Rb oscillators

[Grahame]

Chuck

I'm happy to help you...

Start here and come back to me with any questions you have.

http://www.sgitheach.org.uk/atomic.html

Grahame

[gregebert]

Grahame - Have you been able to detect any drift/inaccuracy of the Rubidium timebase ? I would imagine that even comparing to GPS it would take months, and perhaps years, to notice any drift.

I was really surprised to see that a DS3232 in one of my devices worked out to be 6e-7 after 6 years.

[Mac Doktor]

On Apr 19, 2021, at 2:53 PM, 'Grahame' via neonixie-l <neoni...@googlegroups.com> wrote:

Start here and come back to me with any questions you have.

http://www.sgitheach.org.uk/atomic.html

I first looked at this several years ago and seeing it again I have to say it's as beautiful as ever.

I know you've been asked a thousand times already...but what was the total cost of this project? I realize that some prices may have gone up slightly since then. The Dekatrons if nothing else.

Terry Bowman, KA4HJH
"The Mac Doctor"

https://www.astarcloseup.com/

“...the book said something astonishing, a very big thought. The stars, it said, were suns but very far away. The Sun was a star but close up.”—Carl Sagan, "The Backbone Of Night", Cosmos, 1980

[Grahame]

Hi Terry

I never added up the cost. It was mostly built out of the junk box and used home made PCBs. The expensive parts are the Rb oscillator (off ebay), the case (from Farnell/Newark) and the custom cut and engraved front panel. I would estimate £500 perhaps?

I have pulled the existing clock apart and I am getting ready to rebuild it:

The previous clock allowed comparison between the Rb oscillator and UK mains frequency (a jumper selects 60Hz operation). The next clock will have a number of different clock ticks measured and will include a GPS module and a DS3232 for example.

We moved house a couple of weeks ago so progress will not happen until the house is straight and the workshop rebuilt... The clock will be documented - all open design and open source. I should have some PCBs available as well.

All good fun.

Grahame

GM4EKI

--
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+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/98C7FC21-BF9F-4AE0-BC1B-8D5A6FE155E5%40gmail.com.

[Grahame]

Hi

Against NTP, I noticed no measurable drift over the years that the clock ran for. It did have problems with the back up PSU which made the clock unreliable if it was powered down for too long. I never compared with GPS. The main comparison was against UK mains stability. In the short term there was noticeable differences but UK mains frequency is manipulated so in the long term it is dead on 50Hz so synchronous  clocks stay accurate.

The DS3232 is a fantastic chip and my choice also for clocks not linked to NTP or GPS.

Grahame

--
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+...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/25697d78-b1ae-47a7-9de5-ba3ba3540e24n%40googlegroups.com.

[Kevin A.]

On the subject of stable free running time bases, for an extra $10+ over the DS3232 (2 ppm) you could use a DOT050V-010.0M VCTCXO as a clock source to get 50 ppb stability. Just trim out the voltage control (with an 18+ bit dac) to be bang on 10 MHz against a good counter. 

It is a temperature compensated crystal, so "small" footprint compared to big oversized oscillators that get you into the single digit ppb for much more cost, size, power. Diminishing returns and all that... 

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/28c73729-c5c6-d89c-c300-43055a663d5b%40googlemail.com.

[Kevin A.]

**Ovenized**, not 'oversized'... or are those the same?