This is an A.M. autodyne receiver packed around a single JFET. The operation is fairly simple: The 1N34 input diodes
mix the incoming R.F. and local oscillator signals to produce a 455kHz sideband, which is isolated by the I.F. coil and
470pF capacitor. The output is then rectified and filtered to produce audio. Although lacking in features, it's an easy
circuit which performs well, picking up most local stations to some degree.
It's important to do the calculations and coil winding as accurately as possible. If done right, you should have a
nice little reciever to experiment with. Note: nearby switching power supplies, such as those in computers, may
prevent the circuit from working properly.
The radio was designed to be used with a dual-ganged tuning cap (scrapped from old
tube radio) and some hand-wound coils. C1 is the larger side of the tuning cap, and C2 is the
Turn C1 clockwise about 10% and measure the value, say 390pF. The broadcast band ratio is 1650/530 = 3.11.
To find the lower end of the dial, do 390 / 3.112. This results in 40pF.
The oscillator must track 455kHz higher than the broadcast band, and operates at 985 - 2105kHz. Measure C2 with
C1 set at 390pF and 40pF. My values were 160pF and 16pF. The ratio is SQRT(160 / 16) = 3.16. But the oscillator
band ratio is 2105 / 985 = 2.14. A tracking capacitor must be added to C2 so that the ratio matches 2.14.
It is found with (160 + 16) / 2.142 - 16 = 22pF. In this circuit, the tracking capacitor also serves
to couple the oscillator to the JFET gate.
The antenna will be (159150 / 530)2 / 390, or 231.2uH .
The oscillator value is (159150 / 985)2 / (160 + 22), or 143.4uH. The I.F. coil value is
(159150 / 455)2 / 470, or 260.3uH. Now the coils may be wound to the appropriate values. (There are
many online calculators available for this purpose.) It's better to make coils with similar width and height,
i.e. don't make long skinny coils.