I had a young neighbor boy ask me if he could learn a little about electronics and build a project. We started to have a small class on Monday nights. So we worked through the summer learning about electronics. I decided it would be good ideal to build a crystal radio. We used a schematic out of the December 1997 QST. It was developed in the early 30's by Elmer G. Osterhoudt. Selectivity problems plagued most crystal sets of the day due to the proliferation of broadcast stations. Elmer devised a solution. He determined that poor crystal receiver selectivity was caused by the headphones swamping the tuned circuit and lower its Q. The net result was an inability to separate individual signals. Elmer improved selectivity considerably by connecting the detector near the cold (ground) end of the tuned circuit to minimize the swamping.
Here is a picture of the schematic. L1 is wound on the outside of the rear portion of a outlet box enclosure and consists of 32 turns of 22-gauge enamel covered wire. The coil is tapped nine turns from the grounded end. The tap is made by tightly twisting about 1 inch of the magnet wire and scraping off the enamel insulation. You can secure the windings to the enclosure with transparent plastic tape. I used two different sizes of wire because that is all I had in the shop. They both worked equally well. Here is a couple of pictures of the complete project.