General > General Technical Chat
Once upon a time in the '70'S in WirelessWorld...........
<< < (4/6) > >>
Gyro:
I was looking through the archive a while back looking for an electrostatic headphone design article that I remembered, it seemed a nicely quirky design at the time - drilled perspex sheets, aquadag coatings, clingfilm diaphrams etc complete with HV push-pull drivers and bias supplies. Looking back further I found that there had been no less than three such designs over the years!

They had some great innovative designs and articles over the years, including things like Fleming reminiscing about his triode valve.
MarkMLl:

--- Quote from: StillTrying on January 30, 2020, 07:53:36 pm ---Why buy a £3 module when you could build a monstrosity instead.

--- End quote ---

I suspect that with a design like that it would be easier to extract the phase of the 60kHz signal and to characterise the delay before the start-of-second edge.

Commercially produced modules ostensibly using the MAS6180C chip are very likely (I find) to be actually using the MAS6180C1, which doesn't expose the carrier (QOM pin n/c). I've not tried probing a radio-corrected quartz movement yet.

MarkMLl
Ian.M:
There's also the issue that the MSF fast (100 baud) timecode was discontinued in 1998 [ref.], so even if you did build it, it would do sweet F.A. unless you also built a modulated 60KHz timecode generator to sit beside it!

However, the receiver circuit on page 2 of the PDF StillTrying linked should still be viable, just take its output from IC4 and  decode the slow (1 baud)  code with an Arduino or similar.  Googling MSF decoder sketch finds plenty of people that have done the software side of it, or if you want to roll your own, the signal specs can be found [here].

If you are a period purist, a minimal Z80 system could be used in place of the Arduino.
StillTrying:

--- Quote from: MarkMLl on February 06, 2020, 09:42:51 pm ---
--- Quote from: StillTrying on January 30, 2020, 07:53:36 pm ---Why buy a £3 module when you could build a monstrosity instead.

--- End quote ---

I suspect that with a design like that it would be easier to extract the phase of the 60kHz signal and to characterise the delay before the start-of-second edge.
--- End quote ---

I would think finding the exact point of the seconds edge would still be difficult because the received 60kHz ends up looking like the last red trace.
http://xtrsystems.com/vlf/wwvb_test.html


www.burningimage.net/clock/sensitive-60khz-receiver/
MarkMLl:

--- Quote from: StillTrying on February 07, 2020, 12:09:16 pm ---I would think finding the exact point of the seconds edge would still be difficult because the received 60kHz ends up looking like the last red trace.
http://xtrsystems.com/vlf/wwvb_test.html
www.burningimage.net/clock/sensitive-60khz-receiver/

--- End quote ---

My recollection from the time when various radiocode clock designs were published is that the phase of the MSF signal was significant. The NPL themselves say that the UTC second marker "is transmitted with an accuracy better than +-1 ms.", but I'd expect the edge precision to be fairly constant for each second.

The 60 kHz signal is intended to be a frequency standard, although I'm not sure how the NPL would describe it: probably tertiary. By counting the carrier- at least while idling- it should be possible to derive a high-precision 1 Hz signal. And in principle it should be possible to characterise the Tx modulation performance to derive high-accuracy UTC seconds.

MarkMLl
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod