[neonixie-l] Re: Rare B8971 Nixie tubes

['Greg P' via neonixie-l]

Hmmm




On Thursday, June 20, 2019 at 2:06:59 AM UTC-4, Robert L wrote:
>
> As Michael says... H
>
> [image: IMG_20190512_082331 (Large).jpg]
>
>
> [image: IMG_20190512_082252 (Large).jpg]
>
>
>  
>

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/2b3b10ad-f1dd-4026-bb7b-a4479d065255%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: Rare B8971 Nixie tubes

[Nicholas Stock]

H

Sent from my iPhone

> On Jun 20, 2019, at 03:18, 'Greg P' via neonixie-l 
>  wrote:
> 
> Hmmm
> 
> 
> 
> 
>> On Thursday, June 20, 2019 at 2:06:59 AM UTC-4, Robert L wrote:
>> As Michael says... H
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>>  
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to neonixie-l+unsubscr...@googlegroups.com.
> To post to this group, send email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit 
> https://groups.google.com/d/msgid/neonixie-l/2b3b10ad-f1dd-4026-bb7b-a4479d065255%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/2977224F-B0A9-4F8C-864C-E8C1987E9F47%40gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Re: Rare B8971 Nixie tubes

[martin martin]

Looking cool

On Thu, Jun 20, 2019 at 07:18 Nicholas Stock  wrote:

> H
>
> Sent from my iPhone
>
> On Jun 20, 2019, at 03:18, 'Greg P' via neonixie-l <
> neonixie-l@googlegroups.com> wrote:
>
> Hmmm
>
>
>
>
> On Thursday, June 20, 2019 at 2:06:59 AM UTC-4, Robert L wrote:
>>
>> As Michael says... H
>>
>> [image: IMG_20190512_082331 (Large).jpg]
>>
>>
>> [image: IMG_20190512_082252 (Large).jpg]
>>
>>
>>
>>
> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+unsubscr...@googlegroups.com.
> To post to this group, send email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/2b3b10ad-f1dd-4026-bb7b-a4479d065255%40googlegroups.com
> 
> .
> For more options, visit https://groups.google.com/d/optout.
>
> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+unsubscr...@googlegroups.com.
> To post to this group, send email to neonixie-l@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/2977224F-B0A9-4F8C-864C-E8C1987E9F47%40gmail.com
> 
> .
> For more options, visit https://groups.google.com/d/optout.
>
-- 
mcvei...@gmail.com

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/CAKYv7NQM3-0Vvr0uTPZRbGQ09i%2BHao2ehXi_ZWdrXH9sxGgERQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[Paolo Cravero]

Outstanding Bill! And it will be programmed in Assembly!

Are you using some tricks to improve the reliability of "reading" AC line?
Either HW or SW. In Europe AC time is currently running 32 seconds late
since the beginning of the year. It will catch up, as it did in 2018 when
the grid had "lost" 300 seconds in 3 months. Every time I think my IN-14
clock has troubles with AC pulses, but in fact it's the grid running
noticeably fast or slow.

Paolo

On Wed, Jun 19, 2019 at 1:43 AM Bill Stanley 
wrote:

>
>
> I promised to post some details on my current Nixie nightmare. I present:
>
>
> The BRS Clock
>
> The current version of the clock is in 2 (or more) PCBs. This was done to
> allow changes to the display type by only either changing the display board
> or the display cluster boards. The board can be used with Nixie tubes (10
> cathode or 14 cathode “British Flag” types), LEDs, possibly VF or other.
>
>
> The CPU board:
>
> A pair of 16 pin headers on top of the CPU board connects it to the
> display. The CPU supplies 12V, 5V and 170V with the 170 able to be margined
> by the CPU. I2C connects “things” on the display to the CPU. The time data
> is sent in a serial format (clock+data) along with a transfer strobe to
> write the data to the tubes. For Nixie displays, the HV5812PJ-G chip does
> direct drive to the cathodes, there is no multiplexing. Additional signals
> include a room light sensor an 11 pins defined for control like decimal
> points, symbols and “pimp lights”.
>
>
> The CPU is currently a Silicon Labs C8051F360
>
>
> Time base. There a 4 possible time base sources for the clock. In
> increasing accuracy order:
>
> The CPU time base: This is the worst but with some work can be trimmed up
> somewhat. If this is the only time base, the clock will lose time in a
> power failure.
>
> AC line clock: The primary power source is 12V. It can be AC or DC. If the
> power source is 12V AC, the line frequency is sampled (50 or 60Hz detected)
> and used to run the clock. The AC line has good long term stability, much
> better than the CPU clock. As with the CPU clock, a power failure will
> reset the clock date/time.
>
> DS3231S: This Maxim chip along with a coin cell has very good accuracy and
> will remember the date/time during a power failure.
>
> GPS: An optional Trimble Lassen GPS receiver can be plugged onto the CPU
> board. It provides a very accurate 1PPS signal and UTC date/time for the
> highest accuracy. Other GPS receivers that support the NMEA sentence
> structure and a 1 PPS signal could also be used. The GPS has a battery
> backup to allow faster cold starts.
>
>
> 6 pushbuttons (plus reset) allow user setting of date, time and setup
> parameters. The parameters are saved in EEPROM so are available after a
> power-down. 1 button sets/unsets DST so the twice a year change is simple.
>
>
> There are a number of connectors to plug in various external devices. They
> are
>
> Indoor temperature/humidity: A small sensor will allow the display the
> indoor temperature and humidity every 5 minutes.
>
> Outdoor temperature: A different small sensor will display the outdoor
> temperature at the same time as the indoor, if connected.
>
> PIR/room light: A Passive IR motion sensor and room light level sensor
> will allow shutdown in darkness when no-one is around. The room light
> sensor can adjust the brightness of the pimp lights automatically.
>
> MP3 player: A simple MP3 player is connected to play different sounds at
> different times. The current code plays the Westminster Chimes on the
> quarter hour and the hour count at the top of the hour. Many other things
> could be done.
>
> Meter DAC: My friend wanted a “Steam Punk” look to the clock so we
> incorporated 3 D/A converters (12 bit accuracy) driven at a 1/10 second
> update rate with the H/M/S. He will use these to drive vintage large analog
> meters so the time will be read as analog.
>
> Stopwatch: Another request was for a work clock that displayed a running
> stopwatch every 5 minutes. Three control switches are provided to
> start/stop/reset the stopwatch.
>
>
> Displays
>
> The displays are mounted on a daughter card plugged into the CPU. This
> allowed me to only modify the display to run with a number of different
> display types. The following information is for Nixi display tubes.
>
>
> Tubes: Designed using the ‘QTC’ concept from pvelectronics in UK. This
> mounts the tube (solder or socket) on a small PCB that plugs into the
> display board with a 12 pin 0.1” connector. I have created adaptors for
> about 8 or so different tubes. New tubes are quick to create. The original
> used IN-14 + IN-19 tubes
>
>
> Bottom lights: These lights shine thru the base of the tube thru a hole in
> the QTC board. I call these “pimp lights” and come in 2 flavors; a standard
> single RGB LED that cycles automatically thru multiple colors and a
> dedicated WS2812B RGB LED that is individually programmed with 256 levels
> of each color. The WS2812B 

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[gregebert]

A genuine DS3231 does a remarkable job of keeping accurate time. I have one 
in a wristwatch prototype board that has not been updated for more than 4 
years, and it's still within 1 minute of the real time.

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/0e68eaf6-4af3-401c-877c-069354048578%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[Jeff Walton]

One trick for using line frequency is to rectify, and wave shape to get a 
square wave.  Then set up a monostable one-shot with a period just shorter than 
the line frequency (or doubled full wave).  This will prevent any extraneous 
noise from getting through to the divider circuit that feeds the clock logic.  
I used a scope and superimposed noise on the input to help fine-tune and verify 
the one-shot timing.  I built my first clock over 40 years ago with TTL logic 
using this method and the accuracy was within seconds/year.  I didn't have 
temperature compensated crystals or components back then.  But after many 
years, the method has proven to be very reliable.  The US grid has been a good 
timing source for a non-critical application. Jeff Walton 
 Original message From: Paolo Cravero  
Date: 6/20/19  11:57 AM  (GMT-06:00) To: neonixie-l@googlegroups.com Subject: 
Re: [neonixie-l] Presented for your consideration...The BRS Clock Outstanding 
Bill! And it will be programmed in Assembly!Are you using some tricks to 
improve the reliability of "reading" AC line? Either HW or SW. In Europe AC 
time is currently running 32 seconds late since the beginning of the year. It 
will catch up, as it did in 2018 when the grid had "lost" 300 seconds in 3 
months. Every time I think my IN-14 clock has troubles with AC pulses, but in 
fact it's the grid running noticeably fast or slow.PaoloOn Wed, Jun 19, 2019 at 
1:43 AM Bill Stanley  wrote:

 

I promised to post some details on my current Nixie
nightmare. I present:

The BRS Clock

The current version of the clock is in 2 (or more) PCBs.
This was done to allow changes to the display type by only either changing the 
display
board or the display cluster boards. The board can be used with Nixie tubes (10
cathode or 14 cathode “British Flag” types), LEDs, possibly VF or other.

The CPU board:

A pair of 16 pin headers on top of the CPU board connects it
to the display. The CPU supplies 12V, 5V and 170V with the 170 able to be
margined by the CPU. I2C connects “things” on the display to the CPU. The time
data is sent in a serial format (clock+data) along with a transfer strobe to
write the data to the tubes. For Nixie displays, the HV5812PJ-G chip does
direct drive to the cathodes, there is no multiplexing. Additional signals
include a room light sensor an 11 pins defined for control like decimal points,
symbols and “pimp lights”.

The CPU is currently a Silicon Labs C8051F360

Time base. There a 4 possible time base sources for the
clock. In increasing accuracy order:

The CPU time base: This is the worst but with some work can
be trimmed up somewhat. If this is the only time base, the clock will lose time
in a power failure.

AC line clock: The primary power source is 12V. It can be AC
or DC. If the power source is 12V AC, the line frequency is sampled (50 or 60Hz
detected) and used to run the clock. The AC line has good long term stability,
much better than the CPU clock. As with the CPU clock, a power failure will
reset the clock date/time.

DS3231S: This Maxim chip along with a coin cell has very
good accuracy and will remember the date/time during a power failure.

GPS: An optional Trimble Lassen GPS receiver can be plugged
onto the CPU board. It provides a very accurate 1PPS signal and UTC date/time
for the highest accuracy. Other GPS receivers that support the NMEA sentence 
structure
and a 1 PPS signal could also be used. The GPS has a battery backup to allow
faster cold starts.

6 pushbuttons (plus reset) allow user setting of date, time
and setup parameters. The parameters are saved in EEPROM so are available after
a power-down. 1 button sets/unsets DST so the twice a year change is simple.

There are a number of connectors to plug in various external
devices. They are

Indoor temperature/humidity: A small sensor will allow the
display the indoor temperature and humidity every 5 minutes.

Outdoor temperature: A different small sensor will display
the outdoor temperature at the same time as the indoor, if connected.

PIR/room light: A Passive IR motion sensor and room light
level sensor will allow shutdown in darkness when no-one is around. The room
light sensor can adjust the brightness of the pimp lights automatically.

MP3 player: A simple MP3 player is connected to play
different sounds at different times. The current code plays the Westminster
Chimes on the quarter hour and the hour count at the top of the hour. Many
other things could be done.

Meter DAC: My friend wanted a “Steam Punk” look to the clock
so we incorporated 3 D/A converters (12 bit accuracy) driven at a 1/10 second
update rate with the H/M/S. He will use these to drive vintage large analog
meters so the time will be read as analog.

Stopwatch: Another request was for a work clock that
displayed a running stopwatch every 5 minutes. Three control switches are
provided to start/stop/reset the stopwatch.

Displays

The displays are mounted on a daughter card plugged into th

Re: [neonixie-l] Rare B8971 Nixie tubes

[Bill Stanley]

Hi Greg;

Sorry for the delay, the real world called for some attention.

I looked up the tube and it shouldn't present any problems, but you will 
need to make some choices
The QTC mounts I make are base on a 12 pin connector. Your tubes would 
require at least a 16 pin. This is no problem as I would have needed a new 
adapter for any new footprint.
This will require more driver chips and a code change to a longer data 
strea, trivial
We will need to work out the display characters (more personal taste) as is 
see a .gif uses different segments than I used for my LED clock, trivial 
ROM change

My base CPU will work with only firmware changes, the question to you is 
about the tube carrier and what you want it to look like.
Assume 6 tubes for a H/M/S clock
Assume tubes are spaced TT..TT..TT (any colon between?)
Assume Clock will display with "character numbers" for clock, What for 
date?   JUN 20 19 format or something else.
Do you want int/ext temp/humidity? If so, what will be the display format?
You have full alpha ability, anything else to display ("EAT AT JOES")?
Will you want pimp lights (base illumination)?
Anything else?

We can go with a modified small display PCB which is a little under 2" 
depth, width determined by tube spacing or with a standard display hub with 
3 20 pin ribbon cables to 3 display clusters. Possibly wasier to build into 
a case, your call

If we get started, I should send you a clock with either a small display or 
a cluster display for you to get a feeling about. On your end, I will need 
the loan 1 or preferably 2 tubes for development

Also, I am an electrical guy, anytime I get case work done, it is a friend 
or contracted to small machine shops. I can't help with a case but will 
gladly work with and modify designs for whoever you have doing the case.

Anything I missed?
 
  -Other Bill-


On Tuesday, June 18, 2019 at 6:23:01 PM UTC-7, Greg P wrote:
>
> Hi Bill,
>
> Looking for direct drive clock that can also generate words since these 
> tubes are alphanumeric.  I have zero programming skills so hoping something 
> will present itself.
>
>
>
> On Tuesday, June 18, 2019 at 4:52:45 PM UTC-4, Bill Stanley wrote:
>>
>> Hi Greg;
>>
>> I have a ongoing clock design where I separate the CPU from the display. 
>> The main CPU board uses a SiLabs 8051 class CPU with LOTS of features (some 
>> might be considered a bit crazy).
>> The CPU generates all voltages (the 170 is adjustable under firmware for 
>> higher-voltage tubes. The cathode data is serial to the display drivers 
>> (using HV5812PJ-G). This was so I was no locked into the
>> 10 cathode Nixie format, I can do Panaplex, LEDs, of 14 segment British 
>> Flag displays, just a Small Matter of Programming.
>>
>> For display, I have the small display board with 9 IN-14 + IN 19 tubes 
>> with QTC sockets. I have the large display for 6 Z5600M Nixie + 3 IN-15 or 
>> IN12+1N15. I also have a display Hub that breaks out the display
>> signals to 3 (H/M/S) separate 20 pin ribbon that connects to a 3 tube 
>> display cluster.
>>
>> I used the basic design to create a "Back to the Future" time console for 
>> a friend where the date is displayed with 14 segment LEDs
>>
>> Were you just looking for a clock (or something more)?
>>
>>
>>   -Newer Bill-
>>
>>
>> On Saturday, October 13, 2018 at 6:49:32 AM UTC-7, Greg P wrote:
>>>
>>> I wish someone would make a clock or a kit for the 5971's.  I have a 
>>> bunch of them I'd like to light up.
>>>
>>>
>>>
>>> On Friday, October 12, 2018 at 6:15:57 PM UTC-4, Pramanicin wrote:

 The 5971’s are cute too... 

 Sent from my iPhone 

 > On Oct 12, 2018, at 15:12, Michail Wilson  wrote: 
 > 
 > Those are nice. 
 > 
 > I didn't know they even existed. 
 > 
 > Are you sure NOS 7971?  Seems that NOS doesn't exist for those. 
 > 
 > Anyway, I think you're be happier with the 7971 even if not NOS. 
 > 
 > Michail 
 > 
 > 
 > -Original Message- 
 > From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] 
 On Behalf Of Kevin A. 
 > Sent: Friday, October 12, 2018 1:01 PM 
 > To: neonixie-l 
 > Subject: [neonixie-l] Rare B8971 Nixie tubes 
 > 
 > Hey guys, 
 > 
 > Just thought I'd share some pictures of my latest purchase; six 
 Burroughs B8971 tubes. These babies are in great shape, and as far as I 
 can 
 tell are quite rare. 
 > 
 > I'm going to trade 1:1 with a buddy of mine for some NOS 7971s. I 
 originally wanted the bigger tube, even though they are more common. These 
 are really for a guy who has it all and craves the rarity, aka my pal. 
 > 
 > Anyways, thought I'd snap some pictures and show them here before I 
 trade. Take a look: 
 > 
 > -- 
 > You received this message because you are subscribed to the Google 
 Groups "neonixie-l" group. 
 > To unsubscribe from this group and stop receiving ema

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[Bill Stanley]

Hi Paolo;
Programmed, past tense. I am working on putting the last 5% of features 
into the code now (you know, the programming part that takes 95% of the 
time).
I am US based so not sure about the stability of other grids. My first 
clock (RTL based 49 years ago, details in my "about me" post). used R/C/D 
filtering into 6 hex inverters. At the 3.6V logic level + the base of the 
transistor exposed made for a very noise prone unit. I actually started a 
design of an XO to replace the AC line but never finished it. My next line 
sync clock was (and still is) my Heathkit with Panaplex displays. That 
clock would drift slightly but was very usable for the timebase cost. My 
only trick is to use an opto-isolator to generate the clock.

Slight division, a project I played (have PCBs and some code) with was to 
build a plug-in replacement (8051 based) for the Mostek chip in the 
Heathkit clock. It would use a modern time base + GPS. If there is any 
interest, I could post the work I have done and perhaps restart it if 
anyone else is interested..

Jeff; For my clock line time base, I decided on a brute force filter. I use 
a opto-isolator powered by the rectified AC line with the assumption that 
any noise would need enough energy to light an LED via a large resistor to 
get thru. Not impossible but a fairly high bar. Since that signal does go 
to a counter/timer, I could create a software one-shot to add a bit more 
protection, I will think about that (and there goes 5% more overrun on code)

Gregbert: Interested in the "genuine" qualifier and what is behind it. The 
original clock board was put in a box in late 2015 with a CR2032 coin cell 
powering the DS3231. When I restarted the project in early 2019, the clock 
came up with under 1 minute off  Great chip and really long battery life

Back to the AC line, In the US, a few years ago there was a push to reduce 
the AC line accuracy based on the claimed cost to keep the line cycle count 
balanced. Not sure what happened to that proposal, anybody remember?

  -Other Bill-


On Thursday, June 20, 2019 at 10:57:16 AM UTC-7, Jeff Walton wrote:
>
> One trick for using line frequency is to rectify, and wave shape to get a 
> square wave.  Then set up a monostable one-shot with a period just shorter 
> than the line frequency (or doubled full wave).  This will prevent any 
> extraneous noise from getting through to the divider circuit that feeds the 
> clock logic.  I used a scope and superimposed noise on the input to help 
> fine-tune and verify the one-shot timing.  
>
> I built my first clock over 40 years ago with TTL logic using this method 
> and the accuracy was within seconds/year.  I didn't have temperature 
> compensated crystals or components back then.  But after many years, the 
> method has proven to be very reliable.  The US grid has been a good timing 
> source for a non-critical application. 
>
>
> Jeff Walton 
>
>  Original message 
> From: Paolo Cravero > 
> Date: 6/20/19 11:57 AM (GMT-06:00) 
> To: neoni...@googlegroups.com  
> Subject: Re: [neonixie-l] Presented for your consideration...The BRS Clock 
>
> Outstanding Bill! And it will be programmed in Assembly!
>
> Are you using some tricks to improve the reliability of "reading" AC line? 
> Either HW or SW. In Europe AC time is currently running 32 seconds late 
> since the beginning of the year. It will catch up, as it did in 2018 when 
> the grid had "lost" 300 seconds in 3 months. Every time I think my IN-14 
> clock has troubles with AC pulses, but in fact it's the grid running 
> noticeably fast or slow.
>
> Paolo
>
> On Wed, Jun 19, 2019 at 1:43 AM Bill Stanley  > wrote:
>
>>  
>>
>> I promised to post some details on my current Nixie nightmare. I present:
>>
>>
>> The BRS Clock
>>
>> The current version of the clock is in 2 (or more) PCBs. This was done to 
>> allow changes to the display type by only either changing the display board 
>> or the display cluster boards. The board can be used with Nixie tubes (10 
>> cathode or 14 cathode “British Flag” types), LEDs, possibly VF or other.
>>
>>
>> The CPU board:
>>
>> A pair of 16 pin headers on top of the CPU board connects it to the 
>> display. The CPU supplies 12V, 5V and 170V with the 170 able to be margined 
>> by the CPU. I2C connects “things” on the display to the CPU. The time data 
>> is sent in a serial format (clock+data) along with a transfer strobe to 
>> write the data to the tubes. For Nixie displays, the HV5812PJ-G chip does 
>> direct drive to the cathodes, there is no multiplexing. Additional signals 
>> include a room light sensor an 11 pins defined for control like decimal 
>> points, symbols and “pimp lights”.
>>
>>
>> The CPU is currently a Silicon Labs C8051F360
>>
>>
>> Time base. There a 4 possible time base sources for the clock. In 
>> increasing accuracy order:
>>
>> The CPU time base: This is the worst but with some work can be trimmed up 
>> somewhat. If this is the only time base, the clock

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[gregebert]

>
>
>
> Gregbert: Interested in the "genuine" qualifier and what is behind it. The 
> original clock board was put in a box in late 2015 with a CR2032 coin cell 
> powering the DS3231. When I restarted the project in early 2019, the clock 
> came up with under 1 minute off  Great chip and really long battery life
>
>>
>>
There are apparently some fake DS3231/DS3232 ICs out there that have 
timekeeping problems. So far, I have found 2 different bugs on different 
modules with the cheap RTC modules bough on Ebay.

   - Rollover problem on Dec 31: Date goes to Sept  02
   - Accuracy problem: Losing time at the rate of more than 1 minute per day

When I use units purchased from Digikey, I have no problems.

BTW, this chip averages about 1-2uA current consumption, then every 100 
seconds it runs an internal routine that takes a lot more than 50uA; the 
spike pegs my 50uA mechanical meter.


My first few nixie clocks use the AC-line for timing; I do see some 
day-to-day wandering, and in addition it seems to accumulate a few extra 
seconds per month. I have a pretty good low-pass filter for the 
timing-extraction, so it must be low-frequency noise spikes. All of my new 
clocks use a RasPi for automatic timing, so I have not bothered to 
trace-down the line-frequency stability.

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/b5e4317b-fefa-41f9-9940-f31b3eb08714%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[Terry Kennedy]

On Thursday, June 20, 2019 at 12:57:28 PM UTC-4, Paolo Cravero wrote:
>
> Are you using some tricks to improve the reliability of "reading" AC line? 
> Either HW or SW. In Europe AC time is currently running 32 seconds late 
> since the beginning of the year. It will catch up, as it did in 2018 when 
> the grid had "lost" 300 seconds in 3 months. Every time I think my IN-14 
> clock has troubles with AC pulses, but in fact it's the grid running 
> noticeably fast or slow.
>

Apparently this was (is?) due to the former Yugoslav states:

https://www.reuters.com/article/serbia-kosovo-energy/serbia-kosovo-power-grid-row-delays-european-clocks-idUSL5N1QP2FF
(fair use excerpt): 
*"Serbia, Kosovo power grid row delays European clocks*

*SARAJEVO, March 7 (Reuters) - European power grid lobby ENTSO-E urged 
Serbia and Kosovo to urgently resolve a dispute over their power grid, 
which has affected the broader European network, causing some digital 
clocks on the continent to lose time."*

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/5bf802ef-76f5-4b26-a8c2-a2eca765eb3f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [neonixie-l] Presented for your consideration...The BRS Clock

[Mac Doktor]

> On Jun 20, 2019, at 4:42 PM, Bill Stanley  wrote:
> 
> Jeff; For my clock line time base, I decided on a brute force filter. I use a 
> opto-isolator powered by the rectified AC line with the assumption that any 
> noise would need enough energy to light an LED via a large resistor to get 
> thru. Not impossible but a fairly high bar. Since that signal does go to a 
> counter/timer, I could create a software one-shot to add a bit more 
> protection, I will think about that (and there goes 5% more overrun on code)

Have you tried using a ZCD (zero-crossing detector) opto-coupler followed by a 
Schmitt trigger? I built a thyristor-based variable light controller that way 
years ago and it worked perfectly.


Terry Bowman, KA4HJH
"The Mac Doctor"

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to neonixie-l+unsubscr...@googlegroups.com.
To post to this group, send an email to neonixie-l@googlegroups.com.
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/17CFADC0-A0F8-43A3-9B3B-335532F65FC9%40gmail.com.
For more options, visit https://groups.google.com/d/optout.