignition off?

[G6Tanuki]

The clocks in current-decade cars generally use GPS or RDS - they get a pretty rapid 'time-sync', same as later VCRs and current PVRs sync their clock to over-the-air timestamps.

"radio controlled" alarm-clocks, and my radio-controlled wall-clock in the shack -- yes they do seem to take a while to 'catch up', even in a relatively RF-quiet environment. I'm not sure if the 'radio' bit of such clocks runs continually, or whether (as I suspect) it only fires-up occasionally to listen for the time - this would be much better as a battery-conservation strategy.

[In times-past I had a lot to do with the Internet "Network Time Protocol" - the strategy here being to take repeated time-sync data from several sources of different quality and use this to make minor, incremental changes to the local-clock. Eventually the local-clock 'converges' on the external-clocks and harmony is achieved. There may be user-definable constraints applied to whether you 'publish' the local-clock for use in timestamps in logfiles, financial transactions etc. if it is too divergent from the external clocks.

In a way it's a bit like the "Huff and Puff" VFO-locking-to-a-crystal scheme!

See https://en.wikipedia.org/wiki/Network_Time_Protocol for more info]

[KeithsTV]

Many of the radio controlled clocks only check for an update once a day so will show an incorrect time until it updates.

There is the MSF transmitter at Anthorn in Cumbria https://www.npl.co.uk/msf-signal on 60kHz and the DCF transmitter on 77.5kHz near Frankfurt https://www.ptb.de/cms/en/ptb/fachab...ime/dcf77.html although the data formats are different.

Keith

[julie_m]

I've got a Boots radio controlled travel alarm clock, which was a Christmas present several years ago. It has interchangeable modules (referred to as "antennas" in the instructions, but they have several contacts suggesting more electronics inside) for time signals in use in different parts of the world (UK / USA / Japan West, Europe, Japan East). It will try to self-synchronise several times a day, at 01:00, 02:00, 03:00 (to be sure of catching a daylight saving change), 09:00, 15:00 and 21:00, but you can also force it to do so with a button press.

My cordless land line phones synchronise themselves to the timestamp from an incoming caller ID packet; so if a handset is fully discharged and then recharged, it will show a nonsense time from the moment it has enough juice in the battery to work again until someone rings.

[Tractorfan]

Hi, and thanks for your replies.
I think the radio/alarm was originally bought in the Netherlands, so could be a different system from here in France, possibly? Our house has very thick walls, so a weak signal to a small ferrite rod makes sense.
The car is a twenty year old Audi which had an ancient GPS system before I removed it, so the clock isn't dependant on that, as I thought it might be.
Cheers, Pete.

[julie_m]

My Boots clock had one antenna module for all of Continental Europe, on 77kHz. I don't know for sure if that's a single powerful transmitter or a network of smaller ones in several countries.

[KeithsTV]

The 77.5kHz signal is from a single transmitter. I get a good signal here in Bradford. Also a strong MSF signal on 60kHz.

I believe there is/was a French station which used the 162khz broadcast transmitter with the data modulated onto the carrier, probably in a similar manner to the data modulated onto the Radio 4 LW transmitter on 198kHz. The data format is the same as the DCF transmitter.

I have a clock/timer with the option for synchronising the time with a radio input. Being of French origin it used the 162kHz signal but works equally well with a 77.5kHz DCF receiver.

Keith

[duncanlowe]

Many 'radio clocks' use the Frankfurt time signal even if sold in the UK. That can give issues with signal strength especially if there are steel structures nearby. It can even happen with UK MSF tuned ones in a bad environment, rmembering that a bad environment doesn't have to be RF noisy, just have high LF attenuation.

You Audi probably also has a DCF clock module attached to the back of the instrument pack. UK cars of the era didn't have them but I do remember the coding option being there. If I'd added the coding it would have only given me a fault code for loss of communication with the add on receiver.

[duncanlowe]

Also just to add, there are other ways that radio clocks can get their time. Rather than 'radio controlled' clocks, many clocks have radio receivers built in, as in FM or DAB. Both of these now also carry coded time information (via RDS for FM, or in the basic data for DAB) that the clock can sync with. In those cases many will only sync at specific times, unless the receiver part is actually in use.

As an example of how this can be irritating, I have a Sony DAB that I uses as my bedside alarm clock. If the receiver isn't used fro a few days, it's internal clock can get ahead of the BBCs clock. So the alarm goes off and I wake and turn it off. But in the mean time it has resynced to a minute earlier, and when 5:29 comes round again it comes back which is irritating if I've not noticed and am in the bathroom. (Irritating for 'those still asleep').

[stainless]

I have several radio controlled clocks, some Anthorn, some Frankfurt. I always note that there is a time difference between the two varieties - usually a few seconds, but have seen a difference of 19 seconds. The error varies! I suspect neither of them is right!!

What do other people on here use as a time reference??!!

[jjl]

Originally Posted by G6Tanuki

The clocks in current-decade cars generally use GPS or RDS - they get a pretty rapid 'time-sync', same as later VCRs and current PVRs sync their clock to over-the-air timestamps.

That's probably true in general, but my 2012 VW Passat requires manual setting of the time, even though it has RDS, DAB radio and GPS. I've often wondered why this is.

John

[G6Tanuki]

Originally Posted by jjl

That's probably true in general, but my 2012 VW Passat requires manual setting of the time, even though it has RDS, DAB radio and GPS. I've often wondered why this is.

John

You may find that - if you leave it long enough without manually-setting the time - it will eventually get it right.

The whole principle of 'convergence' of time-sources is based on the difference between local-time and network-time: the further-off the local-time is from network-time the longer it takes to converge. If your local-time starts at 00:00-01:01-1980 it can take a while for it to creep sufficiently-close to 'network-time' for the clock's logic to decide it's now "close enough" that it will enable the display.

Giving the local-clock a 'hint' by manually-setting it to something close-to-network-time can dramatically reduce the time taken to converge.

Paradoxically, but understandably, though your local-clock may be close to converging with its network-clocks, it may choose to still show 00:00 - on the basis that this is a positive indication of "I don't know the exact time", rather than it showing a time which is close-but-not-right.

Professional "network-clocks" as well as showing the time, offer quality-information in the form of 'scatter' and 'offset' data which reports the differences between local-time and the network-time derived from different sources.

[SiriusHardware]

Originally Posted by duncanlowe

Many 'radio clocks' use the Frankfurt time signal even if sold in the UK. That can give issues with signal strength especially if there are steel structures nearby.

And one side effect from those, typically sold by German-origin stores like ALDI and LIDL, is that they may show Central European time by default. You might have to go into the menus and add a one hour offset to get the clock to show the correct UK time.

On a purely analogue radio controlled clock of European origin you may just be able to pull the hour hand off and put it back on offset by one hour.

I have done quite a lot with MSF type radios and i can say with a half (well third, it was me after all) decent bit of firmware they can give a "time fix" in less than a couple of minutes.