[HN Gopher] Raspberry Pi Nixie Atomic Clock
       ___________________________________________________________________
        
       Raspberry Pi Nixie Atomic Clock
        
       Author : geerlingguy
       Score  : 81 points
       Date   : 2021-09-07 06:16 UTC (1 days ago)
        
 (HTM) web link (github.com)
 (TXT) w3m dump (github.com)
        
       | freedomben wrote:
       | Confirmation bias struck me and I read this as "Raspberry Pi
       | Nixie Alarm Clock" and now I'm disappointed. I really need an
       | alarm clock and would love to build one out of a pi. I see a ton
       | of blog posts and stuff. Has anyone actually done that and could
       | recommend?
        
         | LCoder wrote:
         | I'm actually starting this right now with my two teenage kids
         | who are interested in going to college for CS. We are using an
         | esp32 instead of a Raspberry Pi, but I'm having them work
         | through all the hardware and software requirements to get sound
         | (I2S) output, display, WiFi, etc.
         | 
         | My end goal is to have them create a backend api server for
         | syncing data to the alarm clock, and then a mobile app so they
         | can change settings remotely. I figure it is a good learning
         | project for them to touch a lot of disciplines (hardware,
         | embedded firmware, circuit design and layout, 3d modeling, web
         | services, mobile apps) to see what they gravitate towards.
        
       | beervirus wrote:
       | Why not just use NTP to get time data from someone else's (more
       | accurate) atomic clock?
        
         | willis936 wrote:
         | NTP isn't going to beat a GPSDO by any metric, especially when
         | the LO is Rb.
        
         | jrockway wrote:
         | NTP is really not that good. I'm looking at "chronyc sources"
         | on my GPS-disciplined clock, and relative to GPS time, the NTP
         | sources are +/- 100s of milliseconds off. I'm sure most of this
         | is the flakiness intrinsic in a consumer-grade ISP, but even
         | after measuring them over a long period of time they tend to
         | drift at about 1 part per million. (That's a second off every
         | 12 days.) Comparatively, a tuned TXCO connected to the same
         | clock is about 0.009 parts per million off, and many would
         | consider that unacceptable.
         | 
         | GPS is actually an extremely good time reference. On the
         | ground, you can be sure of the time to about 50ns, and the
         | internal clocks on the satellites are kept synchronized to UTC
         | time. (Which variant of UTC depends on the satellite
         | constellation. The US uses the US Naval Observatory for GPS;
         | China has their own version of UTC for BeiDou, etc.)
        
         | geerlingguy wrote:
         | I'm guessing this is more of a fun project than serious--the
         | display on the nixie tubes won't refresh quickly enough to
         | demonstrate the rubidium oscillator's ability to hold
         | nanoseconds of time.
         | 
         | But in the case of potentially _also_ using this board as a
         | Stratum 1 clock (it has GPS /GNSS built in...), why would it
         | not be as accurate as pretty much any other clock that serves
         | up NTP time?
         | 
         | It would surely be a lot more accurate than any sync via NTP
         | going over the public internet.
        
           | willis936 wrote:
           | You _could_ push nixies down from +200 us to roughly + /- 5
           | us with a current sense circuit feeding a tracking circuit
           | (either a hardware PLL or digital DLL) with feedback from the
           | PPS gen circuit.
           | 
           | Most of the inaccuracy of nixie displays (that are done
           | properly and conventionally) is that the nixies require a few
           | hundred us to build up enough charge to ionize the gas. If
           | you send your HV trigger out before the actual start of
           | second then you can account for the static part of this
           | offset.
           | 
           | Reducing the random part would be... a trick. Low and
           | constant temperature would be a decent starting point.
           | 
           | My next ideas are based around taking out the feedback loop
           | and replacing it with static offsets based on an empirical
           | model. Taking measurements of the nixie state transition
           | delay (ie 1 to 2 takes a different amount of time than 2 to
           | 3) would provide useful information. You could take
           | temperature measurements and induce temperature over a range
           | and build a matrix of the nixie state transition delay
           | measurements.
        
       | at_a_remove wrote:
       | I have seen a few alternatives, but are there any nixie
       | replacements which have the features of a nixie (elegantly-shaped
       | lettering, warm glow, solid stroke as opposed to dots) without
       | the voltage issues and rarity? Lixie seems to have stopped
       | producing, etc.
        
         | anfractuosity wrote:
         | I guess it sounds like you're looking for an alternative
         | technology? but in case you haven't seen them, these folks make
         | them - https://www.daliborfarny.com/
        
           | newman314 wrote:
           | I was just looking at these this AM. They are $$$$ but so
           | beautiful.
        
           | scrumper wrote:
           | They have a very interesting project of a clock custom built
           | for NASA to calibrate high speed cameras. Apparently these
           | tubes can switch 100,000 times a second and their lack of any
           | flicker makes them well suited to this purpose:
           | https://www.daliborfarny.com/project/calibration-display-
           | for...
           | 
           | I've worked with vacuum tubes a bit in audio amplifiers, but
           | this seems wild, honestly. Does anyone know how Nixie tubes
           | can manage such a feat? 100,000 times a second? It's
           | incredible.
        
         | tyingq wrote:
         | Someone tried edge lit acrylic:
         | https://hackaday.io/project/165079-lixie-ii-the-newnixie-for...
         | 
         | What appears to be a commercialized version:
         | https://www.banggood.com/Geekcreit-Pseudo-Glow-Tube-Clock-DI...
        
       | willis936 wrote:
       | This has to be my favorite thing in a while.
       | 
       | I've been doing small hobby iterations of gps nixie clocks for a
       | decade. I'm using a pi in my latest iteration, but man this
       | project really goes the distance. It does everything I want to do
       | in my most ambitious vision of the project and then goes another
       | 50% further. Really excellent stuff.
        
       | geerlingguy wrote:
       | When I was working on testing Facebook's PCIe Time Card on my Pi
       | CM4 last week (it has GNSS and a rubidium atomic clock built in),
       | I found out about this project, and thought it looked amazing!
        
         | cyberge99 wrote:
         | My first thought when I saw the headline was that it was one of
         | your projects building upon the RTC unit you have.
        
       | jrockway wrote:
       | This is a great project! I was expecting a run-of-the-mill GPSDO,
       | but this is way better.
       | 
       | I've been looking for more time sources to add to my own GPS
       | clock: https://github.com/jrockway/beaglebone-gps-clock (which is
       | one of those boring ones that everyone has).
       | 
       | I have been thinking about buying a rubidium oscillator from eBay
       | for fun, but basically anticipated the problem the author had --
       | they have been mistreated for years before finally landing on
       | eBay, and aren't any good. GPS is my only frequency reference, so
       | I wouldn't really be able to fix it; I'm at the mercy of my very
       | obstructed view of the sky.
       | 
       | I'd like to be able to measure how badly out of sync I get when I
       | fall down from 16 satellites to 4 satellite throughout the day. I
       | can sort of intuit that from the data I get right now (I send
       | "chronyc sourcestats" to influxdb every 30 seconds, and you can
       | see that when the GPS signal gets bad, the frequency error in my
       | system clock also gets bad.), but I'd like to measure it
       | conclusively. Will probably spend a bunch of money on the problem
       | and become even more unsure which way is up ;)
       | 
       | The link to the "time to digital converter":
       | https://www.ti.com/lit/ds/symlink/tdc7200.pdf is very
       | interesting, and I totally forgot that I should be measuring AC
       | powerline cycles to see how they're doing relative to GPS.
       | 
       | (In the past, I also incorporated a WWVB receiver, but get such a
       | terrible signal in New York that I didn't get any usable data out
       | of it. I then dropped the ceramic antenna and that was the end of
       | that project. I have since realized that I can buy a phase-
       | modulation receiver inside of a clock, throw away the clock, and
       | try that, however. Ordering one right now, actually! Thanks HN!)
        
       | barrettotte wrote:
       | This is the kind of stuff I love to see; an absolute mad man
        
       | nirav72 wrote:
       | I've been wanting to build a Raspberry Pi controlled Nixie clock
       | - but instead of a Atomic Clock, I'm going just use a USB GPS
       | dongle. Also set it up as a NTP server for machines on my home
       | network.
        
         | magicalhippo wrote:
         | I'm working on just that. The GSP NTP server is handled by a
         | different Pi, but makes little difference. I used Chrony and
         | it's working like a charm[1].
         | 
         | I'm settling for a regular RTC[2] rather than an atomic clock
         | for the fallback though.
         | 
         | Designing the PCB's is taking a bit of time though, me being
         | rather new and Nixie tubes being 170V does slow things down a
         | bit.
         | 
         | [1]: https://news.ycombinator.com/item?id=28372861
         | 
         | [2]: https://www.nxp.com/products/peripherals-and-logic/signal-
         | ch...
        
         | Raineer wrote:
         | I've had a Pi GPS clock running for 6 years now, it's
         | fantastic. It's not only an NTP server for my house but it's
         | been contributing to pool.ntp.org for the duration as well. I
         | adore it.
         | 
         | Mine was built with a much older version of this board - https:
         | //store.uputronics.com/index.php?route=product/product.... I
         | like using the HAT as a dedicated solution instead of the USB
         | route.
        
         | sowbug wrote:
         | Good luck with the project. Are you adding the GPS component as
         | a hobby challenge? If it's going to be an NTP server, then it
         | could also be an NTP client.
        
           | nirav72 wrote:
           | Its more of a hobby project. Have a spare Rpi 3b and Usb GPS
           | dongle from another project. Plus, saw some cool projects on
           | the nixie sub. Figured, it would make a nice looking retro
           | clock in my home office. The USB dongle is a big finicky. So
           | if I can't get that to work - then I might just make it into
           | a NTP client and let it sync up with a public server. Nothing
           | on my LAN or homelab is mission critical where I need that
           | level of time accuracy.
        
       ___________________________________________________________________
       (page generated 2021-09-08 23:00 UTC)