Casio G-Shock time sync radio(gshock.casio.com) |
Casio G-Shock time sync radio(gshock.casio.com) |
Keep in mind antenna placement matters with these clocks. Different parts of the room can have very different signal strengths. Away from other electronics and near a window is all better.
Hope there were more stations...
https://www.amazon.co.jp/-/en/Seiko-GP219W-Office-Satellite-...
Looks great, has a quiet sweeping hand, handles DST automatically provided you set the timezone correctly.
I bought it in part because of this feature and it was quite sad. These quartz watches are good but they still drift substantially after a few months.
Thankfully I heard of the DCF77 emulator app thanks to this thread and I will use it.
Why do we have to set the time on our clock radios or on our microwaves or whatever? Why don't they just always use this Tech. I don't understand. All you have to do is allow people to store an offset for the time zone, and RTC issues are a thing of the past.
I felt so overwhelmed with excitement that I ran over and woke my sister up to tell her my watch was syncing with the atomic clock in New York. She did not nearly share my level of enthusiasm over the subject matter at hand.
Of all of the more stereotypically autism-related memories I have of little me, that one is one of my favorites. My passion and excitement at the time was so cute. :') <3
My Protrek basically doesn't sync if I'm at home, and as it tries to start sync after midnight the auto sync near-never works. I installed DCF77-emulator app on phone to get around it.
My radio-synced clock works fine but comes with big (BIC lighter size but double the thickness) external antenna.
Probably much worse in microwave that's somewhere inside the house and not near window
> All you have to do is allow people to store an offset for the time zone, and RTC issues are a thing of the past.
Just to clarify, we're talking about people who couldn't bother to put $0.2 capacitor to hold the RTC battery for the few hours the power is usually down, and you're expecting them to put whole antenna and radio in ?
Yeah, it isn't feature anybody will choose white goods for so it isn't in.
https://www.meinbergglobal.com/english/info/time-synchroniza...
> Why don't they just always use this Tech. I don't understand.
The bit's that get your oven to work don't need the time. Adding the bits for this is a cost and a thing that can fail because it can't receive the radio signal that equals a Dead on Arrival for your oven and a service call at least. When it's on a watch it's likely near your wrist which is likely near a window which likely means it can get the radio signal to sync.
https://www.dhs.gov/sites/default/files/publications/GPS-PNT...
https://phys.org/news/2011-06-power-grid-disrupt-clocks.html
https://medium.com/@bramus/grid-time-or-why-are-clocks-in-eu...
https://www.entsoe.eu/news/2018/03/06/press-release-continui...
https://www.nist.gov/publications/time-and-frequency-electri... ("Due to the efforts of Henry Warren, inventor of the Telechron electric clock, electric power companies have been a source of time and frequency reference for the public for over a hundred years. However, advances in technology and changes in the electric power industry have generated a movement within the industry to end the time-reference service.")
That said, the most power line synchronization is useful for is clocking not time, and to be very honest, its only a consistent clocking source in some places (who perform adjustment to ensure a long term value of 50/60 Hz), which means it can be used to provide a clock source to keep a device in sync, which already knows the time, but cant be used for cold start of a device.
.. it's not really precise 50Hz, just +/- 50Hz.
.. there are more than just proof of concepts, you can tell from a video, when it was made from the backround noise from the main grid.
https://www.umiacs.umd.edu/publications/seeing-enf-natural-t...
.. a few years ago, was the European electric network several minutes to late, because not every country did pay for the electric power.
https://www.nationalgeographic.com/science/article/european-...
Similar problem with GPS being used for clock sync in mobile/cell base stations in much of the US. You can buy an (illegal) jammer off Ali express and bring down a tower for very little money.
This would likely result in lot of returns, warranty claims and so on. Just not worth it in general.
...as opposed to what? How is a blinking 12:00 a better experience?
Probably not expensive to implement, but if you’re building a lot of them, I imagine the ~4cents per unit (maybe more, just guessing) adds up fast.
Prof: Why did you stopped adding passenger side handles? How much one cost?
CarGuy: Around €1.50
Prof: That's not much?
CarGuy: A new car leaves the factory every 60 seconds.
Prof: Oh.
The real question is why does your microwave gives time in the first place. Sounds like they had a display and didn't know what to show on it while the microwave wasn't actively used, I'd say drop the clock altogether
And they are too cheap to even put a capacitor in to cover for most short power failures
The main reason is lack of reliability or better to say strong dependency of reception success from location of the receiver. My G-shock watch never synchronizes when in kitchen but gets sufficiently strong signal in the bedroom. Wrist watches are more mobile and have better chance to be in a good signal reception spot at least once in a while. A microwave, on another hand, is stationary and if out of luck (weak signal) that it is permanent. Not a good solution for time sync (and possibly complains from dissatisfied customers).
I have seen this automatic radio thing once in a very old alarm clock owned by my grandfather but that was probably in the end of the 90s.
https://en.wikipedia.org/wiki/Utility_frequency#Time_error_c...
0 = Standard Time
1 = DST Ending
2 = DST in effect
3 = DST Beginning
There is a plot device for a Tom Clancy novel in this little factoid.
https://github.com/hzeller/txtempus (A DCF77, WWVB, JJY and MSF clock LF-band signal transmitter using the Raspberry Pi)
ios and mac app: https://forum.audiob.us/discussion/49109/pips-a-radio-time-s...
(time sync over rf is super cool; I would be super interested if anyone has links to field surveys of time signal availability, potential gaps in coverage, and maps of said data.
A bit unrelated question: there are many web SDRs that let you listen to radio frequencies/stations such as [2]. Are there web / Emscripten SDR decoders that run in-browser? i.e. a full browser listening + decoding experience? For example, decoding a DCF77 time signal, or even just morse code?
[0] https://www.nist.gov/time-distribution/radio-station-wwv/wwv...
[1] https://github.com/henningM1r/gr_DCF77_Receiver/tree/main/py...
All of which is a false economy. As I buy what is actually a cheap mass production watch as if it were some sort of heirloom piece that I will be passing down to my grand children. Anyhow, whats the point of being a geek if you can't overthink things sometimes?
I got the battery off eBay for about £16.
Straps are what give out first and I've been through about 4 where it's become unsightly or otherwise degraded.
It routinely will set itself to a totally random time far into the past/future. I've wondered if maybe contemporary wireless devices interfere with the time sync radio. Another possible factor is the watch itself is ~20 years old.
One works perfectly, most of the others are usually good, one gets off a couple of times a year (so therefore useless), and my Waveceptor has gone nuts maybe a couple of times.
It seems to be all about signal strength (and a little bit about battery strength).
I see at least one of those watches has GPS reception.
I have it disabled. Once it was set to a random time, and I lost an important meeting because the alarm didn't ring at the correct time. The usual time doesn't deviate more than a second or so every few years, when I need to change the batteries and set it again so it's useless.
This automatic feature would be perfect if it acted as a "press this button to setup" or at startup, but the issue is that the messages are sent very sporadically so you either need to wait almost a full day for it to sync (clocks are not always listening, so it needs to coincide the "push" and the "listen") or set it manually.
You press the receive button on the Casio.
> but the issue is that the messages are sent very sporadically so you either need to wait almost a full day for it to sync
The signal is constant. You probably had bad reception. In case of DCF77 it needs one minute to transmit "date + time + parity". I have seen my G-Shock syncing in under two minutes.
https://www.junghans.de/en/collection/watches/junghans-radio...
A nice feature, but not usable in Thailand, where I live.
I also think most (all?) timepieces with a GPS antenna will sync with good enough precision when you have line of sight to the sky.
It syncs almost every night, and takes into account DST just fine.
"A radio station at Rugby was first operated by the Post Office from 1926, with the call-sign GBR. From 19 December 1927, it broadcast a 15.8 kHz time signal from the Royal Observatory which could be received worldwide. It consisted of 306 pulses in the five minutes up to and including 10:00 and 18:00 GMT, with a longer pulse at the start of each minute. "
(The usual Casio G-Shock time deviation is under 20 seconds per month, so there's that.)
Works flawless in Seoul and Beijing. A month in South Bay and flawless too.
There is no signal for this back home in Singapore. What I do - since we also don't know when they're gonna shut off the radio towers - is use an app on my Tab S8 Ultra, set my watch's time zone to Tokyo, leave watch right next to speaker and go do something else. After some beeps and minutes : my watch time is readjusted.
No fiddling.
- Looks like classic G-Shock.
- shows correct time always (time sync)
- never runs out of battery (solar charging).
Apparently Garmin have some Smartwatch that actually got low enough to also be pretty much constantly on Solar only power but after owning original Pebble I kinda healed myself from smartwatch hype before the fashion for them even started...
It's very large, though. I haven't found a radio-set 'square' Casio in the Baby G-Shock range which would be a better size.
I bought it maybe 10 years ago. I think it's "origin" series or something like that. They still sell them in ebay for ~$100.
The closest modern replacement is the GMW-B5000, but that's quite large and expensive compared to what they were doing in 2008.
Using the HDMI lead interferes with the rPi GPIO signal being used to supermodulate the RF. As long as you don't drive the HDMI, its fine. I think its possible HDMI is an RF radiator.
The java/javascript trick with headphones works too.
With a $20 antenna placed near the window, together with a modern receiver, it is easy to get nanosecond level of precision for GPS systems. Some receivers even have in-door precision specification.
GPS not only work indoors, in terms of timing, it works actually pretty well indoors.
There's no need to 'solve' for time - it's included in the raw data packet from any single satellite.
As any 'solution' carries an inherent error margin, the desire for an excess of sats doesn't stem from solving for time but the reduction of error.
the GPS module itself is probably bigger than most of these watches, and the module itself probably costs more than most of the watches too! (probably something like $100). even if you can power it, which, the power consumed is probably 3-4 orders of magnitude more than the watch.
on the other hand, you have a point about receiving a signal and setting the clock, and I wonder if you couldn't parasitically drive this off the cell network... is there a local reference clock signal in the cell system that you could pull timestamps from and set the clock? cell signals are much much stronger than GPS and don't require constantly tracking satellites to maintain a fix. and while you still need some RF components (as does this casio watch, it's still acting as a RF demodulator!) they can presumably be much simpler and cheaper and lower-power.
(edit: actually I see some people here commenting some watches apparently do exactly this, so your "use the time reference from a single satellite and accept the skew from lightspeed" idea probably does work, although it's a smartwatch and runs its battery quickly.)
really I think most people do not even need continuous updates though. This is something you could build into a "charge cradle" for the watch and have the time reference sent via a data carrier on the inductive charge signal. which does open GPS sync back up a little bit (although wifi/bluetooth would be cheaper and just as good for most people) since the GPS can live in the cradle too.
people are talking about solar watches etc and to be honest if you've got 1-2 years of battery life who cares about solar? Pop it on the charge cradle once a year and charge it inductively. Inductive charging might actually be simpler/cheaper than even those micro solar panels watches and calculators use, and everyone has gotten into the habit of charging their smart watches overnight anyway. And if you do that, it allows you to do the cosmetics of the watch however you want, since you're not tied to having a big solar panel in part of it.
Or for some mayhem just stick a few of them under seats on public busses / trains set to activate at the same time.
The clock was an upsell long ago, but now that the whole device is a commodity product, and the clock is using parts already in the machine, I guess it’s just not worth removing.
I have many clocks with WWVB time. Many miss signals and become off, one even gets an error once in a while and has the time many hours off. Not all decoders are the same apparently.
That said, do these radio signals contain the exact time for that timezone or is it more of a synchronization signal that goes "it is now exactly 0 minutes, 0 seconds" once an hour? Since the image shows that the signal crosses some timezone boundaries.
I was able to use the FSK decoder, it worked really well. For those curious, the C-based decoder extensions are here [0]
[0] https://github.com/jks-prv/Beagle_SDR_GPS/tree/master/extens...
https://www.e-education.psu.edu/geog862/node/1724
Edit: I would welcome corrections though if I am misunderstanding. Perhaps if you're not interested in precise timing you can read the satellite time from the broadcast signal?
Edit 2: OK, I think I understand your point now. The satellites broadcast their timestamps directly. To find a precise position / time you need to find a correction to the receiver clock. But if you don't need a very precise time, you can just use the satellite timestamp.
I had a ~$15 radio-synchronised clock when I was quite a long way from the transmitter, in a solid brick building. It would get a signal in some locations and orientations in the building.
I don't have firm information on how good the signal reception was, as of course it would keep ticking even if it didn't have a signal.
However, I will dig it out and see if it can be resurrected.
https://en.wikipedia.org/wiki/Utility_frequency#Time_error_c...
somehow it's way less frustrating to only have to adjust the clock 30 minutes or whatever. you're not gonna be standing there for five minutes holding the button, you know? even though in practice it's not that big a deal and doesn't happen that often, it's a little human nicety.
and if you don't have time to adjust the clock at that exact moment, you can just mentally correct "I was out of power for 30 minutes so it's 30 minutes slow right now" rather than having to remember "I lost power at 3pm so that's 12:00...".
And on other hand if marketed people expect such features to work. Even indoors or in basement. Next to window facing right direction in lucky meteorological conditions is tough ask for majority. Geeks will geek out when it works once in a bluemoon. But regular people expect it to work every day or even every week.
And I think few people expect their wrist watch to synchronize over radio in a basement... And if they are, they need to understand radio better.
There is little drift on my G-shock anyway. Probably it would still be extremely accurate if it hadn't synced at all during the 7 months I was there.
I would understand manufacturing pain, and the ROI just doesn't justify it. But saying that it's solely to do with costs doesn't line up, IMO.
Stellantis shares tons of parts and software. Peugeot 2008 and Opel Mokka is essentially the same car. They share the same shifter, electronics and software stack sans the skin applied to displays and LCD geometry. Even 3008 has the same hardware with the same software. Heck, even the "lane hold" button is exactly the same one in all three cars. Same for parking brake switch, down to the blinking pattern of indicator LED. I bet all three use the same 8 gear auto gearbox, with different ratio sets, too.
After Fiat bought Chrysler, they started to use the same steering wheel in Dodge Charger and Fiat Tipo. Newer Fiats come with Mopar brakes and fluid subsystems (steering, cooling, etc.)
Key fobs are even funnier. In practice, every company uses the same key fob. A Lamborghini has the same fob with a VW Polo, sans the logo. Same for Stellantis group cars.
When you start to notice "chassis sharing" in commercial market vehicles, it stops being funny and starts to become ridiculous.
On Aston Martin key fobs, when the rubber coating wore off, it literally had the Volvo logo underneath. Ford parts bin.
> I guess, but still, these seemingly large numbers are still a minute percentage (e.g. probably 0.00001%) of overall costs/revenue that are relatively way, way more.
Do you realise that your estimate, of €1.50 per car being 0.00001% of the costs, puts the manufacturing cost of a car at €15m?
Every button needs a bit of plastic moulded, a (good quality) button, backlight, often LED to signal whether it is on or off, and whole control block needs microcontroller to pack it into CAN bus and CAN bus connector to send.
Make whole thing a touch surface and you're saving buck or two on buttons alone and your plastics don't need to articulate (another savings).
Move that to the touchscreen controls, and as screen is already there, more savings!
I just wish we ended up with aircraft style MFD's instead of touch screens. You know those things with a screen and a row of buttons along each side.
700k in advertising would make a much bigger impact on revenue than 700k on door handles; unless people are that passionate about door handles that they go somewhere else.
Even if your current advertising budget is orders of magnitude larger, the force multiplier on investing in advertising is bigger than the door handle multiplier. So you’re just burning money.
Of course that’s all very cynical. Make the effort to reward companies that build the best product they can.
My understanding is that this is now going back and car manufacturers are reverting this decision.
So yeah, why not spend the saved money on advertising? I don’t know the complexity of such decisions. There are some manufacturers that do take a stand not to cut as many corners. Over time they can become known for their design. Apple (at least until the mid 2010s) I’d say was one such company.
By the way, as was pointed out, £700k seems negligible to companies that size for something that might potentially torpedo a car if enough people hate that part and refuse to buy it based on those grounds. But I’m not an expert.
However, I'd rather have a boring car which I don't need to constantly maintain instead of an exciting one which needs a tune-up every weekend but that's just me.
It's not that I can't. It's rather I won't.
Sometimes they might think the higher end of the market is worth dominating by making the best product possible.
Other times that might be overkill.
It’s all very dependant on the company. Glad I don’t have to make those decisions! I just pretend to understand them on the internet heheh
It costs them $0 because consumers already expect a touch capable screen for infotaintment.
> I just wish we ended up with aircraft style MFD's instead of touch screens. You know those things with a screen and a row of buttons along each side.
Yeah it's a dream. Anything touch is annoying to use when driving. MFD-like also have advantage of being able to be touch typed, once you know in which menu you are it's always same sequence, and as those are physical buttons it's far harder to miss-press something.
Bingo, it would be a much different calculus if most potential customers didn't expect or want touch infotainment systems.
But, it's a commodity item, produced in many more numbers w.r.t. a A/C temperature/fan encoder w/custom shape. A bog standard Renault Clio touchscreen is comparable to a good (not high, not top) quality Raspberry Pi screen you can get from RS or AdaFruit.
The screens on Opel Mokka I drove (it has two, one for dash, one for infotainment) were extremely good at sunlight, but the dash one was so small, around 7" IIRC, to control costs. So, even if they don't pay for R&D of the screen itself, the industry is so cost sensitive, that they will cut costs relentlessly to be able to sell more of these things at a lower cost and with higher profit margins.
Also, these LCD dashes has a couple of hidden LED indicator lights around to communicate fatal things in a failsafe manner.
Car manufacturers do not bear the CapEx required to develop these screens, but bear the CapEx required for developing and prototyping physical buttons.
> I just wish we ended up with aircraft style MFD's instead of touch screens. You know those things with a screen and a row of buttons along each side.
Industry is evolving on that direction now.
