CRYSTAL SET 2007 By far the best performer that I've built and easiest to use with a calibrated dial. Previous dual-tuned sets were almost impossible to get the two resonant circuits set to the same frequency if there wasn't a strong station to tune up on. With this one you start by shorting out the antenna tuner by setting the capacitance and inductance to zero. Then tune the main dial until you hear a [faint] station. You can then peak it up using using the antenna tuner and the various controls on the receiver. I calibrated the dial (after picking the trimmer and padder capacitors to get the fequencies for each band spread across most of the dial) by sticking a Post-it over the capacitor shaft (cut out for the shaft first) and marking every 100KHz with a pencil using a signal generator as a source. Use the lowest coupling for the antena and detector for the best accuracy.  I then read off the angles using a protractor and created the dial and the rest of the front panel layout  in Corel Draw. Any of a number of drawing prorams would work. You can also scan in the Post-it to a layer by itself and use it directly for guidelines. The optional MOSFET detector is a noticeable improvement when using lower impedance phones and the low-Z detector tap but doesn't do much for a crystal earpiecs. The set uses a variation on the "contra wound" coil  [seen on schmarder.com and also on Ben Tongue's site] on a torroid.  I've seen several torroids wound half in one direction and then at the midpoint the wire is crossed over and through the core and the windig continued on from there eventually ending back near the crossover. This technique is said to reduce capacitance between the start and end of the winding. I've also learned on previous sets that just tapping down on the coil for a lower inducance for the higher frequency part of the band doesn't always work very well as the unused part of the coil still contributes losses. So breaking the winding at the midpoint and using the two halves in series for the lower part of the band and in parallel for the upper part makes sense and seems to work well. Due to variations in the ferrite the two windings may not have exactly the same inductance but that doesn't seem to be a problem as long as they have the same number of turnes. PDF schematic and notes here

ANTENNA TUNER ON TOP OF CRYSTAL SET ( Love those Cigar boxes!)