TDF demodulation circuit

[toto83]

Heelo there, this is my first post on the forum, I saw LOTS of project about the good ol' DCF-77 german transmitter but i found no project on the TDF (former France Inter) transmitter of Allouis which has WAY MORE power than the german one. The only problem is that contrary to the DCF-77, the TDF signal is phase-modulated and does not have the same logic levels.
This is why I am asking for help : I successfully created the antenna and i could get the signal but that's all. Anyone could help me creating the decoder.

Thanks,
Tom.

[Benta]

Did you find this already?

https://en.wikipedia.org/wiki/TDF_time_signal

[toto83]

Yes but my knowledge about PLL is really poor and I don't know how to design the receiver.

[Benta]

The datasheet on this device will show you how such receivers are structured (click on the PDF symbol, datasheet is in English):

http://www.hkw-shop.de/Empfangstechnik-AM/Zeitzeichen-Empfaenger-IC-UE6015-DIT.html

It will NOT work with TDF directly, and the demodulator will have to be changed.

[Benta]

Oh! Here you go:

https://www.elektormagazine.com/labs/low-cost-frequency-standard-disciplined-by-france-inter

If you view the thumbnail for the schematic, you can actually read it :-)

Otherwise you'll need a subscription.

[babysitter]

Would you mind to share your antenna design? I am trying to get a DSP expert into making some code for different LW timing stations for ARM Cortex-M (possibly parallel reception from several sites at once.)

[StillTrying]

Yes but my knowledge about PLL is really poor and I don't know how to design the receiver.

Here's some reading then 
CMOS Phase-Locked-Loop Applications (Rev. B) - Texas Instruments
www.ti.com/lit/an/scha003b/scha003b.pdf

If your 162khz is larger than about 50mV peak to peak you should be able to apply it direct to a 4046.
If you can design the 4046 to have a small frequency range centered on 162khz, then when it's locked on to the 162khz you should see a low amplitude version of the triangle waveform from the wiki on it's VCO/filter.
A self-leveling comparator would turn the triangle wave back into square pulses.
In theory.

[toto83]

I found a 10368Khz crystal ! (64*162Khz) is it possible to divide the frequency by 64 to do the PLL Stuff? I dont know what a 4046 is or how to use it. I think I also need a IF Transformer. I will publish my antenna design.
Thanks,
tom

[JPortici]

Oh! Here you go:

https://www.elektormagazine.com/labs/low-cost-frequency-standard-disciplined-by-france-inter

If you view the thumbnail for the schematic, you can actually read it :-)

Otherwise you'll need a subscription.

and the descent in the timenut dungeon carries on...

[toto83]

Oh! Here you go:

https://www.elektormagazine.com/labs/low-cost-frequency-standard-disciplined-by-france-inter

If you view the thumbnail for the schematic, you can actually read it :-)

Otherwise you'll need a subscription.

and the descent in the timenut dungeon carries on...

What do you mean?

[JPortici]

https://www.eevblog.com/forum/chat/test-equipment-anonymous-(tea)-group-therapy-thread/

[mark03]

Would you mind to share your antenna design? I am trying to get a DSP expert into making some code for different LW timing stations for ARM Cortex-M (possibly parallel reception from several sites at once.)

   Yes, this is the way to go.  A basic front end followed by ADC and everything else in software.  I have the requisite knowledge and have been itching to try it for a while now.  Only problem is lack of spare time...

Does anyone know where I can pick up a 60-kHz front end with loopstick antenna, amplification, etc, where all I have to do is digitize the output?  That would help.

[Benta]

JPortici, your condescending attitude here is not well received by me.
I find the project interesting.
LF AM reception is not on the curriculum nowadays, but in no way complicated.
Concerning the 4046: I've never had success using the datasheet application examples. In my opinion the datasheet is plain wrong. However, when using an active loop filter, it's a dream to work with and locks quickly and stably at 0 deg phase difference.

[babysitter]

Would you mind to share your antenna design? I am trying to get a DSP expert into making some code for different LW timing stations for ARM Cortex-M (possibly parallel reception from several sites at once.)

   Yes, this is the way to go.  A basic front end followed by ADC and everything else in software.  I have the requisite knowledge and have been itching to try it for a while now.  Only problem is lack of spare time...

Does anyone know where I can pick up a 60-kHz front end with loopstick antenna, amplification, etc, where all I have to do is digitize the output?  That would help.

From our group if we really go on. I intended to sample a wide range from some shady russian low frequency stations over the europeans including our beloved DCF77, maybe having a look at CHAYKA and LORAN family up to at least TDF. First level of candy would be to actually use the phase encoded signals and carrier phase to predict the time mark really good, next layer of candy would be using correllators (if one time mark is found its almost predictable where another one would appear) and the third layer would be giving statistics about each second including the TOA difference, turning it into a nav receiver.

But I will stick between the antenna and the ADC, you will be glad to get software which is not made by me.

[JPortici]

JPortici, your condescending attitude here is not well received by me.

too bad for you, i was joking

I find the project interesting.

me too, it made for a very insteresting read and it could very well be on my to-do list. i never did much with RF and i think i need a frequency standard to check how well clocks are calibrated in my home lab.... (and then i'll certainly need another better one one to check if the DIY standard is accurate.... and then... do you get the GAS/TEA joke now?)

[Benta]

The OEM vapor barrier is actually an integral water control part of the door. Like several other places, Mazda didn't build doors to stay dry, but to channel water from above, through safe channels, and out drains at the bottom.

Yes, thank you. The problem is, subtle jokes are not well understood in writing, only in personal communication. On a forum like this, it has to be really explicit, also keeping in mind how many non-native English speakers are here.

[toto83]

 
Obvious joke is obvious. now how do i do to tune an IF transformer to right frequency? i ve been strugling with a 433khz IF transformer.
the demodulation circuit is practically done.
Will send asap

[StillTrying]

How accurate do you all think the second marker is from these LF time signals?

The UK's 60khz only says +/- 1ms. You could make an adjustment for distance, but I can imagine when you take into account the tuned circuits and filters at both the TX and RX, and then demodulation, that it's impossible to determine exactly when the carrier changes amplitude or phase.

http://www.leapsecond.com/pages/sony-wwvb

[toto83]

Actually 0.3ms so the solution is to feed the phase comparator with the unmodulated carrier frequency (generated with a crystal oscillator, so constant phase) -> ADC-> rising edge detector-> time signal decoded

[toto83]

look at this image we can see first the signal from the PLL, two the signal out of the DAC, three the signal after rising edge detector (the signal we want) and fourth is a sequencing signal every 200ms based on the 59th second (which is not modulated, we can see it at the begining of the first signal). So I'm about to build this thing and tell you the results.