Sunday, February 28, 2010
Saturday, February 27, 2010
Casting a Plastic Radio Case
Having been attracted for years to colorful plastic radios, particularly Catalin, I decided to see if I could recreate one. The word "Catalin," as with Coke or Xerox, is a brand name that in the 1940s became a more universal name for colorful cast phenolic plastics used in radios, jewelry, and cookware. Catalin plastic can be opaque, translucent, solid colored or swirled. You can see examples of Catalin radios in the eBay auction images.
After I drew up the design I wanted, I made a plexiglass model. On it I painted 8-10 layers of latex to make a mold. Then I mixed two part epoxy with cadmium yellow acrylic paint and red marine epoxy coloring to cast the case.
The yellow louvers were cast in plastic tubes from the hobby shop; the recessed knobs were cast on the hemispherical end of a cigar tube surrounded by straws, arranged gatling gun style. That formed the fluted knobs.
The translucent tuning dial was cast out of the same plastic with a mold that used the top of a spice jar featuring a nice ring. There had to be a better way than waiting for eight layers of latex to dry, and sure enough, hot melt glue makes a dandy mold for small parts.
The translucence was achieved by vigorously mixing the resin & hardener so that tiny air bubbles formed, spreading themselves uniformly throughout the mixture.
I used Microsoft Publisher and my ink jet printer for the dial numbers. The ring of numbers is printed on overhead projector transparency plastic. I then cut out the circle and embedded it in the curing epoxy. The edge and surface imperfections mysteriously and beautifully disappear when it's embedded.
How does the radio perform? I was able to optimize the circuit so it plays AM at room filling volume, gets 20 locals up to seventy miles away, quite a feat with only two transistors and an internal antenna.
Here's what's inside. The two transistor reflex circuit is a supremely elegant and efficient circuit. The circuit I used was designed by G.W. Short and published in 1968 in Radio Constructor, a British periodical.
I actually built the circuit first and was looking for a worthy cabinet for it when I got the idea to cast plastic
Comparing this picture with the image above, you can see how the tuning and volume knob shafts mate with the knobs.
|This side view shows the bending of the chassis, which started out as a 4" x 10" piece of aluminum from the hardware store. The ferrite bar antenna (with the green end at top) is from a radio I found in a garbage can on the side of the road.|
I etched the circuit board and modified a few components to optimize the circuit. During final assembly I shortened the bottom shaft to make it match.
This back view is what shows when you take the back off the radio. The battery is situated for easy access.
I've had the gears for years in my junkbox; they are from a defunct copier, and I intended to use them in part a of a telescope drive. Here they are configured as a fine tuning control. Because tuning capacitors only rotate 180 degrees, the 1:2 ratio also permits the dial numbers to be spread around the full circumference of the dial.
The entire reflex transistor construction article can be found here.
Then scroll down to Silicon Reflex, by G.W. Short.
Adapted from an article by A.J. Crighton published in the August 1978 edition of Everyday Electronics Magazine
After the introduction of the Ferranti ZN414 Integrated Circuit in the 1970's there were many articles published that provide details of simple radio circuits that used the I.C. The ZN414 is no longer made, but a new and almost identical I.C., the MK484, can substituted in all the ZN414 circuits and has proved to be a least, if not more, effective.
In the 1970's the ZN414 could be quite expensive to buy compared to cheap transistors and so this article was published in Everyday Electronics magazine as an inexpensive alternative. Today some readers may find it difficult to obtain the MK484 so this circuit still holds great value.
This design uses three cheap transistors, the BC548 as can be seen from the diagram below.
The tuned circuit, L1/C2 selects the required station, transistor TR1 providing the r.f amplification. The signal is then passed to TR2, a small amount of the signal is fed back from the collector to the base of TR1 via the tuned circuit. A mixture of both radio frequency and audio frequency circulates in this feedback loop, the r.f. reinforcing that coming via the tuned circuit, while the a.f. is passed to the audio stage.
Transistor TR3 provides audio amplification that is enough to drive a crystal earphone which is plugged into the socket SK1
Most of the components are mounted on a small piece of plain matrix board having 13 by 7 holes as shown below:
Wiring details for the matrix board and other components
Modification required to the open frame jack socket
to make an 'on-off' switching arrangement, and
other layout details
Connections to the components are made using single cored wire on the underside of the board, this layout enables a smaller layout than using stripboard, however some experimenters may still wish to alter the layout to suit construction on 'strip board'.
The aerial coil L1 is 80 turns of 32 s.w.g. enamelled copper wire close wound onto a small ferrite rod. The jack socket serves as both a connection for the earphone and a power switch. One of the switch contacts must be bent slightly out so that they 'make' contact when the phone plug is inserted rather than the usual 'breaking' action when the plug is inserted. As can be seen from the circuit diagram the power switching is applied in the negative side of the circuit.
The tuning capacitor C2 is a typical miniature polyvaricon type that is widely available from electronic component suppliers and which can be seen in most pocket type transistor radios. If this variable capacitor has two gangs then usually only one gang will be required although some types may require that the two gangs are wired in parallel in order to be able to tune to the lower end of the medium wave band.
USING THE MEDIUM WAVE MINI RADIO
Before switching on for the first time, double check that all the components are wired up properly and that the battery is connected the the correct way round. When you are happy that there are no mistakes plug in the earphone, this will energise the circuit and some background noise should be heard in the earphone.
Adjusting the tuning control should produce a few local transmitters, the number of stations available will depend on the location and the strength of signals available. The tuning is quite sharp and therefore some care is needed when using the control. The ferrite rod is directional and so the radio should be rotated in both planes for maximum signal pick-up for each radio station.
If a whistle occurs then this can be reduced by increasing the value of the 2.7k resistor, R2, to 3.3k Ohms. It may be possible to replace this fixed value resistor R2 with a sub-miniature preset resistor to make fine adjustments.
When the plus input is positive , it pass through the antenna signal. When the plus input is negative , it does not pass through the antenna signal. Therefore antenna signal is
The RF signal is passed from the antenna through C1 to the tuned circuit made up of L1 and C2．One end of L2 feeds the RF signal to the base of Q1 for amplification and the other end ties to the junction ofR1and R2 to supply bias to the transistor．A 0.02-μF capacitor，C3，places the“D”end of L1 at RE ground.
The amplified RF signal is fed through C6 to a two-diode doubler/rectifier circuit and then on to the volume control，R6．The wtper of R6 feeds the detected audio signal through C9 to the junction of R1，R2，and the“D”end of u．The“D”end of L2 is at RF ground，but not AF ground，allowing the AF signal to be passed through L2 to the base of Q1 for amplification. The junction of the 2.5-mH choke and T1 is placed at RF ground through C5. The amplified audio is fed from this junction to the input of the 386 audio amplifier，U1，to drive the 4"8-Ω speaker．The single transistor has performed a dual duty by amplifying the RF and AM signals at the same time．
Radio circuits built by Don Cross
Reflex AM receiver – August 2009
I am very pleased to report that I have finally built a radio receiver that is good enough that I can enjoyably use it!
This receiver is a modification of Charles Wenzel's Two Transistor Reflex Radio. Instead of a ferrite AM loopstick antenna, I use a magnetic loop antenna, and I added an LM386 amplifier stage to drive an 8-ohm speaker. There is a switch to select whether this amplifier should be used to power the speaker or whether crystal earphone listening is desired. I use six AA rechargeable NiMH batteries to provide the 8V power supply, but a 9V alkaline battery or a 6V lantern battery work just fine. I think even a small solar cell array would suffice to power this thing if you were camping on a sunny enough day!
|Prototype receiver on breadboard. I will soon move all parts currently mounted on the breadboard to a soldered circuit board. I will also add interface connections with screws on the wood base, so that I can swap out other receiver modules to the same wooden box.|
|Update – 25 August 2009. The finished AM receiver. The detector and audio amplifier circuitry is now soldered to a permanent circuit board. I added a handle to make it easy to carry the radio around.|
|Close up view of the inside of the finished receiver. I have the schematics folded up inside a small zipper-style plastic bag, attached to the inside of the wood frame using Velcro. The screw connections facilitate removing the circuit board and replacing it with other experimental designs.|
Notes about this circuit
- The magnetic loop antenna is depicted as a transformer. It is actually a pair of hand-wound rectangular loops of enameled magnet wire, about 5.5 inches by 4.5 inches. If I build another one of these receivers, I will wrap maybe 16 or 18 turns instead of 20 on the tuner side to bring down the inductance so that I can tune higher on the AM band.
- The antenna is highly directional. You have to rotate the radio to aim it at the desired radio station. More specifically, for strongest reception of a nearby station (i.e. not receiving by skywave propagation) the radio station must lie in the plane of the loop, and that plane must be at right angles to the ground. (Apparently the magnetic component is polarized horizontally.) If you rotate the radio 180° around a vertical axis, its reception of a given station will be the same. This directionality can be used to advantage. Where I live there are two radio stations close together in frequency: 1030kHz and 1060kHz. However, it is fortunate for me that they are about 60 degrees apart in terms of compass direction from my house. Selectivity of this receiver is pretty good, and with proper aiming I can almost completely block out one of the two stations and focus on the other. I even put felt feet on the bottom of the chassis (the kind used on chair legs to prevent scuffing the floor), so that I can easily spin the radio around without scratching the table it is on!
UPDATE (27 August 2009): Try out my online Azimuth/Distance calculator page to determine the exact compass direction and distance of a radio station! Use Google Maps or Google Earth to find your exact latitude and longitude. Use this as "Point A" in the calculator. Use the FCC online AM radio station database to find the coordinates of the transmitter. Use this as "Point B". The calculated azimuth tells you how many degrees clockwise from North the station is from your location.
- The Band switch selects whether you are tuning the lower or higher end of the AM broadcast band. When the 220pF capacitor C1 is connected in parallel with the tuning capacitor, the tuner can reach down to well below 540kHz. With C1 disconnected, the tuner can reach as high as 1100 kHz. Where I live there aren't any stations I care about above this frequency, but in the future the aforementioned decrease in the number of turns in the magnetic loop antenna should better fit the AM band.
- The Output Select switch (see lower left in schematic) toggles between driving the high-impedance crystal earphone or the 8-ohm speaker via the LM386 amplifier. In earphone mode, the receiver draws a miserly 1.5mA of battery current; that could make your battery last for hundreds of hours! (At least, I think so. I am still in the process of testing practical battery life with my NiMH rechargeable batteries.) In speaker mode, the current consumption is more like 8mA to 15mA, depending on the volume setting.
- When the Output Select is set for earphone mode, the Volume knob has no effect. Use the Regen to get the best volume you can without distortion. You won't get head-banging rock concert volume, but you will be able to hear talk radio quite clearly.
- In speaker mode, you need to adjust both the Regen and Volume knobs to get proper balance between volume and distortion.
- If the 470K resistor R7 is replaced with a wire, you get problems with oscillations (so-called "motor-boating"). Looking at the LM386 data sheet, I see that there is a 50K resistance between the amplifier's positive and negative inputs. Apparently R7 limits the audio-frequency current flowing through C7 enough to stabilize the circuit. When I started this project, I had a wire instead of R7 and I was trying to figure out how to fix the oscillation problem. I disconnected the wire and noticed that if I touched one of my fingers to the wiper pin of the Volume potentiometer, and another finger to pin 3 of the LM386, the radio suddenly started working! I played around with various resistors in the range 100K to 1M and settled on 470K as a good value. If you build this receiver, try various values; another might work better for you.
- The potentiometer I use for the Volume knob has a built-in SPDT switch. I use this as my power switch: when Volume is turned all the way down, it clicks off.
- My design is a lot bulkier than is really needed, but I did this on purpose to make future experimentation easier. I want a lot of room for my large hands to move around! I want to be able to re-use the chassis, antenna, and controls for other experimental circuits, so I'm designing it to allow the receiver circuit board to be swapped out with other designs.
- The LM386 data sheet gives an example value of 10µF for the capacitor I call C10 in the schematic. I started out with this value, but I found the audio to sound "muddy" when the volume knob was turned up too high. In fact, there was a tendency to oscillate past a certain point. After experimenting, I settled on 2.2µF as a better fit. It reduces bass while leaving midrange and treble strong, and allows better fidelity sound at higher volume settings. I tried as low as 0.2µF, but at that point the sound was too tinny and wasn't capable of the volume level I wanted.
Experimental determination of inductance L and parasitic capacitance Cx
Tonight (28 August 2009) I had fun with an experiment to figure out the inductance of the tuning coil (the 20 turns of magnet wire), along with its so-called "parasitic" capacitance. I tuned the receiver to 6 different radio stations. For each radio station, I turned the receiver off, disconnected one of the tuning capacitor wires, and measured the tuning capacitor with my capacitance meter.
Now in each case, I know both the frequency of the radio station f and the tuning capacitance Ct. If you look at the schematic above, you will note that the 220pF capacitor labeled C1 (which I will call Cf here) can be included or excluded from the tuning circuit, using the "band" switch. I was able to tune two of the radio stations (WDBO and WORL) with either band switch setting, because they broadcast on frequencies included in the overlap of both bands. I recorded the band switch setting in all measurements. Overall, I took a total of 8 measurements of the 6 stations, with WDBO and WORL having 2 measurements each (one for each band switch position).
My goal was to solve for two unknowns: inductance L and parasitic capacitance Cx. The total capacitance of the tank circuit is
C = Ct + BCf + Cxwhere B indicates the state of the band switch: when that switch is closed, the capacitor Cf is included and B=1. When the switch is open, Cf is excluded andB=0.
The relationship between frequency, total capacitance, and inductance is
LC = 1012 / (2πf)2The factor of 1012 is needed to correct for the fact that f is expressed in kHz and LC is expressed in pF*μH. If "pure" units of Hz, F, and H were used, this numerator would simply be 1. The squared reciprocal of kHz makes the answer 106 times bigger than it would have been with Hz, because now we are dividing by a number that is a million times smaller (a thousand squared). By definition, there are 1012 pF in one farad, and 106 μH in one henry. Putting all of this together, we get a correction factor of 10(12 + 6 − 6) = 1012.
I made a spreadsheet that used this formula to calculate the value of LC for each station frequency f, recorded the tuning capacitor Ct measurements, and had a box where I could modify my guess for the value of Cx. When I divide the calculated value of the product LC by the total capacitance C, I get a theoretical value for the inductance L. By trial and error, I settled on an optimal value for Cx = 38.5pF, and modified the nominal value of Cf to be 223pF to make all the Lvalues as close together as possible. Here is table that shows what my spreadsheet looks like:
station f [kHz] LC [pF*μH] Ct [pF] B C [pF] L [μH] WFLF 540 86867 218 1 479.5 181.16 WDBO 580 75298 155 1 416.5 180.79 WDBO 580 75298 378 0 416.5 180.79 WORL 660 58150 65 1 326.5 178.10 WORL 660 58150 288 0 326.5 178.10 WYGM 740 46257 221 0 259.5 178.25 WONQ 1030 23876 91 0 129.5 184.37 WSDO 1400 12924 33 0 71.5 180.75
Based on these data, I have the following estimates:
I used two standard deviations around the mean value for the uncertainty of L.
L = (180.2 ± 4.6) μH
Cx ≈ 38.5 pF
(Update: 9 September 2009) I realized I should go back and measure the resistance of the magnetic loop antenna. It turns out to be 3.0 Ω.
This kind of analysis should help me design better hand-wound magnetic loop antennas in the future. Ideally, I may be able to design a tuner that doesn't need a band switch to cover the entire AM band. Another possibility is that I may be able to design a tuner that uses a variable inductor instead of a variable capacitor to tune to various stations. In general, it will be interesting to compare experimentally determined coil behavior with the values I get from the various formulas for estimating inductance based on coil geometry.
A Reflex radio is similar to the Regenerative receiver design, in that both use a controlled amount of positive feedback of the amplified signal, to reinforce it and obtain extreme sensitivity. Just what is fed back is what sets them apart.
The Regen takes a fraction of the amplified RF signal and feeds it back in phase with the signal coming in from the antenna. As the amount of regeneration is increased, a point is reached where the detector breaks into self-oscillation at the signal frequency (actually, just slightly off from it, which allows a beat note to be formed for CW reception). For the AM broadcast radio listener, this means a loud squealing. Wouldn't it be nice if we could get tremendous gain without that annoying howl? The Reflex design essentially feeds back just the demodulated audio to the RF front end;
the input transistor is pressed into double-duty: it amplifies both the RF and the AF (audio frequencies) at the same time. This gives a large gain without a heterodyne squeal so common to regenerative receivers.
In the circuit above, an incoming station is tuned by the front-end LC circuit. A tap in the lower 3rd or 4th of the inductor feeds the base of Q1; this low-tapping is done for impedance matching; i.e., the L-C tuned circuit would be detuned and its Q lowered from the loading effect of having a 600 ohm transistor base hanging off it, if the transistor were not tapped down on the inductor.
The RF gets amplified by Q1, then fed into a diode-cap pair which basically acts like the detector in a crystal radio, but with RF preamplification. Since the capacitor filters the RF out by shorting it to ground, we find the bottom of the coil at ground for RF while still "hot" for audio frequencies. Now the demodulated (by the diode) and filtered (by the cap) audio is put back into Q1's base. The emitter bypass cap is of a value that gives voltage gain at audio frequencies. The pot controls how much emitter bypassing there is and therefore how much audio feedback occurs. So the transistor is amplifying both RF and audio at the same time! And now, when the "volume" (emitter bypassing) is turned up, we find that there is some audio distortion, but no heterodyne squeal.
Notice also that the diode doubles as a path for DC bias for Q1's base. The 33 mH inductor acts as an RF choke / lowpass filter, along with the caps on either side of it. The 3.3k is Q1's collector load resistor, and you could put a crystal earphone from the bottom of the 3.3k resistor to ground, for a 1-transistor reflex with surprising audio volume (with a decent antenna).
By adding a second audio stage whose collector load is the primary of a small 1k :8 ohm audio transformer, we find that we can run Q2 as a Class-A amp, and that this stage has enough power to drive a decent 4", 8 ohm speaker to a comfortable volume if nearby stations are strong. I found that the cheaper the speaker, the less adequately
the receiver performed.
Considering all the trouble radio designers have gone through in years past to come up with the slickest Class-AB power amp design, it's amazing that we can get good volume with one transistor that isn't even a power amp! (The 2N3904 is a small-signal NPN; it is not a power amp) I've also noticed that the audio is very clean-sounding-- no crossover distortion in this little amp operating Class-A.
Like any simple receiver, this one has its "-isms" and annoyances. One is the lack of AGC (automatic gain control, or automatic volume control). It seems more pronounced in a Reflex than, say, in a Regen radio, that some stations are weak and some are LOUD.... we've grown accustomed to AGC in our superhets and don't realize how spoiled we've become until we listen to a receiver that doesn't have an AGC circuit. Also, it seems to be easily overloaded by strong nearby signals.
Like the crystal radio, it seems to come alive at night, but may not pick up much of anything during the day (again-- AGC normally takes care of the gain adjustments between day and night reception). But I still recommend you build a Reflex! It will amaze you that such a simple radio can work so well.
A very simple idea
Thursday, February 25, 2010
Recently one of my young friends, wanted to listen to SSB signals on the 40M band. We built a VWN QRP and placed it near a 3 band philips radio. The long wire was connected to the telescopic areal of the radio. ( Approximately 10 meters) . We hanged the other end of the wire on a nearby tree. The received signal strength on the radio showed a quantum jump. We slowly tuned around 40m band and located some SSB signal. Then the VFO was tuned to the receiver frequency. The SSB signals became very loud and clear. Then we changed the radio and tried the experiment on a cheap Chinese radio ( 10 bands ) . This radio was having a frequency counter. It also received the signals beautifully.
If you are interested in ham radio, this can be your first project. A big thanks to OM Vasanth VU2VWN for designing such a wonderful VFO. Here are the details of the VFO,
Wednesday, February 24, 2010
Build an Air Variable Capacitor
IntroductionVariable capacitors are useful in a lot of situations. But adjustable plate capacitors bigger than 1000 pF are difficult to find, and those that are available tend to be inconveniently large. There are several good tutorials on the Internet for building rotatable air variable capacitors like those found in old AM radios. However, because it's difficult to cut sheet metal into a curved shape while keeping it perfectly flat, the plates have to be kept far apart, which gives them a very low capacitance. The round shape also tends to be bulky. I woke up one day and decided to invent a new design that would be (1) compact and (2) have a high capacitance (preferably above 5 nF). It should be easy to construct using readily available parts. Should be easy, I thought. Next time maybe I'll know better.
|Element||Aluminum 6010||Aluminum 6011|
The idea was to make the plates from the aluminum in a Venetian miniblind, and use an immobilized screw system to convert rotation into translational movement. As the table above shows, Venetian blinds use a high-silicon containing aluminum alloy such as 6010 or 6011. This makes them much springier than ordinary aluminum, allowing them to be easily cut with scissors or a paper cutter without deforming their overall shape.
As shown in the equation below, the capacitance of a flat plate capacitor depends on the area A, the number of plates (n), and the relative permittivity (εr), also known as dielectric constant, of the medium between the plates. Capacitance is also inversely proportional to the distance d between the plates, which means that the biggest improvement can be made by moving the plates closer together. The equation for a capacitor is
where dimensions are all in meters. This is where the great idea of using Venetian blinds comes in. The paint coating on my old Venetian blinds was about 10 microns (0.01 mm) thick. The distance between two painted plates would then be only 20 microns, making the denominator really small. The paint that separates the plates would also have a dielectric constant much higher than air, increasing the capacitance by another 3 to 4-fold.
Not much to the theory, actually.
The mechanical part turned out to be the easiest. A screw is threaded through a large Plexiglas or Delrin (polyacetal) nut that is prevented from rotating by the plates and by two height-adjusting screws. When the lead screw rotates, rotational motion is converted to linear motion which pulls the capacitor plates farther apart.
For the lead screw, take a five-inch piece of 8-32 threaded steel rod and grind the threads on both ends, leaving a 10.5 cm threaded area as shown in the diagram below. Grind a flat area on the front end for a knob. Cut two 1-5/8" wide pieces of 1/16" x 1 inch angle aluminum for the front and back. Drill holes just big enough for the thread-free part of the lead screw to fit through. The rod must fit tightly in those holes. Mount the aluminum angles underneath the plastic body using countersunk holes for three 4-40 screws in the front. In the back, use two 1-1/4" long 4-40 screws to hold the aluminum angle to the body. They will also hold the plates.
Side view showing four plates.
The other side, viewed from same orientation.
Cut a 3/8 x 5/8 x 3/4" piece of Plexiglas or Delrin and tap an 8-32 hole through it lengthwise to make a square plastic nut. When drilling Plexiglas, drill very slowly, otherwise it will heat up, melt, seize, and possibly crack. If you use metal instead, you will need to find a way to insulate the aluminum plate from the main screw, so the capacitor is not shorted out.
Closeup of nut assembly. A gear and rack system or ball screw system could also have been used.
Attach the 1-1/2 x 3/4 x 1/16" aluminum platform to the plastic nut using two 3-48 screws. Mount two one-inch long 4-40 screws on the aluminum platform so that they are parallel to the two screws in the back. These will hold the front plates. Tap the plastic nut and the aluminum platform and add a couple of height adjusting screws as shown. These prevent the nut from rotating as the knob is turned.
A spring is essential in order to prevent play in the main screw. Cut a rectangular flat spring from spring metal. Rather than trying to bend it, which will only fatigue the spring, leave it flat and drill two holes in it. Use a 2-56 screw and nut to attach it to the back. The spring in the photo below was from an old wind-up clock that I took apart in 1967. That spring sat around for 41 years before I finally found a use for it. An alternative source of spring metal is those blue metal bands that are used in shipping.
To drill a hole in the spring, place a block of aluminum underneath and clamp the spring to it.
Closeup of spring.
Next take an 8-32 nut and grind it perfectly flat on one side. Grind the edges off to make it round. Then fix it in place on the main screw by drilling a small hole and screwing a 00-90 screw through it, or by jamming it against a second nut, so it extends about 1 mm away from the start of the threads. This provides a flat surface to press against the spring. If this is not present, the point where the screw threads start will always be slightly uneven, and the screw will move forward and backward slightly as it is turned.
The finished screw should turn evenly and have no back-and-forth or front-and-back play, and the plastic nut should not rotate when the knob is turned.
Unfortunately, my great idea of using the paint from the painted Venetian blinds to separate the plates didn't work as planned. If I simply cut the plates with scissors, the edges were never flat enough, and the uninsulated edges would contact each other. So I had to strip the paint off and coat them. But there's a trick to get around this.
The stripping can be done using ordinary paint stripper, which contains methylene chloride. Be sure to do this outdoors or under a fume hood, as this stuff is highly toxic. Other chlorinated solvents, like chloroform, will also work. Solvents like lacquer thinner, turpentine, isopropanol, hexane, acetone, and ethanol do not work.
Cut the strips last, once the mechanism is finished. The strips must overlap by at least 1 mm at the widest setting. Cut a few extra ones, because when you drill the hole, the top and bottom one will probably get all crunked. Clamp the strips together and slowly drill one hole through all of them simultaneously. Then screw them together through the first hole before drilling the second one. To get rid of the curvature, flatten the pieces by gently rolling them against a cylindrical object like the axle of an empty wire spool.
Spacer and plate.
Next, make a large number of spacers from a stripped piece of aluminum. Make them oversized so it's easier to drill the holes. Drill the holes in the same way as for the plates, then cut the spacers to the correct size with a pair of scissors.
When I started this project, I hoped the paint on the Venetian blinds would be sufficient to keep them apart. This turned out not to be the case. If you make the plates from the inside portion of a vane, it's necessary to add more insulation to prevent the plates from short-circuiting. This can be done by adding thin plastic sheets between the plates, or by anodizing the plates. Spray-painting the edges did not work very well.
Another trick is to use only the ends of the vanes. These ends are already flat, so there's no need to strip and coat them. Plus, they are rounded, which reduces arcing if you use the capacitor at a high voltage.
Method 1: Anodizing
Anodizing is probably the best way to insulate the plates. Anodizing leaves a hard black or gold coating that doesn't conduct electricity. The coating is thinner than a plastic sheet (5 to 18 microns, or 0.005 to 0.018 mm thick, compared to at least 0.025 mm for 1 mil plastic). This means you will get a much better capacitor.
Here is the procedure for anodizing. Chromate anodizing works better than plain sulfuric acid anodizing for this type of aluminum.
- Strip the paint by soaking the aluminum strips for 5 min in chloroform.
- Wash the strips again in chloroform and wipe any remaining paint with a paper towel. Wear gloves and avoid touching the surface of the plate.
- Screw the plates together, separating them with metal washers. Connect the plates to the (+) terminal of a DC power supply capable of producing several amperes at 15 volts. One amp is enough to anodize two square inches in about 30 min. Less current will take proportionally longer time.
- Connect the (-) terminal to a piece of aluminum. Put both electrodes in a 100 ml beaker, keeping the end with the holes out of the solution. Make sure that only aluminum is in contact with the solution.
- Fill the beaker with 15% sulfuric acid containing 1% potassium dichromate. Place the beaker in a larger water-filled container to absorb the heat.
- Apply sufficient current to create about 15V across the terminals. Hydrogen and oxygen gas will be produced, and the solution will heat up. It must be at or slightly below room temperature for best results.
- Check the resistance of the coating, then rinse and seal the pores in the aluminum oxide coating by boiling in water.
- Neutralize the chromate and dispose as toxic waste.
One advantage of anodizing is the high dielectric constant of aluminum oxide (9.34), which is higher than vinyl (3.5-4.5), polypropylene (1.5), air (1.0006), and even glass (4-7). Since capacitance is proportional to the dielectric constant, anodized aluminum should theoretically have 9 times higher capacitance than an air capacitor.
Before installing the plates, add washers if necessary so that both rows of plates are exactly the same height. Then add one plate separated by two spacers on each side, alternating sides so the plates are interleaved. Add another No. 4 washer on top and screw in place. Check for shorts.
Method 2: Plastic sheets
Alternatively, you can use a plastic sheet to separate the plates. This is the easiest and most reliable method if a high capacitance isn't needed. The best material is 3M 9457 Adhesive Transfer Tape, which is 1 mil (0.0254 mm) thick. Some types of packing tape are also 1 mil thick. Completely strip the paint from the plates and stick one piece of the tape on top of each plate, making sure to avoid bubbles. The tape should extend about 1/2 mm from the edge. Leave the part of the plate near the holes uncovered. Put tape only on one side of the plate. Install the plates, using two spacers between each plate, and make sure the plates are interleaved (see diagram). Add a No. 4 washer on top and screw in place. With 25 plates total, this capacitor could be adjusted from 90 to 4500 pF.
Method 3: Use the ends
An even easier way is to use only the left and right ends of the Venetian blind to make your plates. Drill the holes on the side away from the end of the vane. Then dip the side with the holes in paint stripper to remove the outermost 7 millimeters of paint. The rounded outer edges are sufficiently flat that there's no danger of a short circuit as long as the plates are flat, clean, and have a uniform shape. If short circuiting should occur, coat the edges of the plates using a Magic Marker to insulate them. Although this method of making plates is easy, it has two disadvantages: first, you will use eight vanes from your Venetian blind instead of only two or three; and second, the uninsulated edges are fairly close together, so high voltage could still cause arcing. For receiving antenna applications, this is not a problem. The paint is only about ten microns thick, so you can get a high capacitance with only a few plates. Two spacers per plate is sufficient.
Side view showing plates.
With 16 plates, this capacitor could be adjusted from 90 to 3400 pF. By increasing the number of plates to 24 and reducing the total number of spacers from 48 to 36, it was possible to increase the maximum capacity to 8400 pF. However, it is not recommended to use fewer than two spacers per plate, because it causes the plates to bend.
The Venetian blind variable capacitor has several advantages over the conventional type:
- It is very easy to get a very high capacitance and a good high/low capacitance ratio. With only 24 plates, it had about 4500 pF. It had a minimum of about 90 pF, for a ratio of 50:1. This compares with only 15-365 pF for commercial variable capacitors of roughly the same size. If you only need 360 pF, three or four plates should be sufficient (or, you could make the plates smaller).
- Multiple turns eliminates need for gear reduction and allows highly accurate tuning. No expensive gear reduction system is needed.
- It is easy to add or remove plates. More plates can be slapped on with four screws for higher capacitance. This makes it easy to try different designs or coating methods.
- Plastic nut provides smooth, silent rotation.
- It's more compact and fewer plates than a conventional rotating design because the coated plates allow a closer spacing. The overall size was 4.5 x 1.75 x 1 inch. With a little effort, the height could easily be reduced to half of that, or even less. The design is also potentially more precise than the rotating plate design, making it useful for other applications.
- No worries about plates bending and touching each other because they are already touching each other.
- Easier to construct because the plates are rectangular instead of circular. Venetian blinds can be cut with scissors.
- Requires less torque to turn than a geared rotating design. This makes it easy to attach a motor for remote control.
- The knob turns in the direction you would expect: clockwise for lower capacitance and higher frequency.
There are also some disadvantages:
- More rotations are required to cover the entire range.
- For high voltage applications, the plates have to be stripped and anodized or coated with insulating tape. Anodizing is a fair amount of work.
- This turned out to be a lot more work than I originally thought.
Improved version of capacitor
To get better precision, an Acme thread should be used, and the moving platform should be supported on both sides. I made a 1/4"-16 Acme tap and used it to create a nut out of Delrin. On the opposite side, there is another piece of Delrin that slides freely over a 1/4" polished steel rod. Screw threads are cut on each end of the rod to hold it in place. I also used 1/16" steel instead of aluminum. With 20 plates on each side (40 plates total), the capacitance was 350-6530 pF. With 10 plates on each side (20 plates total), the capacitance was 190-3300 pF.