Crystal Radio

Basic Crystal Radio

Basic Crystal Radio Circuit

         \|/       \  /       
          |      +--[]---||--+
       _  |      |  S1   C2  |
       /  |      |           |
 +---||---+-->|--+-------VV--+
 |  /C1   |      |       R1  |
 |        |      |           |
 +--UUUU--+      +-------||--+
 |  L1                   C3  |
 |                           |
 +---------------------------+
 |                            
---                           
 -                            
  • D1 1N34
  • L1 230 uH
  • C1 15-365 pF
  • C2 0.01-0.22 uF
  • C3 500 pF not needed?
  • R1 47 kOhm
  • S1 Piezo Speaker

Radio frequency energy propagates through each section of a crystal radio, provided each section is properly coupled to the next.

  1. Antenna
  2. Antenna Tuner (optional)
  3. Tank Circuit
  4. Detector
  5. Transducer
  6. Ground

Ideally, all of the energy available at your location that was used to encode the audio frequency content of the station of interest is dissipated in the transducer, while the rest is either not intercepted at all or passes to ground. In reality, you will only get to hear some of this energy, and an additional DC current is needed for the detector to function.

                  \|/                             
                   |                              
                   |                              
        .......... |                        \  /  
       /        /  |                         ||   
 +---||---+---||---+--||--+      +--||----+--[]--+
 |  /C1a  |  /C1b     C3  |      |  C2    |  SP  |
 |        |               |      |        +--VV--+
 +--UUUU--+               |      |           R2  |
 |  ::::                  |  ::  |  ==         / |
 +---UU--------------->|--+--UU--+--UU--(A)--VV--+
 |   T1               D1     L2     L3      /R1  |
 |                                             / |
 +-------------------------------------------VV--+
 |                                          /R3   
---                                               
 -                                                

The circuit I like right now is a bit more involved.

  1. Antenna
    • Tuggle Coupling
  2. Tank:
    • T1: Ferrite Toroid (2:1)
    • C1: Dual Air Variable Capacitor
    • Transformer (Mystery) Coupling
  3. Detector
    • D1: Modern 1N60
    • Series Bandpass LCR Coupling
      • L2: 57 mH Series Inductor
      • C2: 20 µF Series Capacitor Bank
      • C3: 2.2-33 pF RF Bias/Charge Pump Capacitor
  4. Transducer
    • SP: Piezoelectric Element (Horn Loaded)
    • R2: 2 MΩ Piezo DC Protection
    • DC and Low-frequency Bypass
      • R1: 100 kΩ Log Potentiometer (should be 1MΩ)
      • L3: 1.5-24 H Inductor (Bogey)
  5. Resistive Coupling (Selectivity Enhancement)
    • R3: 1 MΩ Parallel Log Potentiometer

Circuits to Try

                   \|/              
                    |               
        ..........  |               
       /        /   |               
 +---||---+---||----+               
 |  /C1a  |  /C1b             \  /  
 |        |                    ||   
 +---UU---+--->|---UU--+-------[]--+
 |   L1       D1   L2  |       S1  |
 |                     |    /      |
 |                     +--VV-------+
 |                       /R2       |
 |                          /      |
 +------------------------VV-------+
                         /R1       |
                                   |
 +------------------------||-------+
 |                        C2
---
 -

This works, too!

Also, make the high-Q side be switchable between parallel and serial (bandpass and band-stop) to allow envelope shaping. This configuration may compensate for high diode capacitance in addition to antenna capacitance.

                       \|/
                        |
                        |
        ................|............
       /        /       |          /.         
 +---||---+---||---+----+    +---||-.--+------+
 |  /     |  /     |         |  /   .  |      |   
 |        |     /  |         |     /   |     ---
 |        +---||---+    +----+---||----+      -
 |        |  /     |    |    |  /      |       
 |        | Q~100  |    | /  | Q~1000  |       
 |        +--UUUU--+---|||+  +--UUUU---+       
 |                    /   |        
---                     +-+           \  /
 -                      |      +--||---[]--+
                        |      | 0u15      |
                        |      |           |
              1M   15m  |      |      1M   |
              VVV--UUU--+-->|--+------VVV  |
               |   ===  |      |       |   |
           +---+        |      |       +---+
           |   |        |      |           |
           |  VVV---||--+     VVV------||--+
           |  1M    10p       1M       220p|
           +-------------------------------+

Can the whole thing be Tuggle tuned?

Separate sideband and carrier tuners are variably coupled using a differential capacitor. FIXME Need to figure out how best to couple the tuner to the detector.

             \|/
              |
        ..................
       /      |         / 
 +---||---+   |   +---||---+--+   
 |  /     |   |   |  /     |  |   
 |     /  |   |   |        | ---   
 +---||---+   |   |        |  -    
 |  /     |   |   |        |              \  /
 | Q~100  |   | / | Q~1000 |            +--[]---||--+
 +--UUUU--+--|||--+--UUUU--+            |       0u1 |
 |          / |                         |           |
 |            |            1M    15m    |      1M   |
 |            |            VVV---UUUUU--+------VVV  |
 |            |             |    =====  |       |   |
 |            +----->|------+           |       +---+
 |            |             |           |           |
 |            |            VVV--||--+  VVV------||--+
 |            |            1M   10p |  1M       220p|
 |            +---------------------+---------------+
 |
---
 -

The positive side of the diode may need to be capacitively coupled to the antenna or to the ground.

This one has it all.

                                    \|/
   Tuggle Antenna Tuner              |
             ..........              |
            /        /               |
 +--------||---+---||----------U-----+-(::)-+
 |       /     |  /           :::    |      |
 |             |            +-UUU-+  +-VVVV-+ 
 +--UUUUU---UU-+            |     |         |
 |  = = =   ::              |   / |        ---
 |        +-UU--UUUUU--+    +-||--+         -
 |        |     = = =  |     /  Wave Trap
 |        |         /  |            
 |        +-------||---+           
 |        |      /     |                      
 |        | /          |    /    33n              
 |     +-|||-----------+--||--+--||--+--UUUU+UU--+--+
 |     | /20p Hobbydyne  /5p  | Benny|  ====|==  |  |
 |     |                      |      |      |    |  o
 |     +--------->|-----------+-----VVV     +-||-+   )
 |     |                            10k     |       o
 |     +------------------------------------+-------+
 |             
---                                   40000 : 16 Ohm
 -                                       50 :  1 turn
                                       2500 :  1 mH

Suppose

  • Ltank is 250 μH.
  • Tuning f=1000 kHz ≡ ω=6283185 rad/sec.
  • The total tank capacitance needed is about 100 nF
  • So Ctank will be set around 75-80 pF, less strays.

ω= 1/{sqrt{LC}} right C={1/{omega^2 L}}= {1/{6283185^2*250x10^{-6}}}=

FIXME See Crystal Set Analysis, Gollums Crystal Reciever World, by Berthold Bosch, DK6YY.

        Antenna Tuner
        ~~~~~
 +---+--UUUUU--+
 |   |     L1  |
---  |       / |      +---------->|--+----+
 -   +-----||--+      |              |    |
     |    /C1  |      +-----VVV      |    8)
     | /     / |      |    Rb|       |    |
 +--|||----||--+--||--+--||--+--UUUUU+UU--+
 | /Cd    /Ct    Crfb    Cb     =====|==
 |                                   |
 +-----------------------------------+

Standard Circuits

Torroid Set

\|/
 |
 | Wave Trap
 |      /      Tuggle Antenna Tuner
 | +--||-+               ........
 | | /   |              /      /
 | |     |  +-UUUUU---||--+--||------+
 | +-UUU-+  | = = =  /    | /        |
 |   = =    |             |         ---
 +----U-----+--UU---------+          -
               ::       /
            +--UU-----||--+-------------------+------o
            |        /    |                   |       )
            +-UUUUU-------+--||--+--||--+-UUUU+UU----o
            | = = =       |  10p |  33n | ======= 
            |       +-UUU-+      |      |
 Simplified |       | ===        |      |
 Hobbydyne  |    /  | 7m5        |      v  Benny  
            +--||---+----->|-----+-----VVV
              /                        250k

Note how the Hobbydyne and the Benny work together on either side of the detector to adjust DC, audio, and RF current flows. FIXME But this is incorrect Benny placement according to the sources I've read.

                                +-----VVV
               ::       /       |Benny |
            +--UU-----||--+-----+--||--+------o
            |        /    |            |       )
            +-UUUUU-------+--||--+-UUUU+UU----o
            | = = =       |  10p | ======= 
            |       +-UUU-+      |
 Simplified |       | ===        |
 Hobbydyne  |    /  | 7m5        |
            +--||---+----->|-----+
              /

Wavetraps

The Tuggle Antenna Tuner

Without a coil, the Tuggle configuration is more of a variable coupling between the antenna and the tank than it is a tuner.

 \|/
  |     ........
  |    /      /
  +--||--+--||--+------+
    /    | /    |      |
         |      |     ---
         +-UUUU-+      -
           ~~~~  
         +-UUUU----o
         |      
         +---------o

The Tuggle is more commonly used with the serial capacitor on the ground.

 \|/
  |     ........
  |    /      /
  +--||--+--||--+
  | /    | /    |
  |      |      |
  +-UUUU-+     ---
    ~~~~        -
  +-UUUU----o
  |      
  +---------o

By using one side of a double variable capacitor in series with the antenna to decrease its capacitance and the other side in a tank circuit, the Tuggle circuit is an effective tuner. With good construction, the Q of this tuner should be around 100 and a bandwidth of 5-25 kHz from the low to the high end of the broadcast band. 1).

Ca represents antenna capacitance and C is the value to which both sides of the capacitor are adjusted.

The equivalent capacitance resonating with L is:

C_{eq}=C({2 C_a + C}/{C_a +C})[=C + {1/{{1/C_a}+{1/C}}}—dw]

Clearly, Ceq>C.

Following is a numerical example illustrating the use of the above results. Let C be a variable capacitance with CMIN = 20 pF and CMAX = 475 pF. Let also Ca be 200 pF. Then, Ceq varies between CeqMIN = 38.18 pF and CeqMAX = 615.74 pF.

If we wish to tune the MW broadcast band starting at 530 kHz, then the required inductance L will be:

L=1/{{omega_{rMIN}}^2 C_{eqMAX}}=146.45 µH

The circuit will tune up to:

f_{MAX}=f_{MIN}sqrt{C_{eqMAX}/C_{eqMIN}}=2.128 MHz

If we use for C a variable capacitance with CMIN = 20 pF and CMAX = 365 pF, then

CeqMIN = 38.18 pF and CeqMAX = 494.20 pF, giving for the required inductance L a value of 182.46 uH. The circuit will tune up to fMAX = 1.906 MHz.

Analysis of the Tuggle Front End, Ramon Vargas Patron, 2004.

The Hobbydyne

       Tank
   o           o
   |           |
   |      /    | /
   +----||----|||--+
   |   /Ct   /Cd   |
   |               |
   +-----UUUUU-----+---> Detector
   |     =====    
   |      Lch   
   +-------------------o Common

The Hobbydyne is a capacitive transformer.

“For maximum performance at the low end of the band the RF choke should be at least 5 mH, and preferably 7.5 or 10 mH.”2)

  1. Cd Differential Capacitor: 10-20 pF
  2. Ct Piston Trimmer Capacitor: 10-15 pF
  3. Lch RF Choke: 1.5-27 mH (Bigger is better.)

I think I have a tenuous grasp of how this darn thing works. This circuit seems to have two important resonant modes. The first is the series resonance of the RF choke and only the bottom leg of the differential capacitor. This resonant frequency provides the most efficient power transfer to a high-impedance load from a low-impedance source.

The second mode is higher in frequency and is created by the choke and all three capacitor sections acting in series. This mode transfers power more efficiently to a high-impedance load from a high-impedance source.

The mode created by the choke and the series combination of the bottom of the differential capacitor with the other leg and the trimmer appears to be less important. (However, this mode seems to becomes important when the audio transformer primary is used as the RF choke.)

From this analysis, a 7.5 mH choke seems to be the best choice as it provides a mode 1 resonant frequency range of about 400 to 1800 kHz when used with a 1-20 pF differential capacitor. (The preferred value of 6.8 mH will also cover the band well.) Also, 10 mH may be about the largest practical choke to use as greater inductance should cause loss of the high end of the AM band unless a smaller differential capacitor is used.

For 20pF/10pF/20mH, the mode 1 range is about 250-1125 kHz, assuming the minimum capacitances are both about 1 pF. Mode 2 ranges around 600 to 1600 kHz, so using this larger inductor may be helpful if the antenna has very high impedance at higher frequencies. But I doubt it since this may mean having to constantly adjust the trimmer.

Tank Coil

Tank Capacitor

Detector

FIXME Diodes to test

The FO-215 is supposed to be the best germanium detector, available here.

RF Choke

The Benny

                  Rb
Detector >--+-----VVV     Low End of
            |      |      Output
            +--||--+----> Transformer
               Cb         Primary

The Benny out of context looks like a standard RC low-pass filter.


                    Rb
Detector >----+-----VVV
              |      |
     +---||---+--||--+--o  Output
     |  Crfb     Cb        Transformer
     |                     Primary
 o---+------------------o  
Common

The Benny also forms a capacitive transformer.

“When a transformer is used; the parallel RC* (See [Rb] and [Cb] on the schematic above.) should be connected in series with the low end of the high impedance transformer primary winding. In this case the resistor should equal the transformed effective headphone impedance (PHI).”3) [However, the referenced schematic shows the Benny connected as shown here. I think the 'low end' refers to connecting to the high-impedance primary winding of an autotransformer on the far side of the low impedance winding. Another source written by its creator specifies the 'cold' end of the transformer.–ed.]

  1. Load Resistor Rb: 50-500 kΩ (adjust to equal load impedance) (250k audio-log-taper preferred)4)
  2. RF Bypass Capacitor Crfb: RF Bypass 5) 150 pF - 0.001 μF
  3. Audio High-pass Capacitor Cb: 0.1 μF Audio Bypass “large enough to bypass the lowest audio frequency of interest”6)

When using a transformer, the load impedance is equal to the headphone impedance times the square of the audio transformer turns ratio.

R_b = Z_{out} = Z_{ph} ({t_p}/{t_s})^2 = Z_{ph} {{L_{p}}/{L_s}}

Take the high-pass cutoff for Cb as 100 HZ, and for Crfb as 500 kHz. Consider using two 8Ω headphone drivers in series with a 50::1 transformer and calculate the 3db cutoffs.

R_b = Z_{out} = Z_{ph} ({t_p}/{t_s})^2 = 16*{50^2} =40000=40 kΩ

f = 1/{2 pi RC} right C_{b}=1/{100*2*pi*40000} approx 0.0397 mu F approx0.047 μF

f = 1/{2 pi RC} = {10^6}/{2*pi*40000*0.047} approx 84.7 approx85 Hz

This may be too low to eliminate the 50-60 Hz hum.

  • f = 1/{2 pi RC} = {10^6}/{2*pi*40000*0.033} approx 120.6 approx120 Hz
  • f = 1/{2 pi RC} = {10^6}/{2*pi*40000*0.022} approx 180.9 approx180 Hz

The series resistance for the RF bypass is the sum of the load impedance and the resistor.

C_{rfb}=1/{500000*2*pi*2*40000} approx 3.97 pF approx5 pF

f = 1/{2 pi RC} = {10^12}/{2*pi*2*40000*5} approx 397887 approx400 kHz

Note a ten pF capacitor would still have a 3 dB cutoff of 200 kHz.

ZphonesTransformer TurnsZout=RbCbCrfb
16 50::1 40 kΩ 0.033 μF 10 pF

Low-loss Impedance-matching Output Transformer

 +---UUUUU---+
 |    L1     |  Tank    
 |      /    |
 +----||-----+
 |   /C1     |
 |           |
 |      /    | /
 +----||----|||--+  Hobbydyne   
 |   /Ct   /Cd   |         
 |               |             Rb
 +-----UUUUU-----+-->|--+-----VVV  
 |     =====            | Benny|
 |      Lch             |      |  Audio Xformer
 +------------------||--+--||--+--UUUUU+UU--o
 |                 Crfb    Cb     =====|==   )
 +-------------------------------------+----o

 +--UUUUU--+----+
 |   L1    |    |
 |     /   |   ---
 +---||----+    -
 |  /C1    |
 |         |                  +----o
 |    /    | /    Cb          |     )
 +--||----|||--+--||--+--UUUUU+UU--o
 | /Ct   /Cd   |      |  =====|==
 |             |    Rb|       |
 +--||--+-->|--+-----VVV      |
   Crfb |                     |
        +---------------------+

There is a whole lotta impedance-matching happening.

FIXME Can simplify winding by using one continuous coil (an autotransformer like this) with a zillion taps.

If the crystal set has an output impedance of 20,000Ω, the turns ratio necessary to drive an 8Ω speaker is sqrt{20000/8}=sqrt{2500}= 50::1.7)

You want the largest inductance you can wind,8) so use a big honking high-permeability core.

To wind it, twist a bundle of wires together and wind them around the core. How many wires are in the bundle depends on whether you prefer more taps and more soldering (more wires and fewer turns) or more winding and fewer taps (fewer wires and more turns). Note wire about 30-gauge (28-33) is best for the audio range9). You may want to use one diameter for the primary, and a different gauge or insulation color for the secondary, just to keep them straight.

For example, try twisting together 16 fine wires all long enough to wrap a ferrite toroid with 3*16=48 turns total. Wind sixteen turns all the way around, then add a length of secondary to the bundle and wind all seventeen wires together between the existing winding sixteen times more. Now stop winding the secondary and wind the primary bundle another sixteen times. Secure the coil to the form, then solder one wire from each end together being careful not to make a short circuit.

To solder together the secondary windings, first strip the insulation from all the ends. Then, select a wire from one side and use a continuity tester to find the other end of this wire on the other side. Then solder one end of this wire to any other wire on the other side of the loop. Do not solder together the two continuous ends in a short circuit, and do not solder together two wires from the same bundle. Place consecutively numbered labels (starting with zero) on each solder connection as you make them. Repeat this procedure until you have only two unsoldered secondary ends remaining, one on each side of the coil.

The secondary has sixteen turns, and the primary has 3*16*16=768 turns, a 48::1 coil with sixteen primary taps. As an alternative you can make the secondary a bit longer than the primary and pull (and label) secondary taps after each revolution around the toroid's major radius. This will give ratios from 48::1 to 1::3, depending on the primary and secondary taps selected.

For another example, twist together ten fine wires and a different primary wire, all long enough to make 16×5=80 turns around a ferrite toroid. Wind as above a total of five times around the major radius. The primary will have a total of 5*10*16=800 turns with ten taps and the secondary will have 1*5*16=80 turns with five taps, for a turn ratio range of 50::1 to 1::5.

If you don't know ahead of time how much wire you need or how many turns will fit on your toroid you can still wind these coils with little wire loss. Start with at least half as many coils as you need wires in the bundle so you have as many loose ends as the wires in the bundle. (If your wire is on spools, you need as many spools as you have wires in the bundle.) Gather the ends together and, without cutting any wire from the coils, pull half again as much as you think you'll need for one full revolution through the center of the toroid and measure this length. Now twist the wire you pulled through the loop and wind the core, covering the fraction of the core you intend to cover. (In the first example you might want to cover 1/3 of the core with this first winding.) When you have wound the full revolution, measure how much wire you have left; now you know how much each revolution takes. Wind the remainder of the loose ends, then measure enough off the spools to complete the core. Cut the wires off now, then twist them and finish winding your coil, pulling and labeling secondary taps each time around. Note this method results in one fewer secondary tap, but you won't lose any turn ratio combinations.

FIXME Elaborate on the use of line matching transformers intended for microphones. The Shure Model A95UF looks like it has the following specs configured as high impedance to 19-75Ω.

  • 4300:37.5 Ω DC Resistance
  • 33:1 turns ratio
  • 17,500:16 Ω Impedance

Headphones

Very sensitive audio headphones are difficult to come by these days, even if you are willing to match the low impedance of modern drivers to the detector with some kind of transformer.

I'll be looking at building headphones using piezoelectric (piezo) driver elements.

Make a Crystal Radio from Old Toys

You can make a crystal radio entirely from toys or other household items. You can even build one without using any electronic components at all.

Crystal

  • Rock and pen spring
  • Razor blade, pencil graphite, and pen spring

Coil

  • Spiral-cut aluminum soda or beer can
  • Metal Slinky (may have built-in iron oxide rectifier)

Capacitor

  • Aluminum Foil and a large children's book

Earphone

  • Piezoelectric element from a lighter
  • Piezo speaker from a toy, greeting card, or other annoying item

References

Alternative Schematic

 ^         
 |         
 |     _   
 |     /   
 +---||---+
 |  /C1   |
 |        |
 +---UU---+
 |   L1   |
 |        |
 +--->|---+
     D1   |
          |
 +---||---+
 |   C2   |
 |        |
 +---WW---+
 |   R1   |
 |        |
 +---[]---+
 |  /S1\   
 |         
---        
 -         

Mystery Crystal Radio Circuit

 \|/                 
  |                  
  |                  
 /                   
o   o----+           
|     _  |           
|     /  |    \  /   
+---||---+  +--[]--+ 
|  /C1   |  |  S1  | 
|        |  |      | 
+--UUUU--+  +--VV--+ 
  +-UU-+    |  R1  | 
  |T1  |    |      | 
  |    +->|-+--||--+ 
  |      D1    C2  | 
  +----------------+ 
                   | 
                  ---
                   - 
  • D1 Crystal Detector
  • T1 50::25, 3D3”
  • C1 15-365 pF
  • C2 0.0005-0.001 uF not needed?
  • R1 47 kOhm
  • S1 Piezo Element Speaker

Personal Tools