Bandspreading Techniques
(20 June 99)
    Bandspread in a crystal set?  You must be nuts!  Well, maybe so, but let's give it a shot anyway.  Bandspreading is nothing more than making it possible for you to tune your set more slowly.  Why bother?  Well, if you are in a crowded band, such as in a metropolitan area with lots of strong stations, you have probably already gone for more sophisticated crystal radio designs to allow you to separate the signals.  Even in my location, with two peanut whistles only 50 kHz apart, sometimes I still have to tune very carefully when my rig is set up to be selective so that I can get the station I want, and not hear the other.  In the HF (shortwave) region, with stations fading out and then roaring back with regularity, being able to tune slowly allows me to catch stations that are just out of the mud, but are more often buried by the stronger stations that usually dominate, and which catch my attention easily.  With HF ham rigs, the ability to tune slowly is an absolute necessity; otherwise you sweep right past stations that are only in your selectivity skirt for a few hundred Hz.
    Most crystal sets use a tuning capacitor that has only a180 degree rotation in going from around 10 to 360 pF or so - on the BC band, this gets you through the 1.2 MHz-wide band rather quickly - on HF, you cover several MHz, and a  quick sweep through the frequencies can cause you to miss stations if you are only expecting what you normally hear.
 

    There are two basic types of bandspreading:  Mechanical and Electrical

    Mechanical:  The simplest way to slow your rate of tuning down, or at least let you tune more sharply, is to use a larger tuning knob; your ability to control your tuning rate varies directly with the diameter of the knob, assuming you grab it on the outside edge.  Another method that used to be popular was to have a small diameter shaft turning a larger diameter one connected to it by a dial cord; the cord was usually kept tight by a small spring.  A third way, still much in use, is to have a planetary dial drive which makes several rotations to drive the capacitor through one half a rotation.  Dial cord drives are tricky, and planetary drives can cost more than the rest of the set, so I usually go for the largest knob I can find.
Another thing to watch out for with mechanical bandspread drives, should you try one, is backlash or slop - you reverse your direction, and lose a couple of degrees of turn before the capacitor starts to move.  For inductive tuning, usually consisting of moving a ferrite core in  and out of the coil, the once readily available AM tuning coils used to come with a metal cap through which a screw protruded that allowed you to vary inductance very slowly using a small knob to twist the core in an out.  Compression capacitors also use a screw to let you make several turns from stop to stop, but aren't ordinarily used for main tuning, and are usually for screwdriver use only.  If you do stick a knob on a compression cap with some epoxy, a soda bottle cap should work, and will allow you about 3 or more revolutions of rotation stop to stop.

    Electrical:  Nothing magic about this either.  Here you use two or more capacitors to change the amount of capacitance a turn of the knob gives you.  Shown below are some of the more common ways to electrically bandspread your tank circuit:

 
 
     Figure a. is one of the favorites.  If one capacitor is significantly larger than the other, it is called the band set capacitor, and the smaller one is the bandspread capacitor.  Capacitor size ratios of from 2:1 up to 10:1 should be used.  Keep in mind with all of these  that minimum circuit capacitance is the sum of the minimum of both capacitors.  Since the polyfilm caps I usually use, because they are cheap, only run to about 200 pF, I sometimes just put 2 in parallel to get full AM band coverage (and 360 degrees of rotation at the same time).  Incidentally, since these capacitors are really two sections, a 150 pF and 50 pF, you can get full band coverage with a 200 pF capacitor, with both sections tied together, and only the 150 pF section on the other; use the 150 pF section for the top part of the band and it will slow your tuning rate down a bit.

    Figure b. is sort of a poor man's bandspread, switching in one or more fixed capacitors to increase capacitance in increments, then adjusting the variable as needed.  I used this with my peanut special to get to the bottom of band.  The fixed capacitor should be no larger than the range  of the variable to prevent skipping over frequencies.  

    Figure c. has a fixed capacitor in series with one of the variables.  This reduces the tuning range of the series variable from about the minimum of the variable to the value of the larger capacitor in the series branch.  Good to use if you don't have any small value variables.

    Figure d. taps one variable down on the coil, reducing its effect on tuning range.  In this case, assuming it is the same value as the band set capacitor for example, its effect capacitance-wise over its full range is approximately equal to the percent of the coil it parallels.  If you use a switch to select different coil taps, you can thus vary your bandspread.

    A note on using fixed capacitors:  I have found that some are better than others, and a lossy one can reduce set sensitivity when the fixed cap is switched in.  I have seen this with the readily available ceramic disc capacitors.  This may not happen to you, but just be aware of the possibility.  You might want to try different types here.  Try to use good quality switches as well, and make good solder connections....

    Finally, to use the K.I.S.S. principle, use a capacitor with a range no bigger than you need.  That is, use only enough range to cover that part of the spectrum in which you are interested.  Using a single 365 pF capacitor on the HF bands, one turn of the dial will sweep through several MHz, and no matter how selective your crystal set, you will be sweeping through a lot of dead frequencies and then sweat tuning in and then separating the stations you want to hear.

    You can pretty much do with inductors (coils), what you can do with capacitors, and some of the old designs used a lot of variable inductances.  Putting inductors in parallel has the same effect as putting capacitors in series, and vice versa for series inductors.  My uncle Elmer (I must have had one sometime) told me, however, that circuit Q was highest when you used as much inductance as your frequency range allowed, and to let the capacitor do all the "heavy lifting".  Still, some of those old variable inductors look pretty elegant, and I am sure they work pretty well, even if they are a bit more complex to build.

 

HOME    EXPERIMENTING    DX TIPS    ANTENNAS    TESTING    WAVE TRAPS