WWVB preamp help

[jmw]

I tried making a preamplifier stage and antenna to see if I can receive WWVB. If I can't see it on my scope, there's no point in trying to demodulate it. I made a loopstick antenna using a ferrite rod, 30 AWG magnet wire, and a silver mica 4700 pF cap. The preamplifier is almost a clone of the John Magliacane design from QEX: an instrumentation amplifier followed by an active bandpass filter. He uses a AD620 in-amp, and I chose the INA849 since it's newer and has an edge on noise specs. I also added a buffer op-amp at the end so it can drive a 50 ohm load (my spectrum analyzer).

The antenna and receiver performance check out independently, but I haven't been able to see the signal. I'm either seeing amplified noise in the envelope of the bandpass filter, or a spurious carrier at just under 60 kHz - is that a sign of some interfering switcher nearby? I've tried testing at two locations at different times of day and night, but both are in buildings with neighbors. The pandemic makes it a little hard to call up a friend and ask to come over with gear. Anything else I should try before giving up and putting this on the shelf for a bit?

Some screenshots of the preamp performance to follow...

[TimFox]

Check the website:  WWVB started an upgrade on March 9 that requires occasional lower power periods until March 31.

[jmw]

Schematic: the in-amp resistor sets a 40 dB gain, and the bandpass filter Rs & Cs are set to deliver 20 dB gain at a center of 60 kHz with a Q of 15.



Antenna performance looks good, Q of about 150. (click to expand)



Preamplifier tested by running the TG through an injection transformer, connected to the antenna inputs. There's 20 dB of antennuation on the TG output, so it's actually at -50 dBm. There's a bit over 60 dB of gain at 60 kHz, as designed.



Receiver output with the antenna - either just noise or some spurious signal that's not modulated. Is this interference from a switcher?

[jpanhalt]

Here's the receiver I built using different op-amps.  Worked in Cleveland area even during the day: https://www.burningimage.net/clock/sensitive-60khz-receiver

About half of the 2N3819's I had worked.

[Hexley]

There are some relevant posts in Ed Nisely's blog. He gives details for a preamplifier and an antenna that might be helpful for comparison. See those links in the article at https://softsolder.com/2017/01/31/wwvb-receiver-first-light/

[jmw]

Well, Ed's preamp design (https://softsolder.com/2017/01/23/60-khz-preamp-first-pass/) is almost identical to mine, up to some choices of components. I'm not sure what else to try, maybe buy a commercial loopstick antenna tuned for 60 kHz to compare to my own, or rent an AirBnB in the sticks and bring my gear...

[jpanhalt]

I used a shield loop design made with about 100 ft of coax.  Signal strength was strong at all times.  During the day, I would occasionally lose it.  At night, it was steady.

Tuned it with an old-fashioned aircore variable capacitor.  The final antenna and tuning box are shown in attachments.  It is difficult to insert snippings here, so sorry for the size images.

[jmw]

Determined that the spurious signals were caused by feedback, not interference. The buffer can deliver enough current into a 50 ohm load to be picked up by the antenna. Seems fixable in the next prototype, but I'm still not receiving anything on 60 kHz. 

[bob91343]

I would suspect the antenna.  Unless you are intercepting a decent amount of signal, you  won't see much.  Try a remote antenna, a larger antenna aperture, stuff like that.

[cdev]

John Ackerman had a good idea. He uses a coil of coax tuned with capacitors to resonate at the WWVB frequency with a variable capacitor.https://www.febo.com/time-freq/wwvb/antenna/index.html

[jpanhalt]

My apologies, I usually give references and didn't in this case.

Here's a link that describes the antenna I made in Post #7.

https://www.febo.com/time-freq/wwvb/antenna/

John

[Co6aka]

Schematic: the in-amp resistor sets a 40 dB gain, and the bandpass filter Rs & Cs are set to deliver 20 dB gain at a center of 60 kHz with a Q of 15.

Not the problem, but shouldn't the BUF634A be in the OPA192's feedback loop? 

[jmw]

Not the problem, but shouldn't the BUF634A be in the OPA192's feedback loop?

I don't think so - it's a unity gain buffer and just there to so the 50 ohm spectrum analyzer doesn't load the OPA-192 output too much. But it supplies so much current with a 12V power rail, it can couple to the antenna and start a positive feedback loop. I may not keep it in the design once I no longer need to see things on a spectrum analyzer, but if I do I'll need to add some gain control and/or large capacitance to the buffer's power rail.

[Co6aka]

it's a unity gain buffer and just there to so the 50 ohm spectrum analyzer doesn't load the OPA-192 output too much.

...which is why I'd expect it to be in the OpAmp's feedback loop, if one is seeking the best overall linearity. I've built discrete buffers and have always included them for this reason, so that's why I'm wondering if there's a reason not to. For example, from the datasheet:

[jmw]

Oh, now I understand! That makes total sense now. As I said, I'm not sure if the buffer will stick around as part of this, but if it does, I will make this change in the topology, thank you!