Handmade Radio Components

Okay, so what electronics hobbyist hasn't considered building a crystal radio set at one time or another? I found myself with a chunk of free time during the summer of 2007 and decided that my time had come. I rooted through my old mineral collection and pulled out a few chunks of galena and iron pyrite and measured the voltage drop on several random signal diodes I pulled out of equipment. Supposedly germanium diodes are the best for crystal set work, namely the very classic 1n34. It has a voltage drop of about 0.4V as I recall, and in general germanium diodes will have a smaller forward voltage drop than silicon. However, the rectification diodes in a few computer power supplies that I've salvaged have an amazing 0.2V forward drop, so I've fooled around with using them as well. Also, it some "DX" crystal radios Schottky diodes are used.

Following directions from H. P. Friedrichs' book "The Voice of the Crystal", I made a high impedance earphone/speaker. There is a magnetically biased coil armature made from many individual strands of soft iron or steel wire (I'm not sure which, it was an old spool), and the diaphragm/sound box is a little steel tin that originally held packets of expensive tea. The coil was wound with wire taken from an old cheap guitar pickup that I bought for $3. The wire broke several times during the winding process, but I carefully soldered each connection and tested for continuity and 1kHz impedance as I went along. I am confident that there are no shorts. At 1kHz it has an impedance of 64.6k Ohms, at 120Hz 5.892k Ohms. I can clearly hear the 1kHz and 120Hz signals coming from it when testing. If I ground one terminal and hold the other in my hand, I can hear the 60Hz hum from the house electrical wiring.
A few years later (Feb 2012) I decided to rebuild this speaker with better magnets and some sort of padding for ease of use. I stole the rubber padding and cloth cover from an old set of Chinese sound powered phones. The new magnets are ceramic, and are much stronger than those originally used. I also cut down the long bolts that form the structural frame.
I don't know if I will ever put myself though winding another coil like this. I decided to remove the top piece of the winding bobbin and coat the exposed winding surface with epoxy as I needed a little more space in the rebuilt design. This shows the wires before I added the epoxy The center tube is made from a bic pen casing, so you can get an idea of the tedious thinness of the wire.
This is the underside of the diaphragm. No attempt has been made to isolate the sounding surface from the rest of the sound box/tea tin. The two support bolts float freely in the holes in the diaphragm, and their heads are soldered to the lid (not shown).
This shows the critical gap. This should be as small as possible without contact between the steel diaphragm and the three upright magnetic cores.
This shows the magnets and the arrangement of the speaker driver core. The fine iron wires that make up the core are cut and spread such that the center leg is the magnetic north pole, and the two outer legs route the south pole around and up from the back. This is exactly what is recommended in H.P. Friedrich's book, "The Voice of the Crystal."

#26 Gauge magnet wire "Basket" Coil. Unknown Inductance. Later used in a simple Regenerative Set

This is a classic cat's whisker radio detector. The structure is essentially the same as a point-contact diode, where a sharp thin point of metal lightly touches a piece of crystalline semiconductor. The PN junction forms at the point of contact, and devices of this type can perform useful functions into the range of 100MHz or more.		The diode assembly is shown connected to a simple LC tank. Add a ground and an antenna and you have a radio.
The crystal material I am using in these images is iron pyrite. The whisker is brass, I believe. It was taken from a piece of brass mesh metal screen.		Other minerals that may be used in place of iron pyrite include chalcopyrite, zincite, carborundum, galena, and various oxides of metal. Most rusty objects are just loaded with active semiconductor sites.