PLL stabilized Q-multiplier

On April 6 in the thread "WWVB Receiver Chip" Joreg suggested using a PLL stabilized Q-multiplier instead of a crystal filter in a 60 kHz receiver. Joreg could you, or someone, explain what a "PLL stabilized Q-multiplier" is? Maybe show a block diagram?

Reply to
garyr
Loading thread data ...

A Q-multplier is basically an oscillator where you turn down gain so that it just barely isn't operating.

I'm guessing that Joerg wants to use PLL techniques to keep the Q-multiplier accurately tuned, maybe with a varicap as a tuning element. Varicaps don't have a lot of capacitance but seeing how narrow the tuning range will be it probably would be good enough. Maybe he dynamically wants to keep the gain tuned too.

I'm somewhat dubious of using a Q-multiplier in a circuit without a "passive" bandpass in front of it, but it's possible that a tuned loop antenna is a sharp enough bandpass. I'm on the East coast where WWVB is usually completely dominated over by man-made noise.

I'm used to "Q-multiplier" referring to IF frequencies and "regenerative receiver" used at the original signal frequency but I suppose the principle is the same.

Tim.

Reply to
Tim Shoppa

...

Just a nitpick, but they usually run it at IF so that they don't have to have such a wide tuning range.

Cheers! Rich

Reply to
Rich Grise

My impression is that QRM in the 60kHz region has gotten a lot worse in the past 20 years. Of course in the same time I moved from a small farm in the midwest to the East Coast. I don't know what all the sources are but noise is generally worse during the day than at night so I think it's industrial machinery.

Of course, WWVB increased their output power and antenna array a couple of years ago too, this must've helped some.

Tim.

Reply to
Tim Shoppa

Hmm, could be like a WWVB repeater/booster :-).

Tim.

Reply to
Tim Shoppa

Traditionally, they were at the IF, since at higher frequencies they'd not offer much in selectivity.

But they did appear in front ends in some amateur designs in the sixties, when 9MHz crystal filters appeared and someone wanted to feed the antenna into the mixer without an rf stage. With the minimal front end filtering, someone decided that a Q-multiplier would help.

Of course, at 60KHz the center frequency is low enough that an LC circuit is narrow enough and the Q-multiplier is just a boost.

I'd say the difference between the Q-multiplier and the regen receiver is that the former is just acting on an LC circuit, while the latter includes the detector. Yes, the former is using regeneration, but a regenerative receiver is a bit more.

A quarter century ago when Ralph Burhan was describing various Loran C receivers for use for frequency standard work, I'm sure he used a Q-multiplier in at least one of the published designs. In one, he'd use an off the shelf AM/FM receiver IC, regenerate one of the IF stages for the front end and use the FM detector as a "synchronous detector" (using the limiter to square up the incoming carrier, and the balanced modulator part of the quadrature detector as the mixer, combining the incoming signal with that locally generated "carrier" to get down to baseband).

All this talk about crystal filters and ICs, yet thirty years ago there was the occasional article in the hobby magazines for a WWVB receivers, all the ones I saw used LC circuits for filtering.

Michael

Reply to
Michael Black

Hello Gary,

Tim already explained it pretty well. I don't have a schematic at hand but there are some in older ARRL Handbooks. The last Q-multiplier I built was at least 20 years ago.

Like Tim said, it is basically an oscillator where the feedback is set so it just barely doesn't oscillate. How close you are to that point will determine the Q and thus the bandwidth of the connected LC circuit. Mostly a Q-muliplier was realized by bringing the first IF stage really close to the point of oscillating. In your case it would likely be one of the 60kHz stages since you probably won't have an IF.

PLL control is needed to keep a very high Q LC-circuit at the desired frequency. For that you have to bring the Q-multiplier stage into actual oscillation once in a while, else the PLL divider would have no input signal. With a signal such as WWVB that's no big deal since you don't need to receive all the time.

Nowadays I'd use crystals, or at least a lower frequency IF stage. Switched capacitor filters can make really nice filters at a few kHz but the required chips cost more than a crystal. If you try the Q-multiplier make sure nothing gets radiated out when it oscillates.

Regards, Joerg

formatting link

Reply to
Joerg

Hello Tim,

Which could get people charged as "NIST impersonators" :-)

Regards, Joerg

formatting link

Reply to
Joerg

hand

set

circuit.

really

one

actual

What sort of Q would you expect to be able to obtain with this technique?

but

Q-multiplier

Reply to
garyr

Hello Gary,

That depends on the frequency and how good the LC circuit itself is constructed. It's been a very long time but I was able to get the 3dB bandwidth of a 455kHz stage down to under 500Hz, which would be a Q of around 1000. This was done by hand though, using a top quality variable capacitor on a ceramic plate and keeping the ball bearings nicely lubed. It also had a mil-spec vernier drive. The IF stage and the Q-multiplier were built around tubes. Nowadays FETs would be fine, too.

At 60kHz you can most likely obtain better Q but I doubt that you'd be able to achieve anything close to what a crystal could do.

Regards, Joerg

formatting link

Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.