Here is that inductor being tested with a 150 foot long wire. Air coils in the background added to the adjustment. The capacitor is bypassed.

This is the new form, rather than the fiberglass sticks above.  Form is 4 inches by four inches and will be used in an elliptical band pass filter for 2200 meters.

End mill to the right, laser level on the left for alignment. I can “groove” four of the above forms at the same time.

Form wound with simple #20 hookup wire. This one is about 12uH, Q of 50. Q of 220 with 108/40 Litz wire.

Mechanical set up for a 24 x 24 inch form. Pretty much self explanatory.

Final for the large form. Took about an hour for the first one. Should be able to do two at a time later.

First winding attempt. Inductance was about 3.7 mH and a Q of about 200 at 100kHz.

Spiral Inductor Adventure


The following is a pictorial summary of my spiral inductor development.

A simple Spiral Inductor Calculator can be found at: .

The spiral inductor mathematics can be found at:   (50MB!!)

Scroll down to page 204.

If this link goes 404, I have it on my machines here and will put it up on my server.


Some of these files are large, give them time to load by clicking on the photo.

Link to Litz wire information: . See navigation column on the left.

137.5 KHz Elliptical (Cauer) Band Pass Filter, made of four 4x4 inch spiral inductors as shown above. I threw it together in a coupla hours and obviously needs some refinement. But it works.

Another view.

S21 of the filter.  I got lazy and two of the forms inner diameter is too large, thus lowering the Q and causing a few dB of extra loss. But 3.4 dB ain’t bad.

S11 of the filter. The two above mentioned two spirals need tweaked. But not bad with a return loss of 22 dB.

Next time I’ll try for greater attenuation in the stop band. Really need to figure out how to package it.

2200 Meter Band Pass Filter Schematic.

To get a copy of schematic:

Click on thumbnail to the left.

File/Save As/ give it a name.

Then open in Paint.

Print from paint.