Project "Chronos"
Conundrum, Fri Sept 14 2018, 09:43AMHi, I recently got interested in time and measurement, but ran into some problems with my lack of decent test equipment.
Found an atomic clock (Rb Efratom x100) on a certain auction site but alas its a non working unit.
It seems that they get problems with all sorts of components but if the lamp is really bad (low voltage) then its generally
kaput unless you have access to spare units.
I did look into un-mucking dead lamps using a laser but until now did not have any way to test my hypothesis about using
a blue laser (eg 450nm Osram) to zap the lamp surface and free the Rb atoms.
Has anyone else attempted anything like this?
The usual heat gun trick won't work except in very marginal specific cases where Rb has condensed where it should not
but the advantage with laser is focus can be tuned to +/- 50nm in the Y axis and lamp can be tested in situ to see if there is
any improvement/ partial lock/lock (yay!) etc.
Could probably publish this work on Hackaday if it succeeds.
Re: Project "Chronos"
Sulaiman, Fri Sept 14 2018, 03:35PM
I had an ex-Chinese telecom tower FE560A Rb oscillator
but never met anyone who has resurected a dead Rb tube
even though I read about it.
You could use a GPS-disciplined oscillator
or pick up and decode 77kHz from Frankfurt or 60kHz from Anthorn
or use the very strong 198kHz from Droitwich (typ. divide by 99 then divide by two if discrete logic used)
good stuff here
Sulaiman, Fri Sept 14 2018, 03:35PM
I had an ex-Chinese telecom tower FE560A Rb oscillator
but never met anyone who has resurected a dead Rb tube
even though I read about it.
You could use a GPS-disciplined oscillator
or pick up and decode 77kHz from Frankfurt or 60kHz from Anthorn
or use the very strong 198kHz from Droitwich (typ. divide by 99 then divide by two if discrete logic used)
good stuff here
Re: Project "Chronos"
Conundrum, Sun Sept 16 2018, 09:05AM
Yes, I also have an Anthorn clock.
Interestingly mine went through a phase of not locking which I (erroneously) blamed on a hardware problem.
Seems that heat is bad for these and in this case so is direct sunlight.
When you say *had*, what happened to it?
The unit I am looking at seems to have the low voltage issue so my laser idea might work but it would require
a lot of tinkering and experimentation. Also £100+ is a lot of cash to spend on an experiment.
Would need to sell a lot of my stuff to afford that.
-Andre
Conundrum, Sun Sept 16 2018, 09:05AM
Yes, I also have an Anthorn clock.
Interestingly mine went through a phase of not locking which I (erroneously) blamed on a hardware problem.
Seems that heat is bad for these and in this case so is direct sunlight.
When you say *had*, what happened to it?
The unit I am looking at seems to have the low voltage issue so my laser idea might work but it would require
a lot of tinkering and experimentation. Also £100+ is a lot of cash to spend on an experiment.
Would need to sell a lot of my stuff to afford that.
-Andre
Re: Project "Chronos"
Aard-Wolf, Mon Oct 08 2018, 09:51PM
As Sulaiman mentioned, you could use 198KHz, but there is a small amount of phase jitter due to the data modulated on the carrier. Also, at night, there are problems with signal fading due to varying mixtures of sky-wave/ground-wave signals. I have three 198KHz off-air references, a Quartzlock one, another which was resurrected from ebay (it was an old Practical Wireless design), and a third one from Spectrum Communications (kit).
I also have a DCF 77.5KHz off-air reference (another ebay resurrection), which looks as though it might have been from a laboratory. Even this unit is only at its best during day light hours around midday, as fading is also noticeable at night even even with DCF.
The best reference I've found within my budget is a GPS disciplined unit made by Leo Bodnar Electronics. They are very quick to phase lock, are very sensitive (you can throw the patch antenna down almost anywhere and it works), and they have user programmable output frequencies (anything from a few tens of Hertz up to 800MHZ or so).
I have a few 5MHz and 10MHz oven/double oven references, but these needed setting up against a known reference (in my case the GPSDO). Once set up they remain very stable, and they are fairly cheap to buy, but like I mentioned you need a reference to set them up with (catch 22?).
I did look at Rubidium references, but looking at the cost, even for second hand units possibly near the end of their useful life, I finally opted for the GPSDO. The Leo Bodnar one runs on 5V from the USB output of a PC or laptop. You only need a computer to set the output frequency you require (the default output is 10MHz), once it is set it will retains the settings you have chosen so you can run it off of any 5V power source.
Simpler ways you could try are to manually zero beat a reference to 77.5KHZ, 198KHz etc.
Also, a novel one is to use the 19KHz pilot tone transmitted by FM broadcast stations. Some appear to use quite stable references, and you can obtain the 19KHz from pretty well any cheapo FM receiver. It's something to think about anyhow...
Oh, nearly forgot, the oscillators in most digital/quartz wrist watches are pretty stable too! And very cheap!
For ease of use, accuracy and stability my money is still on the GPSDO
Hope some of that ramble makes sense.
Mark.
Aard-Wolf, Mon Oct 08 2018, 09:51PM
Conundrum wrote ...
Hi, I recently got interested in time and measurement, but ran into some problems with my lack of decent test equipment.
Just how accurate does this "test equipment" need to be?Hi, I recently got interested in time and measurement, but ran into some problems with my lack of decent test equipment.
As Sulaiman mentioned, you could use 198KHz, but there is a small amount of phase jitter due to the data modulated on the carrier. Also, at night, there are problems with signal fading due to varying mixtures of sky-wave/ground-wave signals. I have three 198KHz off-air references, a Quartzlock one, another which was resurrected from ebay (it was an old Practical Wireless design), and a third one from Spectrum Communications (kit).
I also have a DCF 77.5KHz off-air reference (another ebay resurrection), which looks as though it might have been from a laboratory. Even this unit is only at its best during day light hours around midday, as fading is also noticeable at night even even with DCF.
The best reference I've found within my budget is a GPS disciplined unit made by Leo Bodnar Electronics. They are very quick to phase lock, are very sensitive (you can throw the patch antenna down almost anywhere and it works), and they have user programmable output frequencies (anything from a few tens of Hertz up to 800MHZ or so).
I have a few 5MHz and 10MHz oven/double oven references, but these needed setting up against a known reference (in my case the GPSDO). Once set up they remain very stable, and they are fairly cheap to buy, but like I mentioned you need a reference to set them up with (catch 22?).
I did look at Rubidium references, but looking at the cost, even for second hand units possibly near the end of their useful life, I finally opted for the GPSDO. The Leo Bodnar one runs on 5V from the USB output of a PC or laptop. You only need a computer to set the output frequency you require (the default output is 10MHz), once it is set it will retains the settings you have chosen so you can run it off of any 5V power source.
Simpler ways you could try are to manually zero beat a reference to 77.5KHZ, 198KHz etc.
Also, a novel one is to use the 19KHz pilot tone transmitted by FM broadcast stations. Some appear to use quite stable references, and you can obtain the 19KHz from pretty well any cheapo FM receiver. It's something to think about anyhow...
Oh, nearly forgot, the oscillators in most digital/quartz wrist watches are pretty stable too! And very cheap!
For ease of use, accuracy and stability my money is still on the GPSDO
Hope some of that ramble makes sense.
Mark.
Re: Project "Chronos"
Conundrum, Wed Aug 07 2019, 06:35AM
Given that I am trying to measure 1/1M changes in time flow comparable to an altitude change of 10m, quite.
Conundrum, Wed Aug 07 2019, 06:35AM
Given that I am trying to measure 1/1M changes in time flow comparable to an altitude change of 10m, quite.
Print this page