4
\$\begingroup\$

In this article the author dissects a radio-controlled clock that syncs with the NIST transmitter, WWVB, in Colorado. The radio receiver seems to be a single integrated unit:

WWVB Receiver

Of special note is that the author is able to get a binary signal out of the little Temic U4226B IC, which could presumably be processed by a microcontroller.

I have seen this (or something very similar) for sale at Sparkfun but they are no longer available. In particular, the ferrite-core wound antenna is common between the two setups.

What is required to receive the signal that WWVB broadcasts? In other words, what is the circuit doing to pull the signal out of the air and convert it to binary logic? In particular, what is special about the antenna (number of windings, materials, etc) that tune it to the 60kHz signal?

\$\endgroup\$
  • \$\begingroup\$ The European equivalent of WWVB is DCF77, broadcasting from Germany at 77.5 kHz. IIRC the coding of the time information is different between the two. \$\endgroup\$ – radagast Dec 6 '13 at 8:32
  • \$\begingroup\$ The amplifier section in Andy's answer made me wonder if there was a replacement for the obsolete ZN414 and it looks like the TA7642 is widely available. With a bit of experimentation that might be an option for the amp because they're quite simple to use and there's a ton of examples around. You could also use it to find the inductance of your coil by finding what capacitance was needed to tune to a local AM broadcast station (although an LC oscillator might be a more direct way). \$\endgroup\$ – PeterJ Dec 6 '13 at 12:27
2
\$\begingroup\$

There will be a broad range of values for the ferrite core and number of turns that would suit reception of a 60kHz transmission. Firstly, because it is only 60kHz, virtually any ferrite core material would do because at that frequency, the losses in most ferrite materials are going to be negligible.

Secondly, the ferrite core and winding will resonate with a tuning capacitor to form what is known as a tuned-circuit. The general idea here is that the tiny magnetic field from the 60kHz signal received is converted into a voltage (by the coil) and amplified many times by the action of a coil and capacitor. This gives you two things: -

  • A bigger signal to deal with (better of course)
  • Rejection of other signals that are out of the 60kHz band

In other words, signal is improved and noise is reduced. Here's what wiki says about this: -

AM broadcast radios (and some other receivers used at low frequencies) typically use small loop antennas; a variable capacitor connected across the loop forms a tuned circuit that tunes the radio's input section.

Because resonant frequency is dependent on inductance and capacitance there is a broad sway of coil configurations that are suitably tuned by different capacitances achieving the same (or nearly the same) performance but, in general, you would try and make the coil's inductance is high as possible to achieve higher Q. Q is how much the circuit resonates and therefore for a low bandwidth signal you re better with more rather than fewer turns on the ferrite core.

After the ferrite core there is a strong possibility that no further LC tuning is done and that the AM signal is amplified by simple IC's or transistors. The signal will then feed an envelope detector. This will produce a signal that "follows" the amplitude of the 60kHz carrier frequency - the data is encoded as changes of carrier amplitude and so the envelope detector produces an output voltage that is the data. This can be fed into a comparator circuit and then on to a microprocessor for digital decoding of the signal.

\$\endgroup\$
  • \$\begingroup\$ for something as low bandwidth as WWVB, I'd start with a resonant coil but follow that with a digital filter that is just wide enough to include data plus leave room for crystal frequency slop. That will greatly increase the signal to noise ratio because most of what comes out of even a well tuned L-C filter will still be out of band. \$\endgroup\$ – Olin Lathrop Dec 6 '13 at 14:18
  • \$\begingroup\$ Very informative! If you wanted to build an antenna resonant at 60kHz (or any other frequency for that matter), can you calculate the values for L and C, or is it a matter of trial and error to find a resonant configuration? (I seem to remember something about LC circuits from my college physics course but it wasn't covered very well...) \$\endgroup\$ – thirtythreeforty Dec 6 '13 at 18:14
  • 1
    \$\begingroup\$ Ferrite rods of the type in the picture (if specifed correctly) will be able to tell you how many turns you need to wind to get a specific inductance but, in the past I've just taken one from an old radio and used it with a 1nF fixed tuning capacitor and a 300pF trimmer and got this to work to receive the old 50kHz time clock in the UK. This isn't active any more. Trial and error is probably the best way and if you have a signal generator and oscilloscope this helps tremedously. \$\endgroup\$ – Andy aka Dec 6 '13 at 18:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.