The HamBone

Posted: **Wed Jan 29, 2025 11:16 pm**

I had some fun email with KC4NYK about the best diode for a crystal set.

So, I blew the dust off some old memories and found some new devices to recommend;

The article shows why the OA90 series, the old winner generally thought to be OA91, are popular in crystal sets -very low forward voltage.

http://qrp.gr/ultimatextalset/

But there are many others; maybe someone can tell us about the Marconi detector?

A crystal set tuned circuit, ideally, is a high-Q LC tank operating with available antenna-provided low voltages to provide RF to a simple rectifier demodulator. This rectifier needs to be a lowest forward voltage type, or there is insufficient half-cycle RF to make it conduct to provide an audible output.

Can you forward bias a Si diode? Yes, we bias Si varactors every day to make them work, however the joy of the venerable crystal is its independence from external power? A further design consideration is how to maintain high Q of a L-varactor tank, or LC-to-lower impedance biased Si diode. Obviously can be done, but difficult without violating a KISS principle. (A pair of biased Si diodes = a bipolar transistor. Works well, I'm told(!), but needs power to overcome two forward voltages.) A junk-box solution might be a Schottky diode, typical forward voltage 0.3V vs 0.2 for the Ge.

Alternatives are legion: cat's whisker - crystal, and double crystal types. Seems to be much folklore regarding which is best for radio reception, so some crystal radios are designed with switchable multiple diode types. The third example below allows one to choose between two Schottky types, Germanium, or even a MOSFET.... The impedance properties of the MOSFET also improve the tuned circuit's selectivity....

Refs:

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

https://antena.fe.uni-lj.si/literatura/ ... tector.pdf

https://www.aliexpress.us/item/3256807572215972.html

http://crystal-radio.weebly.com/crystal ... sfets.html

Hope it helps! Enjoy the Rabbit Hole...

Posted: **Thu Jan 30, 2025 4:17 pm**

Thanks John! Yep, I'm eagerly awaiting my "official" 1N36 and 1N3666 germaniums as we speak from Surplus-Electronics. If those don't work I pop for the $8 ones ( that's EACH) on United Nuclear's site. Although the very terms "united" and "nuclear" in a business name gives me pause after watching "Severance", lol! And thanks for the links to all the sources.

As I mentioned, Mike, WU2D, said that I could forward bias a wide range of diodes to make them function as rectifiers but I'd lose the panache and corresponding bragging rights to claim "unpowered" radio. That's not as important right now as my being able to evaluate various coils, variable inductors, variable capacitors and associated relationships for my project. Once I understand those parameters I plan to provide a variety of detectors as that's part of the learning platform that this project represents. Kind of like an RF erector set, hihi!

SteamUgly_2.jpg

Thanks again for your guidance. Now let's see if I can actually post this reply, hehe.

Cheers,

rob, KC4NYK

Posted: **Mon Feb 03, 2025 12:05 pm**

Much to think about. Thanks rob, KC4NYK.

Do you suppose we could increase gain 3dB with maybe some of these diode devices, even a MOSFET pair?

Take a gander at viewtopic.php?t=160 for inspiration?

Posted: **Fri Feb 07, 2025 6:46 pm**

Here's a quality xtal set in wood case.
MW / SW, switchable xtals.... OA91, point contact, Si, MOSFET....

xtalradio.jpg

More details are at https://www.aliexpress.us/item/3256806991812035.html

Posted: **Fri Mar 14, 2025 5:23 pm**

Here's a novel thought for Xtal sets:
A FET acting as a half-wave switch detector, with minimal gate current from the LC tuner, so preserving Q, and providing a better headphone impedance match too.

fetdet.jpg

The full article describing this is at:
https://www.creative-science.org.uk/epadmosfets.html

This irritatingly simple improvement is hard to find on a web search, but is really not new.

Or you just try one of the old "gold wire on Ge" diodes, als not easy to find, but ideal of a crystal set -lowest forward voltage and steep slope.
Here's a slightly more advanced circuit, on the same theme:

fet_xtal1.jpg

And the full account to go with it:
https://www.diy-electronic-projects.com ... al-Radio-

Well, hows that for thought? A FET to preserve Q and increase output volume, all in one component.

What's your best detector?

The HamBone

The best diode type for a crystal set?

The best diode type for a crystal set?

Re: The best diode type for a crystal set?

Re: The best diode type for a crystal set?

Re: The best diode type for a crystal set?

Re: The best diode type for a crystal set? -FET no battery?