#48 Crystal Set — "Mine is bigger than yours!
I'm a little ahead of schedule on my #48 radio. I hope to use this one in the 2005 crystal radio receiving contest. That is, unless I build another set like this. That can always happen.
Ok, here is the lowdown on this radio. The base is cut from the same size board as my #47 crystal set. This one I stained using a golden oak color stain. The board turned out beautiful. The front panel is styrene. I prefer to work with Garolite® but since it's rf properties are unknown to me, I felt I would play it safe using styrene. It does have a nice shine.
The capacitors came at a big price. Since very good capacitors are somewhat rare, others were in line at ebay to purchase them. I do apologize to the others I outbid. You sure made me pay up! The detector capacitor is one of those "holy grail" types that Fair Radio sold a few years ago. It has silver plated plates and ceramic insulators. Capacitors don't get better than this. The other one is a three gang type with ceramic stator insulators.
The triple gang is useful as it allowed me to have two tuning ranges, one for shorter antennas and one for the longer ones. This helps give a good balance between sensitivity and selectivity. If you don't have a capacitor like mine, you can get the same result by placing a fixed capacitor in series with the antenna. Some people have had better results with this input circuit by switching the antenna and ground leads. Give it a try and see if it works better at your house.
The detector is the "Hobbydyne" style.
Jim Frederick in Florida
thought of using a differential capacitor and rf choke to improve the selectivity
of a large litz crystal set. I now use the Hobbydyne detector in most of my sets.
This circuit does require the use of an audio matching transformer system as shown
on Darryl Boyd's
site and my own pages.
Perhaps the star of the show is the spider web coils using 660/46 litz wire. I started making my crystal sets with spider coils, even my first set. They work well for me and hold up well. They are easy to wind and now that I have gotten used to how a scroll saw works, they are easy to cut.
The antenna coil is on a 6-1/2 inch (165mm) with a 3 inch (76mm)hub. There are 32 turns of 660/46 litz wire wound on the form. 40 feet (12m) of wire is required to make this coil, including the end lengths to make the connections. This makes a 148 µh coil.
The detector coil is wound on a 7-1/2 inch (190mm) styrene form, also with a 3 inch center hub. This coil has 42 turns of 660/46 litz, producing an inductance of 240 µh. My little spreadsheet tells me that this coil uses about 58 feet of wire. Between the two coils, it looks like 100 feet (30m) of wire will do it. If your heart thumps too loudly at the price of this big litz, 165/46 or 40/44 will work with good results. Besides if your heart is making that much noise, you can't hear your radio anyway.
The coils on this set are much like the ones on my #35 set. By moving the two boards closer and further from each other, the coupling could be adjusted. This set also has adjustable coupling but it is done by rotating the antenna coil on an axis. This varies the coupling between the coils and can be adjusted for optimum coupling. As in #35, the tuning has to be touched up after adjusting the coil.
As of May 2004, I have made a slight improvement in this set. I added a SPDT switch to select between one and two coil operation. The one coil selection qualifies as a "hobby class" receiver in the crystal radio contest, while the two coil radio is an "open class" radio. One other feature is that by tuning in the single coil position acts as pre tuning when switching to two coils. I do like dual use type sets. The switch is one of my brass link types. The switch is located just above the diode select switch.
Here it is July 2004 and I found a slight error with my updated set. When I put in the switch to change the circuit to a one coil tuner, I neglected to provide a return path from the detector side back to ground. I did add a wire between the two points, but I found that there was too much audio hum. Even if I grounded the body of my audio transformer, the hum was too great. I added a capacitor between ground and the cold side of the detector circuit. A .001 still allowed hum but a 100 pf was low enough to kill the hum. There is still a good rf connection between the two halves. The top part is coupled only by 10 pf, so this is ten times the amount. You may experiment with this value.
To test different diodes and use the one that suits the conditions the best, I added a double pole rotary switch to select diodes. One position selects the front panel terminals and the other positions can be used for other diodes soldered to the switch.
The audio transformer is external with this set. Getting good transformers is difficult so I use one on many sets. I can also experiment with other transformers in the quest to find the perfect one with more than 100% efficiency.
This is probably the ultimate in a high performance crystal set. The best litz, the best capacitors, ceramic standoff insulators, it's all here.
As usual I made no drawings before building this set. That files in the face of the old adage "Measure twice, cut once". Somehow I lucked out and didn't ruin any parts, wood or styrene.
I made a change on this set. I added a Bogen T725 audio matching transformer. It was added on the back of the differential capacitor styrene mount. A small Alco rotary switch was added to change the taps. Now the #48 doesn't require an external transformer. The picture is next to the schematic, below. Details on the T725 can be seen on my audio matching transformer page. I won't update the schematic. Just add the transformer to the output.
Here is a very nice replica of this radio.
Finally I want to thank David Williams in the UK for sending me those very nice looking homemade thumb nuts. They put the nice finishing touch on my project.
73 and good crystal DX. Dave - N2DS