High-Voltage Projects for Fun and Science

Home , Autotransformer, Charles Grafton Page, Induction coil, Nicholas Callan, Voltage transformer

High-Voltage Projects For Fun and Science

Book 1

CONTENTS

Square Tesla Coil 2

Jacob's Ladder 7

Classic Induction Coil 9

Electronic Dazer 15

High-Voltage Capacitors 17

High-Voltage Pulse Generator 24

Negative Ion Generator 30

1. GERNSBACK PUBLICATION

As time passes, the availability of parts and sundries decreases or By the Editors of Electronics Now price changes occur. Contact the suppliers and distributors listed and Popular Electronics herein for up-to-date information before you proceed with building Copyright cg Gernsback Publications, Inc., 1992 or assembling the projects. BY VINCENT VOLLONO

¡kola Tesla, considered voltage, which collects on the by some to be the top-capacitance sphere Ngreatest inventor of T HE where causesit an avalanche the electrical age, is today breakdown of the surrounding best remembered for his fas- air, giving off a luminous dis- cinating power-transmission S Q U A R E charge. experiments, using his famous In order to get maximum Tesla Coil. In his original experi- output from the Tesla Coil, cer- ment, he was able to transmit tain conditions must be met. electrical energy without wires First of all, the primary and to light incandescent lamps secondary resonant frequen- located over 25 miles way. cies must be made equal by Today, most similar circuits— tuning the primary coil, L3. like the Tesla Coil described in CC That's accomplished by tap- this article—are used for edu- ping L3 at points along the coil cational and experimental with a clip lead. purposes. Unlike many of the In addition, the setting of modern versions, our circuit the spark gap greatly effects feeds AC to a power trans- the output of the Tesla Coil. former capable of outputting Our Tesla Coil is designed to about 3-kVAC at 20 milliamps. use either a stationary spark The output of the transformer gap or an optional rotary is sent to a primary coil, and is spark gap; both of which must magnetically coupled to a be adjusted for maximum out- secondary coil with a top ca- put. (We'll discuss the rotary pacitance. And theif primary spark gap a little later.) coil properlyis tuned, a spec- If L3 and L4 are coupled too tacular high-frequency, high- co close, coil efficiency is re- voltage output producedis at duced; over-coupling pre- the secondary coil. vents the circuit from resonat- ing at maximum efficiency. Circuit Description. Figure 1 SQUARE That also causes a breakdown shows the schematic diagram between L3 and L4, which can of the Tesla Coil circuit. The cir- produce arcing between the cuit consists of little more than two coils. By increasing the a few coils, a step-up power TESLA coupling between L3 and L4, transformer, and a capacitor. the amount of energy in- Power from an AC wall socket creases in L4 until a "critical is fed to transformer T1 (a small coupling" is reached. neon-sign transformer) which C OIL In addition, the Q of the coils steps the voltage up to about is very important (the Q of a 3000-volts AC. coil is equal to its inductive reactance The stepped-up output of T1 is fed Build this unusual version divided by its resistance). The lower the through Ll and L2 across Cl, causing it Q, the higher the efficiency of the coil. to charge until enough power storedis of Nikola Tesla's most The primary coil was made from a few in the unit to produce an arc across the famous experiment! turns of aluminum grounding wire (so its spark gap. The spark gap—which mo- resistance is very low). The secondary mentarily connects Cl and L3 in paral- effect. The energy fed through L3 is has many more turns of fine magnet lel—determines the amount of current transferred to L4 via the magnetic cou- wire, which by its very nature exhibits a transferred between Cl and L3. pling between the iwo coils. (Because higher resistance than does the wire The arcing across the spark gap of the turns ratio that exists between L3 used in the primary coil (L3). sends a series of high voltage pulses and L4, an even higher voltage is pro- through L3, giving a sort of oscillating duced across L4.) Coil L4 steps up the Rotary Spark Gap. The rotary spark

- 2 - PL1 PL2 METAL OUTPUT you avoid metal enclosures. Because SPHERE the circuit consists of very few parts, its J3 PL3 components can easily be hard-wired Li J1 J2 10p.H together within the housing, using Fig 1 as a guide. There nothingis particularly SPARK critical about the layout of the circuit. Cl* GAP L4* Just be sure to maintain adequate .006 L3 spacing between the individual com-

PARTS LIST FOR THE *SEE TEXT TESLA COIL

Li, L2-10-11.FI, AC-line filter choke L3-6 turns, #I0 aluminum grounding Fig. 1. The Tesla Coil circuit consists of little more than fewa coils, step-upa power wire, see text transformer, and capacitor.a L4-348 turns, #24 magnet wire, see TO text 02 TESLA T1-3-kV, 20-mA, neon-sign 1 A COIL T2 transformer 12V C1—.006-µF, 5000-WVDC ceramic 700mA J4 J5 capacitor (see text) F1-10-amp fuse JI-J3—Banana jack PLI-PL3—Banana plug PL4-3-conductor AC power plug with line cord Metal output sphere, plastic or wooden enclosure, "L" brackets wire, solder, wood, hardware, etc. Fig. 2. Here's the schematic diagram for the optional rotary spark gap. The circuit is made up of 12-volta power tivansformer, bridgea rectifier, 4700-p,Fa capacitor, PARTS LIST FOR THE and 12-volta DC motor. ROTARY SPARK GAP QI—TIP41 NPN silicon power transistor gap is a simple add-on circuit for the which in turn varies the rotating speed Q2-2N3055 NPN silicon power Tesla Coil, consisting of a variable DC of the motor. transistor power supply and a small, 5000 rpm, The rotary spark gap has a stationary BRI-1-amp, 100-Ply, full-wave bridge DC motor. The circuit allows you to vary post (two screws) mounted on a small rectifier the output of the Tesla Coil by adjusting square of perfboard, which face a rotor MOT1-12-volt, 5000-rpm, DC motor the rotating speed of the motor. A ro- (another perfboard square on which T2-12-volt, 700-mA, step-down power transformer tary gap is far more efficient than the four screws are mounted and elec- R1-1000-ohm, /14 -watt, 5% resistor stationary gap because the stationary trically connected together with bus R2-10,000-ohm potentiometer gap could cut out, requiring that the wire). The stationary posts and the rotor J4, J5—Banana jack gap be readjusted. posts are positioned as close as possi- PL5-2-conductor AC power plug with Figure 2 shows the schematic di- ble. The movement of the rotor makes line cord agram of the rotary spark gap, which is and breaks the gap giving maximum Perfboard materials, plastic or wooden assembled as a separate unit. The cir- impulse power, and will not cut out theif enclosure, wood, wire, solder, cuit madeis up of a 12-volt power trans- stationary and rotor posts are set prop- hardware, etc. former, a bridge rectifier, a 4700-p.F erly. capacitor, and a 12-volt DC motor. The rotary spark gap connectedis to Note: A 3-kV, 20-mA, neon-sign Power is delivered to the circuit via a the Tesla Coil through separate wires transformer (TI) is available as part 117-volt AC line cord, and fed to trans- and banana jacks (J4 and J5). When number 720-391 from N. Glantz & Son, 218 57th St., Brooklyn NY former12 (a 12-volt, 700-mA unit), which the circuit is powered up, the current 11220, Tel. 800-522-5120. The 12-volt provides a 12-volt AC output. The output that normally travels through the sta- DC motor for the optional rotary spark of the transformer fedis to BR1 (a 1-amp, tionary spark gap on the Tesla Coil re-is gap (MOT!) is available from H&R 100-PlVfull-wave bridge rectifier), which routed through the rotary gap via J4 Corp., 401 E. Erie Ave, Philadelphia, converts the AC input to provide 12- and returned to the Tesla Coil via J5. To PA 19134, Tel. 800-848-8001. Contact volts DC for the operation of the motor. make the connections plug PL1 into J1, those companies directly for pricing, The output of the rectifier fedis to the PL2 into J4, and PL3—which in Fig. is1 shipping and handling charges, etc. base of transistor Q1, which along with used to connect the output of the sta- Plans for making your own high- Q2 forms a Darlington pair. The output tionary spark gap to L3—into J5. voltage capacitors, such as the one of Q1, which controls the bias pre- required for CI, are available for $3.50 postpaid from Alegro Electronic sented to the base of Q2, is controlled Construction. The author's prototype Systems, Mine3 Mountain Road, by potentiometer R2. Potentiometer R2 of the Tesla Coil was built into a large Cornwall Bridge, CT 06754; ask for is used to adjust the base bias on Q1, plastic enclosure; because of the high item number UHVC400. thereby varying the current through Q2, voltages involved, it is imperative that

- 3 - 3/16 INCH CARRIAGE ponents and wires. BOLT Start by drilling holes in the enclosure "L" BRACKET to pass wires through and for the panel- L2 mounted components. In the author's prototype, three sides of the enclosure Li Ji were outfitted with appropriate sized holes. Ai/8-inch hole was drill in one side of the enclosure, through which a ground wire connects to L3. On another side of the enclosure, holes were drilled to accommodate a dowel rod (which partis of the stationary spark gap), a fuse holder, and the power cord. On the third side, three holes were drilled for banana jacks. It will also be necessary to drill holes in the bottom of the enclosure suitable for Ti's 1/2 INCH WOODEN mounting hardware. DOWEL Begin assembly by mounting the Ti 1/2 INCH FUSE power transformer on the bottom of CERAMIC HOLDER the enclosure. Next connect 10-µHa AC STANDOFF filter choke in series with each of Ti's Fig. 3. The stationary spark gap is made from two 3/I6-inch carriage bolts supported secondary leads, and then connect by "L" brackets mounted to spacers so that they face each other. One bolt is made the free ends of each coil across C1 stationary, while the other is made adjustable so as to vary the spark gap. (see Fig. 1). Note: In the author's prototype, C1 is really two .012-µE 2500-volt AC capacitors that were wired in series to create C1 (giving the capacitor an effective rating of .006 p.F at 5-kV AC). youIf use the same scheme, keep the con- L4 necting leads between the capacitors as short as possible. After connecting the capacitors together, cover the gap between the Iwo units with non-con- NO CONNECTION ductive tape, and connect the jerry- rigged unit in the circuit as shown.

Stationary Spark Gap. The stationary gap can be made from two 3/16-inch carriage bolts (see Fig. 3). One bolt is stationary and the other one adjusta-is ble so that canit be used to vary the spark gap. A /12 -inch wooden dowel is attached to the bolt that is to be adjustable, allowing adjustments to the gap to be made from outside the projects enclosure. The wooden dowel veryis important; one does not want to adjust the gap by touching metal (or any other conductive device), since the gap is adjusted with the Tesla Coil in operation. The bolts that form the spark gap are TERMINAL TO L2 supported by Iwo "L" brackets mounted BLOCK to spacers so that they face each other Fig. 4. The inner winding of L3 should have an area of about 6-inches square. When making (see Fig. 3). The stationary post of the the coil be careful that you do not form wire loops, instead of the continuous coil spark gap connectedis to J3, and the illustrated here. movable bolt connectedis to J2. a somewhat unusual look, the original position. Iwin-turnscrew type barrier Primary Coil. The original primary coil (more-or-less round) primary coil was blocks, which were mounted on four (L3) was made from turns6 of #16 alu- replaced with a square version made blocks of wood. minum grounding wire in a pancake from heavier #10 aluminum grounding The wood blocks (with barrier blocks style winding. However, to give the unit wire. The Coil was formed on four 6- attached) were mounted to the top of

- 4 - the enclosure near the edges, and bus wire was then connected to the barrier blocks to form the coil. Note: The wires do not have to be fed through the barrier strips because of the Iwin-turnscrew arrangement. The wire can be cut to the proper size and screwed into the terminal strip to form the primary coil. The inner dimensions of L3 should be about 6-inches square. When making the coil be careful that you do not form wire loops, instead of the continuous coil illustrated in Fig. 4. When you are finished with the coil, there should be one unoccupied screw terminal at the center and another at the outer rim of the coil. The unoccupied terminal at the outer rim of L3 is connected to ground via a wire that's brought out through a hole in the enclosure. The unoccupied terminal at the center of the coil leftis floating.

The Secondary Coil. To fabricate the secondary coil (L4), the author wound ROTOR about 348 turns of #24 magnet wire STATIONARY POST onto an 8 1/2-inch length of 3 1/2-inch di- Fig. 5. The rotor of the rotary spark gap is made from smalla pedboard square containing ameter PVC tubing. That works out to be four #6 screws that are connected through bare bus wire. The stationary post consists of about 48 turns per inch, covering 71/4 another perfboard square (of equal size), containing two #6 screws that are not tied inches on the PVC tubing. together elecrically. The coil was wound by hand using a simple jig—which consists of little more ball. mattersIt not youif use solida ball gap tois build and mount the electric than a stand for the wire and another or a hollow ball; they will both work motor support. theIn author's prototype for the coil form. When winding the sec- equally well as long as their surface (see Fig. 5), the motor mount was made ondary coil, try to keep the winding as areas are equal. from small blocks of wood assembled even as possible without overlapping The size of the sphere effects the sec- in a "U" shape. woodenA mount alsois any turns. ondary's resonant frequency, so youif used to secure the stationary wood After the coil has been wound, apply use a larger sphere, willit be necessary post in place. clear varnish or polystyrene (Q-DOPE) to retune the primary coil for maximum The distance between the rotor to hold the coil windings in place, and output. A bigger sphere collects more screws and the stationary post screws to help insulate the coil. Next drill smalla energy, causing toit give off a higher must be as small as possible without hole in the center of the Tesla Coil en- output. So experimenting with the top touching in order for the unit to function closure lid, and thread the lower lead of capacitance is highly recommended. properly. After mounting the motor L4 through the hole and connect to it mount in the enclosure, place the the ground end of L3 (as shown in Fig. 1) The Rotary Gap. The rotary spark gap motor in the mount and secure it in and mount L4 in the center of L3 (see is not necessary to the operation of the position with epoxy. Fig. 4). The secondary coil is then se- Tesla Coil. So, youif do not wish to build Next assemble the motor controller cured in place with glue. the optional rotary gap, skip this sec- circuitry on a piece of perfboard, using As an added measure of protection, tion. Fig. 2 as a guide. Note that 12, R2, J4, you can also place clear plexiglass 4- The rotor of the rotary spark gap is and J5 aren't mounted to the perf- inch OD (outside diameter) tubing over made from a small perfboard square board, but instead are mounted to the the secondary coil as seconda layer of on which four #6 screws are mounted rotary-gap enclosure. Once the con- insulation. You might also seal the in- and electrically connected through troller board is assembled check your sulating tube and place mineral oil in it, bare bus wire. The stationary post con- work for wiring errors. allIf checks out, thereby further increasing the tubing's sist of another perfboard square (of solder wires to the appropriate points insulating properties, but that's not nec- equal size), containing Iwo #6 screws on the board for connection to the off- essary for this type of Tesla Coil. that are not tied together electrically. board components. Set the board to The screws of the stationary post are the side for now; willit be installed in a The Output Sphere. The output instead taken out to J4 and J5. Per- moment. sphere—a 11/4-inch steel ball on top of fboard is specified because the holes Mount the off-board components on a plastic spacer—also serves as the top in perfboard make easyit to align the some convenient spot on the en- capacitance. An important point here screws on the rotor with those on the closure. Mount R2 so that you'll have is that the surface area represents the stationary post. easy access to its wiper. Jacks J4 and J5 capacitance not the inner area of the The first step in building the rotary can be mounted in any desirable loca-

- 5 - grounded. You must use a 3-conductor clip with the power off. Tuning L3 and SQUARE TESLA COIL AC power cord that groundedis (earth adjusting the spark gap greatly effects grounded) in the Tesla Coil itself. Do not the output of the Tesla Coil. youIf place tion. Before mounting 12, make sure touch the Tesla Coil while it's in opera- a grounded wire near the output that the transformer leads are long tion. However youif want to show-off sphere, you should get 3- to 4-inch enough to connect to the perfboard your creation, a fluorescent lamp may sparks. assembly. be placed near L4 to demonstrate the If you are using a rotary gap, make After mounting T2 in the enclosure, ionizing power of the Tesla Coil. sure that the screws on the rotor and the mount the perfboard assembly on the Only use properly rated compo- screws on the stationary post are as enclosure using standoffs, and then nents. Do not use an overrated power close as possible. Remember, the complete the wiring between the per- transformer. 3-kVA transformer with 2-a speed of the motor effects the output, fboard assembly and the off-board kV AC capacitor outis of the question. so adjust the motor speed with the vari- components. With that done, plug in An overrated capacitor (for instance, a able power suppW. the line cord and rotate the wiper of R2, 6-kV AC unit) is fine in the circuit. Re- There should be no arcing anywhere. making sure that as you do the motor member the capacitors are AC rated All arcing must be corrected or you'll speed increases and decreases. theIf not DC rated. burn out the turns in the secondary. L3 If circuit does not operate as described, it The rotary gap will work well with this is too close to L4, arcing can occur. You will be necessary to recheck your work, unit, but it may not work well with a may place a 4-inch OD plexiglass tub- correct any errors found, and try it larger unit. A larger unit will require that ing over the secondary coil to help pre- again. everythingIf checks out, the ro- the rotary gap be redesigned. You must vent arcing between L3 and L4. tary gap complete.is also protect your eyes: Do not stare at Be aware that corona discharge (a the stationary or rotary spark gaps; bluish-purple ionization, of the air Caution!!! The most important part of doing so can cause eye damage. around the Tesla Coil) can cause using the Tesla Coil safely.is Never tune breakdown along the secondary coil, (adjust the tap on L3) the Tesla Coil Operating the Testa Coil. With the and loss of power at the output of the when power is applied to the circuit. unit completely assembled, make sure sphere. Proper insulation of L4 will limit corona discharge. You may also notice an output at the top of the secondary coil coming out of the sides. That will take away from the output at the sphere, you could place several layers of tape (Turn off the power first!) around the upper-portion of L4, until the output from the sides of the Tesla Coil is reduced. In operation, the Tesla Coil emits ozone gas, which in large quantities can be dangerous. So use the Tesla Coil in a well ventilated room, and do not operate forit periods of more than to 3 5 minutes at a time. In addition, the Tesla Coil emits a fair amount of Radio-Frequency Interfer- ence (RFI). Coils Ll and L2 help to limit the amount of high-voltage kickback introduced to the AC power line, and help to prevent the high voltage kickback form damaging the power transformer. Even with such precautions, RFI will still be generated at the Here's an inside view of the rotary spark gap; note the tight spacing between the pedboard spark gap and the output -of the Tesla rotor and the stationary post. The wires coming from the stationary post are connected Coil. RFI will effect both AM radio and to .14 and .15, through which the rotary spark gap is connected to the Tesla Coil circuit. television reception. That's why you should not operate your Tesla Coil for Use phenolica plastic box or woodena that all the components are properly more than a few minutes. box to house the Tesla Coil and the ro- installed and oriented. youIf are using The Tesla Coil anis excellent introduc- tary gap—avoid metal enclosures like the stationary gap, start with a gap dis- tion to high-voltage, high-frequency, the plague. In addition, it is recom- tance of about 1/4-inch and tune L3 at and tuned circuits. And after building mended that you use one hand only any point on the third turn from ground. this one, you may wish to build a larger while working with high voltage, and At that point turn the power on; you unit. The author does not recommend wear rubber soled shoes to reduce the should get an output at the sphere. Ad- building a larger unit until you've potential of shock hazard. just the spark gap for maximum output. learned enough about such circuits, The power transformer, capacitor Tune L3 for maximum output, by and the safety precautions that must and coils L3 and L4 must be properly changing the position of the alligator be followed when using them. MI

- 6 - JACOB'S LADDER

A climbing electric arc has held the imagination of science-fiction fans as the symbol for an eerie laboratory!

JAMES, NICOLE, and DWIGHT PATRICK, Jr.

IN MANY SCI-FI AND HORROR FLICKS, ESPECIALLY THE STOCK "FRAN- kenstein" variety, along with weird sound effects and the like, movie producers always feature the fantastic visual effects produced by Tesla coils, van de Graaff generators, and Jacob's Ladders. Of those three devices, the Jacob's Ladder theis easiest to build. With low-currenta neon-sign transformer, converteda flyback transformer, converted auto spark coil, or other similar transformer, you can whip together your own Jacob's Ladder in less than an hour. Because the ladder is so simple, there's no need for detaileda parts list or schematic.a We tell you how to build one as we reveal the theory of operation.

Getting started As we can see in fig. 1, a Jacob's Ladder provides fantastica visual effect. A beautiful electric arc hisses its way up two diverging wires, providing fascinatinga and downright scary effect. The arc starts at the smallest distance between the vee electrodes (Fig. 1-a), and "walks" up the widening gap toward the top of the electrodes (Fig. 1-b). Why does the arc walk up the vee electrodes? You would expect that when the arc starts to jump across the narrow gap at the bottom of the electrodes, that wouldit stay there where the electrical resistance between the electrodes is lowest. What actually happens is that the arc heats the air it passes through, causing toit rise. Because the heated air is ionized by the high- voltage arc, providesit verya low-resistance path, so the current path (the arc) rises with the warm air. Eventually the arc reaches the top of the ladder (the electrodes) and bows upward creating an electrical path that gets longer and longer. At the point where the resistance at the bottom of the electrodes is less than that of the arc-path, the upper arc stops, and newa arc begins at the bottom of the ladder. Thus, what is seen is a continuous climbing arc that disappears at the top of the ladder and reappears at the bottom. It's all lota of fun to watch, providing that you don't poke your finger between the electrodes or get your nose too close. To build Jacob'sa Ladder, you need high-voltagea source that

-7- the llaaddddeerr arcs at the initial setting, move the wires apart until an arc is just sustained. Placing the wires too close together will ruin the transformer over time. When you have established the arc, if it does not move up between the two diverging wires, they must be adjusted in or out.

Better safe than sorry Once your Jacob's Ladder is up and running, a clear Plexiglas or acrylic cylinder around the entire unit will prevent the unthinkable from happen- ing (see Fig. 3). The clear plastic housing should have vent holes at the top and bottom to allow heated gasses to escape, but not so large that the FIG. 1—HERE IS A TIME-EXPOSED PHOTOGRAPH showing the development of an arc at the bottom of the ladder as it climbs to the top. Display Jacob's Ladder in darkeneda room smallest child in the family can get his for best viewing. or her mitts or anything else inside. Generally, such cylinders can be pur- chased from most plastic-supply nected via high-voltage TV anode houses rather cheaply, and are well hook-up wire to two pieces of no. 12 worth the added protection. copper wire used for the vee-shaped You do have to keep in mind that such electrodes that protrude from por- high voltages are extremely dangerous. celain insulators. When removing the and you certainly don't want to come in insulation, be sure not to nick the contact with them. • wire. The connections between the transformer's secondary and the bottom of the insulators must be kept as short as possible and void of sharp bends (see Fig. 2). Exposed high-voltage points were given coata of RTV silicon rubber to prevent any arcing inside the enclosure.

Adjustment and operation The last step in getting your Jacob's Ladder to work is the adjustment of FIG. 2—NOTICE HOW NEAT THE WIRING the vee electrodes. DO NOT MAKE IS. RTV cement is used around the high- ANY ADJUSTMENTS WHEN THE voltage connection points. The line cord's third lead is used to ground the case. UNIT IS TURNED ON. The unit should be turned off and unplugged can deliver around 10 kilovolts at 30 when making adjustments, to prevent milliamperes or more—such as neon-a accidental electrical shocks. sign transformer or high-voltage The electrodes must be close power-supply transformer. The lower enough at the bottom to establish the the current output, the less chance spark or arc-over, with the wires gent- there is for building fatala shock haz- ly angling away from one another to ard. The higher the voltage, the larger form the "V." The initial distance at the separation you'll be able to make the base to start the spark will vary at the top of the ladder's electrodes. with the voltage applied, humidity, The high-voltage source should be altitude, etc.; so, it's pretty much placed well-insulatedin a case or cab- done by trial and error. Start with the inet. If the case is metal, shouldit be wires at the base about an inch apart adequately grounded. In the pho- when using 10 kilovolts or more, and tograph in the opening of this article, move them closer in small increments FIG. 3—A PLASTIC SHIELD SHOULD BE a 7-kilovolt transformer was placed in until an arc is established. But re- USED to prevent inquisitive people from a /14 -inch Plexiglas case with the pri- member to kill the power before each accidentally touching the arc or the elec- mary winding connected via fuse and adjustment. trodes. Be sure to include a vent hole at the top and bottom of the shield so that switch to the 117-volt AC input. The When the distance is correct, the gasses produced by the arc are allowed to output of the transformer was con- arc should start. On the other hand, if escape.

- 8 - Learn the history behind the induction coil and the men that all invented atit once, and then have littlea high voltage fun yourself

BY STANLEY A. CZARNIK

he final quarter of the 19th cen- wireless-communication systems. And Of course, this list of discoveries in tury marked a period of in 1897, Joseph John Thomson an- physics is selective and presented Textraordinary scientific activity. nounced the discovery of the first sub- here to make a point. Every single in- Excitement was everywhere. Creative atomic particle: the electron. novation in the sequence would have intellects flourished as the theory of been difficult, notif nearly impossible, evolution matured, the camera inno- were notit for a single piece of labo- vated the study of the stars, and Freud WARNING!! ratory equipment: the high-voltage contemplated the meaning of his This article deals with and involves subject induction transformer. matter and the use of materials and sub- dreams. Just how, when, and where the in- stances that may be hazardous to health and During that same period, important life. Do not attempt to implement or use the " duction coil was invented are not contributions in the area of experi- information contained herein unless you are • easy questions to answer. In fact, the mental physics were made. For exam- experienced and skilled with respect to such > 1, entire matter submergedis in a good subject matter, materials and substances. Nei- ple, in the late 1800's, Heinrich Hertz measure of ambiguity and confusion. ther the publisher nor the author make any • 4 produced the first radio waves, which representations as for the completeness or the Let's try to disentangle some of the provided a concrete illustration of accuracy of the information contained herein main historical threads. Maxwell's theory of the identity of and disclaim any liability for damages or inju- • electromagnetism and light. In 1895, ries, whether caused by or arising from the Callan's Magnet. In December, lack completeness,of inaccuracies theof infor- Wilhelm Roentgen stumbled upon the mation, misinterpretations of the directions, 1836, Nicholas J. Callan published a strange penetrating power of the X- misapplication of the information or other- brief description of a "new Galvanic ray. In 1896, Guglielmo Mcirconi filed a wise. battery" in The London and Edin- patent application for one of the first burgh Philosophical Magazine. At the

- 9 - ceeded to discuss an alternative way Journal dated May 12, 1836, Charles of doing it. He took a bar of soft iron, Page announced the construction of about 2 feet long, and wrapped it a device for "increasing shocks" in a around with two lengths of copper new way. That new way involved tak- wire, each about 200 feet long. For ing the shocks from a secondary reasons left unmentioned, Callan winding longer than the battery cir- connected the beginning of the first cuit. Recall that Callan's original coil coil to the beginning of the second. was made with two windings of equal Finally, he connected a battery, much length. So, not only was Page's coil smaller than the enormous con- made public before Callan's appara-

Page's Electrootatio Coil, April, 1838, trivance just described, to the begin- tus, it was different. In The History of ning and end of winding one. He Induction (1867), the one book he Charles Grafton Page was probably the found that when the battery contact wrote, Page described the novelty of first to construct high-voltagea was broken, a shock could be felt be- his creation as follows: "the use of a induction transformer. Page followed his tween the first terminal of the first coil longer coil for the secondary current prototype with longa series of and the second terminal of the sec- than that used to transmit the battery improvements. This "Electrostatic Coil" ond coil. is one of them. The magnetic field current" and the production of around the core attracts the metal button Further experimentation showed "shocks from a purely secondary coil at G. The motion is transmitted by the how the coil device could bring the exterior to the primary coil." curved wire at E. The mercury cup shock from a small battery up to the Page's first inductive device was un- contact point is at M. strength level of biga battery. So, Call- like most of his later models. The pro- an tried making a bigger coil. With a totype was a 220-foot long spiral of battery of only 14 seven inch plates, copper ribbon mounted inside a time, Callan was a clergyman and the latter device produced power wooden box. The coil was tapped in physics teacher at Maynooth Col- enough for an electric shock "so six different places with short copper lege, in County KIdare, near Dublin, strong that a person who took feltit strips connected to open mercury Ireland. The battery consisted of 20 the effects of forit several days." cups on top of the box. That allowed zinc plates, each about two feet Callan thought of his creation as a Page to experiment with "sub-spirals" square, and a similar number of large kind of electromagnet; but, what he of various lengths. But note: Ws still one copper canisters to hold the acid. The actually made was a primitive induc- continuous winding of ribbon, not two. system was activated by lowering the tion transformer. Page's retrospective claim to have plates into the acid by means of a used "purely"a secondary coil "exteri- windlass. Callan's cells required about The American Version. Callan was or" to the primary is somewhat mis- 30 gallons of acid. The battery was not the only one working with ways of leading. Strictly speaking, that's just quite powerful and capable of melt- increasing battery power by elec- not the case. Fact is, Page used a sin- ing thin pieces of platinum wire. tromagnetic means. A few months gle winding with multiple taps; he had The generation of higher and high- earlier, similara system had been built built, in effect, an autotransformer. er levels of electrical energy seemed in Salem, Massachusetts by a young very important to Callan, and he pro- medical student with tirelessa interest Simultaneous Invention. In August, in electrical experimentation, Charles 1836, using a coil a bit larger than the Spark gap Grafton Page. Q one just discussed, Page succeeded In a communication to Sill/man's in producing not only an electric Secondary,,

Interrupter

Pri arym /

Condenser

Key Battery Most standard induction coils equipped with automatic Neef-type vibrator mechanisms operate in similara way: When the interrupter contact points are closed, the battery current flows through the primary winding. That creates a magnetic field around the core which attracts the vibrator hammer thereby This is one of the largest induction transformers ever made. The "Great Coil" of breaking the circuit. When the primary Spottiswoode (named after its creator William Spottiswoode) had secondarya circuit current is cut off the points spring back consisting of no less than 230 miles of cable and could generate power enough to and the circuit is re-established. throw sparka some 42 inches long. shock, but a very small electric spark circuit breaker. McGauley attached a between the terminals of a second- soft iron ball to one end of pivoted ary winding. seemsIt clear that justif wire mounted over the induction ap- one person must be given credit for paratus. The ball was situated so as to the invention of the high voltage in- be within the electromagnetic field duction coil, it should be Charles produced by the core of the coil. The Page. Nonetheless, Nicholas Callan opposite end of the wire touched the still deserves a measure of recogni- surface of some mercury placed in a tion; after all, he was working along small cup. When the system was acti- similar lines with similar equipment in vated, the magnetic field near the the same year. core of the coil attracted the iron ball. In his personal history, Page himself That broke the contact at the surface notes that he was the first one, but not of the mercury, shut down the mag- the only one, to generate an induced The coil from Fisher is sealed in a netic field, and released the ball. high voltage discharge in 1836. A rectangular wooden case and comes When the wire at the other end fell spark, he mentioned, was also ob- complete with an adjustable Neef-type back into the mercury, the cycle tained by two Italian scientists, Antinori hammer interrupter. would begin again. and Linari, and published in the De- A very simple mechanism, but it cember 13, 1836 issue of L'Indicatore worked quite well and suggested itself Sanese. I have not been able to lo- elsewhere. Charles Page, without cate the original Italian reference. Yet, knowledge of McGauley's experi- it is possible, even probable, that ments, came up with almost the exact crude high voltage induction appa- same thing at just about the same ratus was being developed indepen- time—yet another instance of simul- dently in not one, not two, but three taneous invention and parallel chan- different places—America, Ireland, nels of thought. Page made the and Italy—at almost exactly the same automatic circuit-breaker part of his time. famous "Electrostatic Coil" of 1838. To test your coil, fashion sparka gap According to A.F Collins, writing in from two pieces of copper wire The Lightning Wheel. Page had 1908, the so-called electrostatic ma- connected to the high-voltage output been in touch with William Sturgeon, a terminals on top of the case. chine was the most powerful coil in well-known English experimenter and existence at the time. inventor, in 1825, of the first elec- The next important development tromagnet. Sturgeon was also respon- came, once again, from Neef and sible for a science journal, The Annals Wagner in 1839. In short, the men re- of Electricity, and, published a reprint placed the mercury/cup arrange- of Page's original article on increasing ment with a vibrating metal strip and shocks in 1837. After that, interest in the contact point and placed the entire new spark-producing machinery sub-system at one end of the coil. The grew rapidly, and improvements were great advantage of the device ob-is announced one after another. vious: eliminatedit the mercury. One special problem presented it- It did not take long for Neef's simple self immediately. The high-voltage im- A miniature gaseous tube illuminator system, or something very similar to it, pulse in the secondary winding can be made with piecea of thin sheet to become one favorite method of appeared only when the primary cir- metal, some wire gauze, and fewa making and breaking the primary circuit was opened or closed. So how to pieces of plain glass. cuit. Now, what to call it? One popular make and break the circuit in the term was "rheotome" (current-slicer). most efficient way possible became a Neef, a physician, used the system for Page liked "electrotome." Callan point of major concern. As turnedit the creation of a rapid succession of liked "repeater." Some preferred to out, there were a lot of ideas on how presumably therapeutic electric honor the inventor with "Neef's ham- to do it. shocks. He called his device the "light- mer." There were regional variations One very early suggestion came ning wheel." too. In France, according to Bec- from Neef and Wagner of Frankfurt- querel, wasit known as a "trembler" am-Main, Germany. wasIt a copper The Hammer and the Wasp. The (lrembleur). And of course, among En- disc, about 6 inches in diameter, lightning wheel, as well as other sim- glish-speaking experimenters, "inter- equipped with 36 strips of wood inlaid ilar devices, had to be operated by rupter" and "vibrator" were destined with strips of metal attached to the hand. Such manual contrivances to become standard references. disc. A flexible conductor made con- were bothersome. What was neces- Credit for the most imaginative ex- tact with the interleaved wooden and sary was some kind of automatic de- pression must certainly go to certaina metal surfaces. When the brush hit vice. English clergyman and amateur metal, the circuit was closed; when it In September, 1837, McGauley, of physicist, Reverend Lackey. Lockey hit the wood, the circuit was broken. Dublin, exhibited the first self-acting noticed that the operation of a typ- the physics of electricity were being made in America; and, perhaps more important, none were really ex- pected. Now, think of the situation on the opposite side of the Atlantic. It could not have been more different. From Galvani to Volta, to Davy, to Oersted, to Ohm, Ampere, Faraday and beyond—everything, but everything, was coming from Europe. So, it must have been only natural to believe that the trend would continue, and in many ways, did.it Charles Page (much, in fact, like Joseph Henry) was the victim of a no- torious indifference to American science by 19th century European philosophers. When Ruhmkorff was recognized and rewarded for the invention of the induction coil, is it likely A high-voltage flasher can be easily made by connecting an ordinary I5-inch that the French officials involved had fluorescent tube to the output of the induction coil, and telegrapha key between your never even heard of the experi- power supply and the input. The circuit makes uniquea code-practice device or menter from Massachusetts. As Salem signaling system. Howe Wales of Scientific American phrased it, the episode was yet an- ical Neef-type device was accom- other "oversight of American panied by a kind of buzzing, "a achievements by European savants constant bee-like hum," as he put it. already too common." The real inven- Apparently, he thought that the name tor of the induction coil was not ig- of the mechanism should recall the nored; he was unknown. insect-like sound it made when it worked and suggested the term "gal- A Spark Coil For Your Home Lab. A vanic wasp." good electromechanical induction coil stillis one of the most versatile and The Prize. There oneis final chapter in the story of the mid-19th century This simple device creates rotatinga induction coil. In 1802, following a circle of sparks between the central Further Reading long visit by Alessandro Volta,. electrode and the inside edge of the Napoleon Bonaparte announced metal ring. Note that the input cables The History of Induction, Charles Page, In- the establishment of a 60,000-franc are kept at least or3 inches4 apart. telligencer Printing, Washington, 1867. prize to be given to the creator of the "Description of a Large Induction Coil," most useful application of the voltaic effect of reducing unwanted dis- William Spottiswoode, The Philosophical Magazine, Volume 3 (January 1877), pp. battery. The Volta Prize was awarded charge at the vibrator points and in- 30-34. to Humphrey Davy in 1806. wasIt not creasing the intensity of the second- awarded again for over fifty years. ary spark. Ruhmkorff, who had been Induction Coils, H.S. Norrie, Spon &Cham- berlain, New York, 1901 Following a revival of the competition building induction apparatus at his by Napoleon Ill, the prize was given to Parisian workshop since 1851, incor- Experimental Science, George Hopkins, Munn, New York, 1902 Heinrich D. Ruhmkorff in 1864 for the porated Fizeau's modification imme- invention of the induction coil. diately. wasIt a good move at gooda Design and Construction of Induction Yes, the invention of the induction time. As Page himself said in 1867, Coils, A. Frederick Collins, Munn, New York, 1908 coil. Heinrich Ruhmkorff did not invent Fizeau's contribution to the Ruhmkorff the induction coil; there is no doubt coil "at once gave a it celebrity which, "The Induction Coil," George Shiers, about that. So why, then, was he given perhaps, mightit otherwise not have Scientific American, Volume 224 (May 1971): pp. 80-87 the prize? Why not Charles Page? The attained." reasons are very complex and, to this Physics, Patents, and Politics: BiographyA of Charles Page, Robert C. Post, Science day, not entirely clear. A Case of Indifference. There is a History, New York, 1976. In 1853, Armand Fizeau, a French wider issue implicated here, an issue physicist, discovered a way of greatly involving the sociology of scientific Note: The books by Norrie, Hopkins, and Collins are currently available in paperback improving the performance of induc- activity. Recall for a moment the editions from Lindsay Publications (RO. tion coils: the connection of capaci-a period: the middle of the 19th century, Box 12, Bradley, Illinois 60915-0012; Tel. tor parallel to the breaker points of the almost 150 years ago. At the time, al- 1-815-468-3668). Call them for price and vibrator. That simple addition had the most none of the major discoveries in shipping information.

- 12 - The coil is sealed in a rectangular power off and loosen or tighten the MATERIALS LIST FOR wooden case (about 8 x 5 x 4 control nut slightly. Make your adjust- EXPERIMENTS inches) and comes complete with an ments conservative, a quarter turn, adjustable Neef-type hammer inter- perhaps; the operational range notis NOTE: Prices listed may have changed. rupter. The unit is made to operate very large. Do not attempt to touch 6-12-volt, 0.5-3.5-amp DC power with DCa input of about to6 10 volts at the vibrator or adjust the points supply when the unit is in operation. Copper wire beiween 0.5 and 3.5 amps depend- 15-inch fluorescent tube ing on input voltage and adjustment Once you get a spark, turn off the Four x 4 4 inch glass plates of the breaker points on the hammer. coil and make the spark gap about 1/4 Induction coil The input terminals are next to the inch larger. Then turn on the power Iron filings vibrator and the high-voltage output and check for a spark. youIf have an Large binding post or Fahnestock clip posts are on top—just as they were on adjustable supply, raise the input volt- Miniature mercury-vapor bulbs the classical induction transformers of age a bit necessary,if but don't over- NE-2 neon bulbs the 19th century. The unit delivers a do it. You can also try readjusting the Salt shaker strong spark at least inch1 long and vibrator. Repeat the procedure until Sheet aluminum or brass, 4 /12 x 2/12 power enough for most of the stan- the gap toois wide for sparka to pass. inches That is one simple way to determine Spark-plug wire (or equivalent) dard high-voltage projects and dem- Steel ring onstrations. the maximum output of your coil. Telegraph key You will, of course also need a When everything justis right, the sys- Threaded rod (or long narrow screw) power supply. Anything that can sup- tem will deliver a thin, intermittent, Wire screen 4 /12 x 2/12 inches ply the required voltage and current lightning-like spark close to 2 inches 1 piece of wood, 7 /12 x 5/14 x /34 inches will work, but one with a continuously long. These long sparks are best ob- 2 pieces of wood, 5 /12 x 1/12 x 1/12 variable output between 6 and 12 served in a dark room. inches volts is recommended. Note: The high-voltage induction coil is Assuming that you have gathered Gaseous Tube Illuminator. There available from Fisher Scientific EMD some of the things listed in the Parts are many, many things you can do (4901 W. LeMoyne, Chicago, IL and Materials List, let's discuss some with your induction coil. With some 60651; Tel. 1-800-621-4769 or experiments you can perform. Keep in scrap metal and fewa pieces of glass 1-312-378-7770) as part number mind that these demonstrations, gen- you can make a miniature gaseous- S-43525 for $121.10. Shipping rates erate a certain amount of ozone; so tube illuminator. vary from place to place so call them you'll want to perform all of your ex- Obtain four plates of clear glass for an estimate. Illinois residents will periments in a well-ventilated room. about 4 inches square. Such glass need to add the appropriate sales tax. plates are standard science items The following items are available from Testing and Operation. The most and can be ordered from a supplier American Science and Surplus (601 obvious thing to do with a spark coil is mentioned in the Parts and Materials Linden Place, Evanston, IL 60202; make a spark. For that, obtain two List. You'll also need some clean wire Tel. 1-708-475-8440): 4 3/4 inches O.D. pieces of bare copper wire each screen and a piece of very thin sheet steel rings, part number 20247 at $2.50 for packagea of two; green about 3 inches long. Do not use cov- metal (like brass or aluminum). The mercury vapor glow lamps part ered wire; the insulation may get hot metal sheets you select should be thin number 3628 at $2.25 for packagea and melt. Bend each of the pieces enough to permit easy shaping with of three. The company requires a into a semi-circle and connect them scissors. $12.50 minimum order and flata to the high-voltage terminals. The Cut the metal sheet to form a rec- $4.00 fee for shipping and handling. idea is to keep the Iwo wires as far tangle about 2 /12 inches wide and 4 /12 Illinois residents will need to add the away from each other except where inches long. Then, fashion a tab at appropriate sales tax. Their catalog is they form the spark gap somewhere one narrow end of the rectangle. 50 cents. above and between the binding Now, cut a piece of wire screen to Clean 4 x inch4 glass plates are posts. To begin with, make the spark match the size and shape of the met- available from Hagenow gap about 1 inch across. Make sure al sheet. The metal sheet and wire Laboratories (1302 Washington the wires are fastened down tight. screen will be used as electrodes. St., Manitowoc, WI 54220) as part Now, look at the vibrator mecha- You're now ready to build the il- number 67C for 49 cents each. The company requires $15.00a nism. You will see a large knurled nut; luminator. Place the metal-sheet minimum order. Their catalog is that is the breaker-point adjustment electrode between two of the glass $1.50. WI residents must add sales control. Screw the nut in or out until the plates. Place the wire-screen elec- tax. points are just barely touching. Finally, trode between the other two. Rest the note polarity as indicated (by color) sheet-metal sandwich on some sort enjoyable high-voltage devices you at the input terminals of the coil and of insulating material, like 4a x inch4 can possess. An excellent example hook up your low voltage DC power block of wood. Now place fewa small designed along classical lines is cur- supply. neon bulbs on the glass surface over rently available from Fisher Scientific, If necessary, set the supply between the metal sheet. Then, add four rub- a laboratory supply company in Chi- 6 and 12-volts. Turning on the supply ber spacers, one at each corner of cago (see the Parts and Materials List should cause the coil to produce a the glass. Finish by placing the wire for more information). strong inch1 spark. it If doesn't, turn the mesh arrangement carefully over the

- 13 - bulbs; the lower glass plate should rest ety. Such tubes are often sold as re- them with a large alligator clip. Con- on the spacers. The tabs on the metal placement parts for photocopying nect the clip to the threaded-rod sheet and the wire screen should be machines. electrode beneath the lower surface pointing away from one another. of the wood. Connect the other wire Connect one high-voltage output The Spark Ring. Here's a device to the binding post. Then connect the terminal on the induction coil to the which will enable you to create an apparatus to the high-voltage output sheet-metal tab, and the remaining infinite variety of randomly rotating of the induction coil. terminal to the screen's tab. Spark- sparks. The visual effect resembles a Finally, obtain some powdered iron plug wire is probably the best kind of wagon wheel spinning under strobea and place about tablespoona of in it wire to use; but any sort of heavy, well- light. Frozen with a photographic an old salt shaker. Now, very carefully, insulated hook-up cable will work. Fi- time-exposure, resemblesit some sort sprinkle some of the metal onto the nally, darken the room and turn on the of weird electric snowflake. surface of the wood inside the perim- power. A high voltage electric field is The simplicity of the spark ring al- eter of the metal ring. Try to distribute created in the space between the lows for a number of equally suitable the iron as evenly as possible, but do plates that causes the neon tubes to construction methods; but, here's one not use too much of it. flicker and glow like so many little easy way of doing it: To begin, you'll When everything is ready, darken lightning bugs. need some wood—one piece about the room and turn on the coil. The You may observe a few stray sparks 7/12 x 5/12 x 3/4 inches and Iwo pieces entire area inside the ring will light up passing over and around the edges 5/12 x 1/12 inches square. Locate the with hundreds of tiny sparks created of the glass plates. A very careful re- exact center of the larger piece and by the gaps between the particles of arrangement of the electrodes (with drill a hole just big enough to accom- iron. If you are not satisfied with the the power off, of course) will probably modate a long screw or length of effect at first, just turn the apparatus solve the problem. If it does not, try threaded rod. The screw should ex- off, remove the iron particles with a using smaller electrodes, larger tend at least 1/2 inch beyond the up- magnet, and try again. Every ap- pieces of glass, or a lower power-sup- per surface of the wood and no more plication of iron will create a slightly ply setting. than inch1 below the lower surface. different pattern of tiny sparks. The de- For an unusual multicolor effect, try Drill another hole (somewhere near signs bear an interesting re- replacing a few of the neons with the upper right hand corner a is good semblance to the Lichtenberg figures small mercury-vapor lambs. The type I place) for a large binding post or discussed in an early article (March used were also used in the Electronic Fahnestock clip. Set the piece aside 1990). Novelty Light featured in the De- for now. Do not feel limited to the use of cember 1990 issue of this magazine. Now for the key component: a per- powdered iron. Any low resistance See the Parts and Materials List for fectly circular ring of metal about or4 material reduced to the appropriate more information. 5 inches in diameter. You may very size will also work; but the effect will be well have something like that in your different. You can try tiny bits of Fluorescent Flasher. If you enjoy collection of building materials. not,If stranded copper wire, metal foil, or working with gaseous conductors, just make a loop from some thick wire, even smalla handful of miniature nuts here's something else you can try. Ob- metal tubing, or a long metal strip by and bolts. However, do not, under any tain a good quality telegraph key (or wrapping aroundit a large round bot- circumstances, use an easily com- some other kind of momentary-con- tle. Or, you can get something ready bustible metallic substance, like tact switch) and hook it up in series made. What didI was obtain a big powdered magnesium, powdered between one output terminal of your steel ring from American Science and aluminum, or zinc dust. They are very, power supply and one input terminal Surplus (see Parts and Materials List). very dangerous. of the induction coil. Now, locate an It's the perfect size and shape. Also, it's Also, keep in mind that the appara- ordinary 15-inch fluorescent tube and heavy enough to stay in one place tus does create some heat in the cen- connect directlyit to the high-voltage without the need for additional hard- tral electrode and metal ring. So, do output with some spark-plug wire and ware. not run the equipment for extended a couple of alligator clips. To keep the Once you have your metal ring, periods of time and watch carefully whole contrivance steady and safe, you'll need to furnish withit a piece of for anything suspicious or unusual, like the fluorescent tube should be ele- flexible conductor for attachment to over-active sparks or scorched wood. vated up and away from the rest of the binding post. usedI some braided the equipment with a stand of some copper grounding cable, but a short Learning More. For more on the sort. For such temporary projects, length of stranded wire will do just as history of induction apparatus, see standard laboratory hardware comes well. possible,If solder the wire to the George Shiers' May 1971 article in in very handy. metal ring. Do not use too much wire Scientific American and Robert Post's With the power supply on, each to make the connection, otherwise biography of Charles Page. For more press of the key will activate the coil the metal ring may not lie flat on the experimental ideas, see Volume II, and flash the lamp. The circuit makes wood. Then set the whole con- Chapter 2 of George Hopkins' a unique code practice device or sig- struction on the Iwo remaining pieces Experimental Science and Chapter 5 naling system. For a still better effect, of wood. of H.S. Norrie's Induction Coils. For ad- try using a green fluorescent tube in Cut two 15- or 20-inch pieces of ditional information, consult the sec- place of the conventional white vari- spark-plug cable and furnish one of tion entitled "Further Reading." •

- 14 - ELECTRONIC DAZER

Never walk in fear with this one-evening project. It won't kill, but it's an effective way to say "Leave me alone!"

By Rick Duker

LITHE ELECTRONIC DAZER IS A MODERN, PORTABLE, PER - sonal-protection appliance. generatesIt high potential energy WARNING to ward off vicious animals or other attackers. It is an aid to THIS DEVICE IS NOT A TOY. We present it for educa- help escape from potentiallya dangerous situation. The de- tional and experimental purposes only. The circuit develops vice develops about 2,000 volts. Higher voltages may be about 2000 volts at respectablea amperage. It can cause you attained by adding additional multiplier stages, but shouldit pain and even damage if you become careless and touch its be noted that those stages will also increase the overall size of output terminals. The unit can also damage property as well the unit. so use it wisely. The Dazer is very compact, being built into smalla plastic You should never use the device on another person! It may case. It is powered singlea nine-volt battery, either NiCad or not be against the law to possess such devicea in your area, alkaline. The high voltage is applied to two electrodes which but if you use it on someone you may be deemed liable in a require only light contact to be effective. When touched with civil and/or criminal action suit. Don't just follow the golden the Dazer, the victim will receive stunning,a but non-lethal rule after constructing the project, instead just don't do unto jolt of electricity that will usually discourage any further anyone. encounters. Included in the article are numbera of instructions on how The Electronic Dazer is a power supply which consists of a to build, test, and operate the Dazer; all of them must be micro-size regenerative amplifier/oscillator coupled to an followed to the letter. Do not deviate from the procedure. energy multiplier section. It should not be confused with cheap induction-type cattle prods. The Dazer is more versatile than other high-voltage stun devices currently being charge. That voltage is rectified and increased by the voltage sold. Those devices are basically high-voltage, AC gener- multiplier section which consists of C2 and C9, and D1 to ators which jam the nervous system. However, the Dazer may D8. The final output is approximately 2000 volts. The neon be used for heating and burning applications, or anywhere a bulb NE1 is used as chargea indicator and indicates that the high voltage DC supply is required. unit is charged and operating properly.

How It Works Construction Referring to the schematic diagram in Fig. 1, the two power As with all projects start out by laying out and identifying. transistors Q1 and Q2, form regenerativea amplifier operat- If you do not wish to make printed-circuita board, then you ing as powera oscillator. When Q1 turns on, Q2 turns on and that shorts the power supply across the primary of Ti. That 01-08 ARE 1N4007 C2-C9 ARE .01 UNITS current pulse induces higha voltage in the secondary of TI. As Cl charges, Ql turns on again and the cycle repeats itself C3 C5 Cl C9 Therefore, a rapid series of DC pulses are generated and stepped up by Il to approximately 300 volts at full battery

Fig. 1—As you can see, although the Dazer is not complex, it contains enough doubler circuitry to pack quite a punch. The oscillator does nothing more than send sharp current pulses through Ti. The back EMF across the secondary winding ELECTRODES is then pumped through the multiplier stage to produce the very-high output voltage across the electrodes.

- 15 - may use perfa board as long as you remember to keep the leads of all high-voltage components isolated. That is to prevent sparks from arcing across your board. A x 4 7.5 cm perfboard is suitable for that purpose. The first components you should mount are the two transistors Q1, Q2, transformer Ti, resistor RI, and neon bulb NE1. Solder them in place (for PC construction) being sure

that the transformer and transistors are hooked up correctly. at Apply smalla amount of adhesive to the base of NE1 to hold it securely in place. Mount DI to D8 and C2 to C9 on the board and make all solder connections. Note proper polarity of the diodes. The off-board components come next. Solder in leads for Si, and the output electrodes. Also solder in the battery clip for Bi. Build the enclosure from some nonconductive material such as plastic. Drill holes for Si, NE!, and output electrodes. Be sure that the output electrodes are about cma or greater apart. Connect the output wires to the electrodes and insert them through holes from inside of case. Thread on the retaining nuts and tighten them securely. Set the circuit board in the case and mount Si, securing it with a nut. That completes the construction.

PARTS LIST FOR THE ELECTRONIC DAZER C1-0.1-µ,F monocapacitor C2—C9--0.01-FLF 400 volt polyester capacitors D1—D8-1N4007 1-kVolt diode NE1—Type NE-2 neon bulb Q1—MJE521 NPN power transistor Q2—MJE371 PNP power transistor R1-3,300-ohm 1/4-watt resistor R2-1,000,000-ohm /14 -watt resistor Good parts layout is the secret to any miniature project. If 1 S1—SPST monetary-contact, pushbutton switch T1- your layout causes the battery to come too close to the high- 1200 to 8-ohm audio power transformer voltage components we suggest you insulate withit tape. ADDITIONAL PARTS AND MATERIALS 9-volt battery clip, 10 x 5 x 2.5-cm plastic case, 7.5 x Operation and Use 4-cm perfboard or PC board, two t32 X - 1 /14 bolts and Activate the unit by pressing SI. NEI will light indicating nuts for electrodes, adhesive for mounting NE1, circuit the dazer is fully charged and ready to use. Notice also that board standoffs (optional), hookup wire, solder, etc. only one pole of NEI will glow indicating DC voltage pres- The following are available from Quantum Research, ent. isIt important to remember that the device holds chargea 17919-77th Avenue, Edmonton, Alberta, Canada T5T even after S1 is off. To discharge, touch the electrodes to a 2S1: metal object and note the healthy spark discharge. QV100K1—Complete kit without PCB (includes all The Electronic Dazer was designed as self-defensea weap- above parts except those following the electrodes in the above list)—write for price. on for use against vicious dogs or other attacking animals. QV100K2—Complete kit with PCB (includes all above The device is most effective when the electrodes contact an parts except those following the electrodes in the above area of low resistance such as skin or flesh. Those include the list)—write for price. snout or mouth since the resistance of those areas are much lower than areas of hair or fur. The electrodes could be pointed to penetrate these areas better. The dazer generates Testing great stopping power. One contact will give powerfula jolt Before inserting the battery and closing the case, fewa test and should discourage any further attacks. measurements should be made to ensure correct operation. The device can burn and heat materials with low resis- With the ground clip connected to the battery, connect VOMa tance. Those include flesh, moistened paper or wood, etc. between the positive clip and the positive terminal of the That makes the unit potentially hazardous to humans. Re- battery. Set the meter for current reading, and press Si. You member, the dazer is not toya but qualitya electrical ap- should measure currenta of approximately 300 to 500 mA. pliance and therefore must be treated accordingly. Use the NE1 should be glowing. utmost discretion with this device. With higha voltage VOM, you should measure about 2000 Another use for this device is as high-voltagea DC power volts on the output terminals. Those measurements indicate supply. It may be constructed as variablea power supply if proper circuit operation. Let the unit run for about one output taps are taken from various stages of the voltage minute. Transistors Ql and Q2 should be warm, but not hot to multiplier section. Remember, always disconnect the battery the touch. Insert the battery in the holder and close the case. and fully discharge the capacitors before working with the That wraps up the Electronic Dazer. circuitry. •

- 16 - Build capacitors that really pack puncha for high voltage fun!

BY ANTHONY CHARLTON MAKE YOUR OWN HIG H VOLTAGE iiPACITORS

s one who supplies parts to pacitors of any voltage and energy dielectric. A dielectric may be solid, those who experiment with storage size for either AC or DC use. gel, liquid, or gas. A capacitor's ability high voltage, getI a lot of let- A The process involves a step-by-step to store energy is measured in either ters and phone calls from frustrated logical approach that we'll present microfarads nanofarads (nF), or builders that go like: "Can you supply here. We'll explain how to plan and picofarads (pF). Micro means one mil- an inexpensive XXX microfarad ca- construct a capacitor, where to get lionth, nano stands for one billionth, pacitor at a working voltage of YYY? materials, safety considerations, tips and pico for one trillionth (farads are My only source wants $249 for one." and hints, and include a few simple also used, but in high voltage work Sometimes, a high price is justified; projects. they are impractically large units). other times, a seller has the only ca- Several factors affect capacitance. pacitors of a special value available, A Capacitor's Description. A ca- The formula for determining capaci- and will soak you for the maximum pacitor consists of Iwo or more plates tance is: dollar. of a conductive material separated It is feasible to build your own ca- by an insulating substance called a C = (0.224I

- 17 - where C is the capacitance in dielectric. alsoIt has the advantage picofarads, K is a constant that de- of being easy to glue to with Silicone WARNING!! This article deals with and pends on the insulator (or dielectric) involves subject matter and the use of RTV or Krazy Glue, and it is readily between the plates (called the di- materials and substances that may be available and cheap. However, it is electric constant), theA is area of one hazardous to health and life. Do not at- fragile, and may contain impurities conductive plate in square inches, d is tempt to implement or use the informa- that allow conductive paths for de- the separation between adjacent tion contained herein unless you are structive arcs. Contradictorily, for your plates in inches, and n theis number experienced and skilled with respect to first capacitor or two, we suggest that of plates. As you may know, different such subject matter, materials, and sub- you try typea made with glass to gain stances. Neither the publisher nor the insulators have different dielectric experience, since they go together author make any representations as for constants. Table 1 shows the values of easily and are cheap. the completeness or the accuracy of the K for some common materials and Many industrial capacitors are oil information contained herein and dis- the peak voltage they can withstand claim any liability for damages or inju- filled. Oil has an extremely high resis- per /34 000th inch (called a mil) of thick- ries, whether caused by or arising from tance, so doesit not measurably in- ness. This rating calledis the puncture the lack of completeness, inaccuracies of crease leakage. Silicone transformer or breakdown voltage. the information, misinterpretations of the oil is the best liquid insulator, but is directions, misapplication of the informa- rather hard to obtain. Mineral oil, on Dielectrics. The better the insulat- tion or otherwise. the other hand, is readily available ing property of the dielectric, the from most pharmacies. Although it higher its resistance, and the less di- has a low dielectric constant, it can electric leakage loss present. In low with high quality steel drill bits or spe- be used in a variety of simple ways to current, high voltage power supplies, cial plastic bits. They must be drilled at make very good high voltage capac- minimizing all sources of loss impor-is 300 RPM or slower to prevent chip- itors. tant to prevent undue power-supply ping and melting, and be sure to For example, a dandy variable DC loading. For that reason, plastics are leave the protective film or paper on capacitor can be made by immers- by far the best materials for large ca- the plastic when working with it. ing junkeda AM-radio tuning capaci- pacitors. A serious project should in- Mylar, Polyethylene, Nylon, and es- tor of the movable-plate type in volve one of the plastics. pecially Teflon are difficult to work with mineral oil so its shaft and connection Lexan, Polystyrene, and Plexiglas in as they are very slippery. The best way leads come out of the container's top. particular are easy to glue, and can to attach plates to any of those mate- If you wish to try this idea, make abso- be cut with a table saw using a plas- rials is to use a glue specifically de- lutely certain the "cold" plates of the tics blade, or a carborundum impreg- signed for the material. Polyvinyl capacitor (the moving plates) are at nated all-purpose cutting blade like chloride (or just PVC) is moderately ground potential. Use a good, large, Zippity-Do (which cheaper).is A sabre slippery. canIt be glued with a PVC non-metal knob for adjustment. A 100- saw with a really coarse wood blade cement, or foil plates can be at- to 365-pF variable capacitor with a 1- will also work (other blade types clog tached using silicone RIV. kVDC breakdown voltage (i.e.. a plate or chip). Such plastics may be drilled Glass is, in principle, an even better spacing of 1 mm) becomes a 270- to 985-pF unit with 7500-VDC breakdown rating. Try pricing a 7500-volt TABLE 1—DIELECTRIC CONSTANTS AND BREAKDOWN VOLTAGES variable capacitor sometime, and you'll see the advantage to this ap- Dielectric Puncture Voltage Insulator Constant per 0.001 Inch Notes proach! You can use mineral oil in designs of Air 1.0 30 1 Window glass 7.8 200 your own, too. Immersion of a home- Polyethylene 2.3 450 made capacitor in mineral oil will Paper (bond) 3.0 200 greatly improve its voltage rating and Polycarbonate (Lexan) 2.96 400 lifetime. Teflon 2.1 1000 Polystyrene 2.6 500 Paper is an excellent dielectric Epoxy circuit board 5.2 700 2, 3 when saturated with mineral oil. Try Pyrex 4.8 335 20-lb. bond computer paper which Plexiglas 2.8 450 has 4a mil thickness. Prepare this inex- PVC (rigid type) 2.95 725 pensive capacitor by interleaving Silicone RN 3.6 550 Polyethylene terphthalate (Mylar) - 3.0 7500 layers of dry paper with aluminum foil, Nylon 3.2 407 4 and then immerse the capacitor in oil Mineral Oil, Squibb 2.7 200 2, 5 until the paper gets saturated. Shellac 3.3 200 One disadvantage to using oil in NOTES: All measurements at 1 MHz unless otherwise noted. home-made capacitors is that the 1Tested with dry air. tape or glue used to bond the assem- 2Tested at 300 HZ using Healthkita IM-2320 Multimeter and homemade capacitor. 3Estimate, based no experiences. bly must be oil-resistant. Silicone RIV is 4Lowest value of 3 types. the best glue for these purposes. 5Estimate. Probably higher. 0.040"A gap withstood over 10,000 volts DC before breakdown in one test. Design Considerations. There are

- 18 - DO NOT TOUCH BALL WHEN JAR IS CHARGED! High Voltage Safety. METAL BALL High voltage is considered any value over 500 volts AC or DC. When you attach a capacitor to high voltage, you are multiply- lbr METAL ROD ing its hazard manyfold. Therefore, experi- PLEXIGLAS PLUG MAKE- menters must take extra precautions to FROM 1/4" SHEET avoid painful shocks and possible elec- trocution. Here are fewa guidelines to follow when working with high voltage: • Label your project in several locations GALLON MAYONNAISE JAR with: "Danger, High Voltage" where appro- ALUMINUM FOIL ON priate. Such a warning label is provided OUTSIDE IS THE NEGATIVE TERMINAL here for you to copy (see Fig. A). Keep children, pets, and curiosity seekers away from the apparatus. Cover all bare leads, wires, connection terminals, and possible points of contact with high voltage putty or a cover fabricated from thick clear plastic. • Work in a dry location. Working in a FOIL ON INSIDE damp basement or workshop courts disas- IS THE POSITIVE ter. Wear rubber-soled boots or sneakers. PLATE Stand on thicka rubber mat.

Fig. A. Curiosity can hurt more than just felines, so use this warning label on WIRE CONTACTS BOTTOM all your high voltage projects to protect the unwary from harm. Fig. I. The classic Leyden jar is the oldest storage capacitor we know of They are easy to make and the materials only cost around $2. • Never put your body in a position to become conductor.a Locate your apparatus away from appliances, metal doors and several things to consider when de- designs first to get used to techniques windows frames, heating ducts, vents, ra- signing and constructing your own and quirks before you invest lots of diators, sinks, or water pipes. All these items can become deadlya ground yourif capacitor. Let's point out each one time and money. body comes between them and high volt- before moving to the construction de- You must also take into considera- age. tails. The first and most important thing tion the voltage that will be applied to • Always pull the plug when working on to concern yourself with safety.is De- the capacitor. That will affect your a high voltage circuit unless you must test it. When testing a live circuit, use utmost spite the romance of high voltage, is it choice of a dielectric and thus its re- caution. Keep one hand in your pocket. foolish to needlessly risk your life. Since quired thickness. Should you use an Use clip-on test leads that are rated twice you will probably be working with le- inadequate dielectric or thickness, the voltage of the live circuit. Use a high thal voltages, observance of all safety sparks or arcs can result. A spark is a voltage probe whenever possible—its in- practices for high voltage (or HV) is temporary breakdown that a lot of sulating handle will help protect you. • Use NE-2 neon lamps to indicate live absolutely essential. For some capacitors will survive, but an arc is or stored high voltage. Bleed off the charge guidelines, see the boxed text entitled serious: it is a path burned into the on capacitors with a power resistor before "High Voltage Safety." dielectric or other component. Arcs performing adjustments. The next aspect to consider is ca- carbonize materials, producing a • Adequate ventilation must be provided for circuits that produce large pacity. youIf have a specific capaci- highly conductive channel that often amounts of ozone such as Jacob's ladders tance in mind, you can design a renders an apparatus useless and or Tesla coils. capacitor using the information pro- very likely dangerous. Except in spe- vided elsewhere in this article. Try one cial cases where the insulator is a of the designs described later. Or per- "self-healing" type (like air, oil, and that often appear in materials that haps you prefer experimenting in- some plastics), singlea arc will ruin the are not highly refined for capacitor stead. Either way, when building for capacitor. use, we must add a safety margin to the first time, we suggest making small To compensate for the impurities the thickness of the dielectric. In the

- 19 - case of DC, a good rule of thumb ais COPPER-CLAD PC BOARD.

50% margin. For example, say you 2" x 6" ALUMINUM-FOIL TAB COPPER MUST BE BRIGHT need a 500-volt DC capacitor using SECURED WITH CRAZY GLUE. SHINY CLEAN AND FREE OF CHEMICAL RESIDUE! polystyrene. Consulting Table 1, note polystyrene's breakdown is 500 volts -e_ ee, 1" EDGE MARGIN eK,.-.".'>r,e,, per mil, thus mil1 is required. Adding .e,,- ,- n e,-;.: 50% gives you 1.5 mils, which is ade- 4:e" 't'',:5,- quate for pure DC. You can always use :,)-- ee' '-iV„•,1"- a thicker dielectric if it's expedient, 'II 1, ,\\9 , \ie- e;› e CA,m\ ,, . providing that you adjust the number \'‘NqAgwrzp,tm,m of plates or their size to accommodate the wider plate separation. It should be mentioned that when making a paper capacitor, you should use 1¡‘\i ,‘,,:,!, ,'\,r, \ \.\\\'\,,,P,,eltekt,\‘'0, \Iii $-\\\,tx\\1).\\,•\„'¡‘,;., \1;,P\\;\i'.\«\t\ ''\'"Îii4ki,'''''.\\.\\l'\‘‘:\' \ S a healthy safety margin since paper is e\k,\,‘t\l\,k,lk,\‘,'\,t,,i.?,1\,!M1..1.4,\¡,e,oil;;\\\.k,»\\0\tk\,,gp\,:`•`\ i\00,\'‘'kî'ie,','\'4,,,'!\\,\,,:'1,‘:.:i,!: not always uniform in thickness. \\,,,:\\\\\\\\‘,\.\¡‘\; \t\v1;,,. ,,:.,,\;,, \,i,\‘‘,,.:\•;\;\.;;,,,\,,,,e,,,.,,.,;...;,....;\,,,,:,,,,i,.,,\;,;\.\\,,:\\,‘\,,,:\i:\:,;, In comparison to AC, DC puts rela- 'r, \"my,e'e \\\,,,,,, qy,\ ,,.e.og\ei \\\'. OM\4\`1\':., ' a • .‘' ',',\ ‘'' '' tively little stress on a capacitor. By 12" 10" contrast, AC reverses the dielectrics' Ï ‘''ikly''P''\' \\\..'.,'\\t,.,.‘:e.b\\ ?„,, , polarity every cycle. So the dielectric in an AC capacitor must have twice .e4\1,,,\e\\t,teel¡il'jI'\I!"\;\t'.\'çe%\iee'tl.,Z•\' 1\t‘el'd''''t \., the thickness required in on equiv- \ alent DC capacitor. Further, when considering dielectrics in AC applica- ' tions, you must deal with the peak voltage—not rms (Root Mean Square) voltage—that they will be exposed to. If you wish to convert arr rms voltage to its equivalent peak sinewave value, 12" multiply byit 1.414. So, to roughly calculate the proper Fig. 2. For single-sectiona capacitor, use one double-sided PC board. For multiple voltage rating needed for an AC ca- sections, use several single-sided boards clamped together or bolted together with pacitor, you first double its required nylon screws. rms voltage rating then multiply by 1.414. To further simplify this calcula- dielectric material to use since hasit a breaks down gradually upon ex- tion, all one needs to do multiplyis the very high puncture voltage per mil, posure to ozone gas (always present AC (rms) voltage in question by 2.828. and thus makes a very compact ca- around high voltage) becoming brit- Now divide the voltage by the punc- pacitor. Plastics are light, so most ca- tle and less resistant to arc puncture. ture-voltage rating to get a prelimin- pacitors will weigh less than ten That brings us to another important ary thickness value. Finally, you must pounds. The toughest plastic is Lexan, consideration: the capacitor's useful add a safety margin of 50% to 100%. which is difficult to crack even with a life. To enhance a capacitor's life The actual percentage depends on hammer and oftenis used for vandal- keep the working voltage at or below the characteristics of the applied AC proof windows. Glass theis worst ma- the rated specification in both DC voltage. For a pure sinewave AC, we terial for a lightweight, durable ca- and AC applications. We discovered suggest 50%a safety margin whereas pacitor, and can even crack under its that charging at no more than 70% of high frequency, non-sinusoidal ap- own weight when lifted. Take all this a capacitor's working voltage resulted plications such as Tesla coils require a into account when selecting your main an amazing 10-fold increase in life- full 100% extra thickness. terials. time for one type of commercial ca- If one is available, equip an os- Of course, the overall cost in labor pacitor. Also, for DC capacitors, watch cilloscope with a high voltage probe and materials should also be consid- out for voltage reversals. yourIf system to visually observe exactly what the ered before constructing a capacitor. has a lot of inductance, reverse volt- circuit is doing so you can determine Calculate beforehand the cost of age swings are always produced. In- the proper safety margin. An os- your materials. Paper and poly- crease the safety margin if a lot of cilloscope will also enable you to de- ethylene are the cheapest. Glass is inductance is in the circuit. Further- tect destructive voltage spikes and the next higher price. Labor time is more, the temperature should be superimposed AC (also called AC rip- about the same with Plexiglas, Lexan, kept below 120°E As mentioned ear- ple) so you can design a capacitor to and glass sheet capacitors. Exotic lier, watch out for superimposed AC, handle those harmful excursions. plastics such as Teflon are not needed voltage spikes, and ringing. These Of course physical size, weight, and unless your application demands ex- types of AC waves can drastically fragility are also important charac- treme chemical and thermal deterio- shorten lifetime. Tesla coils have noto- teristics of capacitor design. If you ration resistance. Polyethelene has rious ringing. To repeat: feasible,if use have size limitations, Mylar theis best excellent chemical resistance, but an oscilloscope to visually analyze

- 20 - your circuit. Often a power resistor in- tor plate, or anywhere the shape of a Whether a capacitor enclosedis or serted in the current path to the ca- conductor changes abruptly (such as exposed, discharge paths must be pacitor quenches ringing. With this the tip of a nail) is called point dis- wide enough to avoid arcs to the criteria under our belts, let's look at charge. canIt be readily observed in case, adjacent plates, terminals, con- some problems your design and con- a dark room at very high voltages. nections, or components. That is es- struction methods should prevent. Small, bright blue pinpoint(s) are seen pecially important in situations where leaking electrons into the air, accom- conductors must be left uninsulated. Signs of Trouble. Your assembly panied by a hissing sound and copi- Note that the space from each plate techniques should seek to minimize ous ozone production. to the edge of the dielectric must be the likelihood of a few possible prob- Once again, insulation and proper wide enough to stop any spark from lems. Luckily, all of them can be pre- ventilation are the proper solutions to "crawling" over the edge of one plate vented at least in part by using ample all these problems, and there are to another. amounts of insulating material such as some specialized techniques to insu- Power leads must be capable of No-arc or Corona Dope and/or high late your capacitors and otherwise withstanding the full voltage of the voltage putty on all exposed areas. A improve the safety of your high volt- charge plus at least a 50% safety plastic case to enclose the apparatus age projects. Let's get to those now. margin. TV anode wire, which comes is also recommended (more on that rated up to 40-kVDC, makes great later). Construction Requirements. A key leads. Vinyl tubing or aquarium air Still and all, you should know what ingredient in a good assembly is a hose may be slipped Over leads to problems the insulation is preventing. proper case. Your capacitor's housing increase their voltage rating. The first problem insulation relieves is must protect againstit moisture, dirt, Make sure the plates are securely the possibility of electrical shock. and accidental discharge. Plastic mounted or they will tend to shift, or Insulation also minimizes the pro- cases for dry capacitors are easy to make a noisy rattle when used with duction of ozone —a gas created make with acrylic sheets glued at all AC. Glue or compress the assembly to when high voltage causes three oxy- corners with Silicone PTV Oil-proof hold secure.it With regard to mount- gen atoms to join together. Ozone has cases can be made for immersed ing, keep in mind that glues that dry a tart, sweet "electrical" smell, and is models, but you will need to rough-up by evaporation of volatilea chemical 100 times as poisonous as carbon the plastic at the sealing edges with might not set properly if "buried" in- monoxide. Beware: it quickly causes sandpaper and use both a bonding side an assembly away from air, and headache, nausea, vomiting, and re- and second fillet glue coating for a could thus become a fire hazard. spiritory irritation. In addition to insulat- liquid-proof seal. Metal cases can be Rolled-up capacitors may be held ing all the exposed HV areas, you made from PC boards cut on a shear securely by wrapping the interleaving should also operate your equipment or large paper cutter and soldered at layers of foil and insulator tight around with good ventilation if it produces the edges. Copper roof flashing an insulating mandrel and then tap- any ozone. (available at hardware stores) works ing with cleara PVC tape. Where nec- Closely linked to ozone generation well too. However when using metal, essary, coat the ends with Silicone RN is corona leakage. is It produced by a always beware of contamination by That will eliminate end-arcing flash- charge being leeched off a highly solder rosin, solder bits, and other over and corona loss. Alternatively, al- charged object by the air. That typ- crud, which can short out plates or though is it somewhat brittle, paraffin ically produces ozone. However, otherwise reduce efficiency. (with a puncture voltage of 250 volts/ sometimes a device (such as a Van deGraff generator) is constructed TOP INSULATING SHEET specifically to display corona discharge, and insulating wouldit defeat 1 1 that purpose. In such cases, good THICKNESS EXAGGERATED TO SHOW AREA OF COPPER ventilation theis only practical means of hazard prevention. Ozone can also be created by arcing, which can occur anywhere. How- ever, ozone production is not the greatest hazard arcing presents. At 50 kV a spark can arc between an uninsulated contact and your body youif come within 2 inches of the contact. Arcing commonly takes two forms: directly through a capacitor's dielectric BOTTOM INSULATING SHEET (as mentioned earlier), or across the EDGE MARGIN (FREE OF COPPER) edges of a capacitor's plates to an adjacent plate. A snapping sound indicates the presence of arcing, so ALUMINUM FOIL CONNECTION TABS keep your ears open. Fig. 3. With this design, you may stack as many plates as you wish, provided there are Arcing from the edges of a capaci- an equal number of plates attached to each lead.

- 21 - BOTTOM PLATE with an insulating dielectric between FOIL TAB FINISHED CAPACITOR them. By now we hope you have a good understanding of the principles and

MANDREL techniques involved in making your own capacitors. Without forgetting safety, let's talk about how to build some simple capacitors, any of which TOP PLATE FOIL TAB can be modified for your application.

A Leyden Jar Capacitor. Leyden INSULATING DIELECTRIC Jars are one of the first types of capacitors made, having been inven- BOTTOM PLATE FOIL TAB ted nearly Iwo and a half centuries NOTE EDGE SPACING ago. Their development was first re- corded in 1745 by Ewald von Kliest. In 1746, Peter van Musschenbroeck of Leyden, Holland experimented further with the invention. We can build our own modernized units with a gallon-size wide-mouthed mayonnaise jar. The project only costs about $2, and goodis to at least 10 kVDC at 2.5 nE Units we've tested at 15 kVDC did not fail; at that voltage, the capacitors

stored just under 1/3 joule each. First select a jar without bubbles, cracks, or blemishes and that has a mouth large enough to comfortably TOP PLATE FOIL TAB slip your hand through. Next, carefully ALUMINUM FOIL clean it out. You'll use aluminum foil

ROLL THIS WAY SEE TEXT inside and out as the conductive plates (see Fig. 1). Cut a foil disk 1-inch bigger than the bottom of the jar. t Now coat the dull side of the foil and 1 inside jar bottom with a thin, even

ALUMINUM FOIL layer of rubber cement. Let both dry INSULATING DIELECTRIC for 10 minutes, and press together. Smooth with firm hand pressure. Avoid Fig. 4. A rolled-up capacitor, like that shown here, can provide the greatest excess wrinkles. Do the rest of the in- capacitance in the smallest space. Note that the dimensions in the side view have been side except the top inch of the bottle greatly exaggerated for the sake of clarity. using three or four pieces of foil. (It is easiest to do the plate in pieces in- mil) is an excellent insulator for the balls make good terminals. Clean stead of all at once, since rubber ce- ends of rolled-up capacitors and the them up with a wire brush or steel ment "grabs" and it is difficult to re- edges of flat-plate type capacitors. If wool to eliminate rough spots. The au- position the foil once contact has you want to use melted paraffin wax, thor uses fishing floats covered with been made.) Now do the outside foil heat the wax only in a double-boiler either aluminum foil or nickel print plate in pieces, leaving the top inch pan, since it if gets too hot canit catch paint for up to 10 kVDC. Split the bob- bare. Check the foils with a continuity fire. Be sure to apply several coats, bin first with a razor blade, remove the tester to determine theif pieces are in allowing the wax to harden between line holder and spring, and glue to-it good electrical contact. Areas of foil each coat. Liquid electrical tape also gether again with epoxy. not in contact can be bridged with makes a great end seal, however is it Furthermore, as you work, keep all strips of foil or nickel-print paint. somewhat hard to find. Try mail-order materials as clean as possible. Not For the top cover, cut Iwo disks of distributors for that product. only will your work have a better ap- clear plastic, one slightly smaller than High voltage terminals for your proj- pearance, but arcs and burn- the rim, the other /14 -inch larger than ects can be made from plastic rods throughs due to contaminants will be the rim. Glue the two pieces together drilled through to accept connection prevented. High voltage easily tracks to form a plug. Drill a /14 -inch hole wires. You may add a nut and bolt on along dust, surface contamination, through the plug's center. Cut and in- top for convenience, However, and even finger oil (which contains sert a length of /14 -inch (outer diame- beyond about 3,000 VDC this method salt). Also, we shall refer to a "section" ter) metal rod or tubing through this suffers from point discharge. Metal as consisting of two conductive plates hole. Attach a ball to its top, and sol-

- 22 - der a wire or small-link chain to its totypes first. The design uses a layered bottom. The wire must make good Sources approach (as shown), and we sug- electrical contact with the foil. Let the gest using only one section as is it diffi- assembly dry for a day with the cover All types of plastics: United States Plas- tics Corporation, 1390 Neubrecht Lane, cult to align and wrap multiple off, to allow vapors from the rubber Lima, OH, 45801; Tel. 800-537-9724. sections. By contrast, a single section cement to dissipate, then cement the Company changes for catalog and re- several feet long is not too unwieldy. cover on with silicone or Krazy Glue. quires minimum-amounta order. Write or Aluminum foil works great in these call for details. capacitors. You'll find the oven/broiler PC-Board Capacitor. Some nifty low High voltage rectifiers and meters: MCM type, which heavy-dutyis foil, far easi- Electronics, 858 E. Congress Park Dr., inductance capacitors can be made er to work with than the plain variety. from pieces of copper-clad epoxy cir- Centerville, OH 45459-4072; Tel. 513-434-0031. Free catalog. Polyethylene and Mylar are the most cuit board (see Fig. 2). For a simple common dielectrics, but you can ex- Iwo-plate capacitor, you can use one Sheet and formed metals, plastics, and precision tools: Small Parts, Inc., PO Box periment with other materials. double-sided sheet. For multiple sec- 381966, Miami, FL 33238-9980; Tel. Looking at the figure, note the ori- tions, use single-sided board. 305-751-0856. Free catalog. entation and shape of the foil plates To prepare each board, start by Information, technical assistance, high (A) and (C). They can be easily se- etching away a 1-inch strip from voltage parts, and kits: Allegro Electronic cured to the dielectric (B) using dou- around all its edges. That process can Systems, 3 Mine Mountain Road, Corn- ble-sided Scotch tape. Note also the wall Bridge, CT 06754; Tel. 203-672-0123 be simplified by first masking off the edge spacing. An outer covering of strip, spraying the bare copper with (9 AM-12 noon EST weekdays). Free catalog. dielectric (D) will prevent the finished an etch-resistant paint, removing the capacitor from having a "hot" case, masking tape, and then etching. which might be a hazard. With those Clean the board after etching, and When building a large capacitor of points in mind, lay the foil out on a rinse with de-ionized or distilled water. this type, we suggest that you use smooth sheet of paper, which in turn Thoroughly air-dry the sections, or use nylon bolts at the corners to hold allit should be laid out on a smooth, hard a blow dryer. Attach strips of alumi- together. The bolt holes should be surface to prevent wrinkling. Carefully num foil to each plate. pre-drilled before assembly, and all assemble the four layers as shown in If you are building a multiple-sec- chips cleared away. Make sure the the drawing. Strive to make them flat tion capacitor, connectthe aluminum plate-to-edge spacing is adequate and smooth. foil strips together as shown in Fig. 3 for the voltage you will subject the Wrap the capacitor "sandwich" and secure them using glue or nylon capacitor to. Add extra spacing youif around a non-conductive mandrel or bolts at each corner. Spray the intend to use bolts at the edges. spool—ideally made of plastic or finished assembly with several coats Glue foil carefully to the top of the glass rod (be careful not to break a of an insulating product, or paraffin. first plate using a small amount of glass rod). Try to make the roll straight If you use the dimensions shown in spray adhesive, Krazy Glue or RN sil- and free of lumps and wrinkles. When Fig. and2 a 0.060-inch gap between icone. Press smoothit and let dry.it A it's all rolled up, secure withit plenty of plates, you can achieve a capaci- photographic finishing roller is handy tape. The author uses clear package- tance of 1.94 nF (1940 pF) per section. for flattening foil. Repeat the pro- sealing tape for this. Now secure the When deciding on the gap width to cedure for the second sheet, orient- positive foil tab (assuming it's going to use, keep in mind that the greater the ing the foil connection tab in the be for DC) to the mandrel using tape. space between successive plates the opposite direction. Keep the plates Finally, coat the exposed ends with an lower the chance of arcing. For exam- and dielectrics aligned as assembly insulating product like silicone RTV. ple, a 1-inch spacing gives you a 30% proceeds. Repeat this procedure for The remaining foil connection tab larger gap than a 20-kV spark can as many sections as you want. Always may be reinforced by rolling aroundit jump. Insulation will further improve keep the final number of plus and a small metal dowel. A nail, or a cut- that margin. minus plates equal. off piece of 1/8-inch uncoated brazing Put an insulating sheet above and rod is suggested. Apply glue to hold The Stacked Sheet Design. This below the last plate and secure the the assembly together. type is virtually identical to our PC assembly with nylon bolts. Do not over Foil tabs can be strengthened by board capacitor, but it can be de- tighten or the center of the assembly adding "ribs" of adhesive from a hot signed to handle considerably more will "bow." Finally, clean the ends with glue gun. Similarly, the tabs can be voltage. You simply substitute sheet a very small amount of isopropyl (rub- made tear-resistant by applying hot plastic or glass dielectrics, and glue bing) alcohol and wipe dry. Smear a glue where they enter the capacitor. aluminum foil in place of the copper coating of silicone RTV over all the Note most problems with this design for each section (refer to the PC edges. come from particle contaminants board capacitor drawing in Fig. 3 as that stretch a dielectric thin in spots needed). All in all, it's an easier design Roll-Up Design. The kind of capaci- where they are trapped by the tightly to build, as it does not involve the tor depicted in Fig. 4 can provide rolled dielectric. Another trouble is in- effort of etching copper, and you can large capacitance in a small size. adequate edge spacing, causing continue to add sections to your origi- They are a little trickier to make than arcing across the ends. Careful plan- nal prototype to increase its capacity stacked-section type capacitors, so ning and assembly will eliminate both as future demands require. you might want to try fewa small pro- headaches. - 23 - If you've ever wanted a high-voltage generator to create neat lightning effects, perform Kirlian photography experiments or play with neon lights, then this one's for you!

A HtGe-avOLTAGE PULSE GENERATOR

By Dale Hilemon

e will describe laboratorya pulse generator using other hand, if a proper return circuit is not provided, an an auto-ignition coil and capable of delivering a equally distressing shock could be had by contact with the W train of pulses having peaka potential up to 30,000 primary circuit. Because the ignition coil is an autotrans- volts. With couplea of minor circuit and construction vari- former, the return circuit for the high-voltage pulse includes ations, the project is suitable for use as an electric-fence the power leads. Therefore, one side of the power supply charger, operating at lowera voltage, but capable of much should, if possible, be Earth grounded. That precaution, higher output current. besides preventing shock by contact with the power leads, Applications for high-voltagea spike are numerous: elec- also precludes arcing within the power supply itself as the tromagnetic and radio-frequency interference studies, elec- high-voltage pulse seeks the shortest return path. trostatic-discharge simulation; investigation of insulation break- Applying that reasoning to the fence-charger option, we down; flammability experiments; strobe effects; etc. A DC can see why fixeda pulse rate is specified, as there is a strong power supply or battery is required, and pulse potential may likelihood of accidental human contact with the fence wire; be varied simply by changing the supply voltage. With a a rate of 60 pulses per minute or less being considered safe. 12.6-volt input, the ignition-coil model delivers its maxi- Also, since there is gooda chance of personal contact with mum pulse, but uniquea multivibrator-driver circuit makes the power or battery leads, a good ground connection is operation possible down to a supply voltage as low as 1.5 essential, as with any electric-fence system. volts, yielding an output pulse of only fewa hundred volts. For maximum safety, we recommend batterya supply for Its pulse frequency is set by a front-panel control, with a the fence-charger system. range from about 0.3 Hz to 20 Hz. If you should happen to reverse the power-supply leads An ignition coil, however, is not well adapted to the fence- to either project, the current-limitation lamp, largea auto- charger application since its output resistance is so high: motive bulb easily seen in the photos, lights brightly to warn typically 10,000 ohms. Thus its output pulse is strongly you. However, the equipment must not be allowed to remain dependent on loading. With shorta fence, long sparks might in this condition for more than fewa seconds. Even if you be struck at risk of igniting brush; while on the other hand, never expect to make this mistake, the lamp should be in- with longa fence, shunting by weeds or by dirt and moisture cluded because it limits excessive surge currents that could may reduce its output voltage below an effective value. Hence otherwise occur under some operating conditions and which for the fence-charger version the RATE prf control must be could blow the power transistor. omitted for reasons of safety. No-load output of the fence-charger option is typically 4 About the Circuit Kv pk (kilovolts peak), or about half that value when con- As shown in Fig. 1, free-running variable multivibrator nected to a 1-mile fence. A car battery powers the fence- Q1 and Q2 drive Darlington power amplifier Q3, which charger model for about four months before recharging is makes and breaks the primary current to coil 1 T as in an needed (at recommended pulsing rate of 20 pulses/min.) auto ignition system. Duty, or "dwell" is a few miliseconds, Two lamps mounted on the circuit board and visible through and the high-voltage pulse is generated at the end of the the see-through front panel are important indicators of the period when the circuit is broken and the field of Ti rapidly unit's performance. collapses through the winding. An unconventional multivibrator circuit was developed Precautions to provide high saturation currents over a wide range of While a single jolt from an ignition coil is itself rarely supply voltages. In this design both transistors Q1 and Q2 traumatic, the resulting-reflex muscle contraction could have conduct at the same time and both cut off at the same time. unfortunate consequences. If a continuous train of pulses Another unique feature: For safety in the fence-charger ap- causes you to involuntarily grasp the high-voltage condiic- plication, the circuit is designed to automatically shut down tor, for instance, you might not be able to let go. On the if driver Q2 should fail to conduct for any reason (fluctuaiion

- 24 - zero volts. When QI begins conducting, and its collector voltage has dropped far enough to start Q2 conducting also, then positive-feedbacka action is initiated, forcing both tran- of supply voltage, intermittent connection, etc.) sistors into saturation. At the same time, power transistor Starting with both transistors cut off; C3 is discharging, Q3 is turned on by the current supplied through R7. its negative plate rising toward ground at ratea determined Dwell is determined by the time constant R6 x C3. When by various series resistances; while its positive plate is held the charging current of C3 diminishes below the value which near zero volts by a relatively low-resistance path through will sustain conduction of QI , then regenerativea action is R6 and R7 and resistora internal to Q3 across its emitter- again established, this time cutting off all three transistors. base junction. It is at that moment the high-voltage pulse is generated. The series combination of C5 and C6 (discussed later) has negligible effect on the charging rate, which is therefore Further Details determined mainly by C3 with the series combination of Capacitors C5 and C6 form a voltage divider which RATE control R9 and resistor R2 (or R2 alone, in the fixed- ensures rapid cutoff of QI; while C6 acts as a bypass to frequency version). prevent QI from being retriggered by pickup of the high- Capacitor C3 discharges fully, and then begins charging voltage pulse. in the opposite direction as its negative plate rises above Dwell must be long enough to permit the field around

R9 2.5M EG +12V RATE 4#r 11

s- - -• (SEE TEXT) R2* W 1. • + IE- Cll 470 R3 R4 10K 100K C5 1000p F 02 + 2N3906 R5 C2 1K 10 Q1 2N3904 02 2 1N914 C4 en. 0.27 C.). NE1 I ,I ' I -0 OUTPUT 400V T1 I I o o Ll i C3* R6" RB L-_ 1 10K 3- 4

CBE 2N3904,6 R7 e100 S2

Q3 B C MJE5742 MJE5742

Fig. 1—The pulse-generator version has variablea rate control and requires different values for C3 and R6 than does the fence charger version. The fence charger version has fixeda rate and uses conventionala transformer instead of an auto coil (see dashed lines).

- 25 - Ti to be fully developed to its steady-state condition un- The rapid collapse of its field when Q3 cuts off, as com- der all anticipated conditions of loading. Although the pared to the relatively slow 60-Hz sinewave for which it period is not critical, it may be set for optimum results is designed, explains how several thousand volts can be with particulara coil or transformer, as described later. developed across the 115-volt winding (E = L dedt). That A higher capacitor value at C3 is specified with the winding will typically be found to measure 30 to 120 ohms fixed-frequency, or fence-charger version, for reasons of DC, while the 12-volt winding will have a resistance of safety. It allows the use of a lower resistance value for around ohm.1 The author has tried many such transformers R2, reducing the shunting effect of dirt or moisture which for Ti, including the Stancor P-8392 and P-8393. (The latter might otherwise cause a significant increase in the repeti- provides somewhata bigger jolt although costsit more than tion rate. That is the reason we specify an axial type for the former.) The problem, however, lies in the breakdown C3, so that its pads are more widely spaced than they rating of the 115-volt winding. would be with radial.a In most transformers of the species, the winding is rated for breakdown at 1500-volts RMS (corresponding to 2100- volts pk), with safetya margin that may vary depending on the manufacturer; the Stancor rating proving remarkably conservative. The author subjected the winding of P-8393a to 40 million pulses of 4-Kilovolt amplitude without breakdown. However, he does not guarantee equally good luck in your application. One way to preclude breakdown with such transformera is to always operate the fence charger with an appropriate load. If your fence isn't long enough to load Ti to 2-3 Kv pk, you could reduce the supply voltage: Say, use 6-volta battery instead of 12 volt. Or you could substitute for bulb Il a type having lowera current rating, and therefore highera resistance. Either of those approaches, naturally, will somewhat reduce the effectiveness of the unit. Otherwise, the author offers transformera specially wound The frequency control is mounted on the see-through front panel behind the ignition coil here. The auto lamp on the for the fence-charger option and rated at 5 Kv pk (see note circuit board limits current, and lights theif power leads at end of parts list). are connected backward. The chimney protruding through the hole in the comer of the circuit board accommodates Other parts the 1/2-in, pipe used in the fence-charger version. A type MJE5742 transistor is specified for Q3, rated at 400-volts under heavy inductive load. However, you can at Power Amplifier some risk substitute the cheaper MJE5741 (350-volt rating) Because the field of Ti, as might be supposed, collapse or MJE5740 (300-volt rating), depending on T1. In any case, through the primary as well as the secondary, the inductiv breaking the circuit to an inductive load is tricky and so if "kick" comprises positivea pulse o n the co llector of Q3 • you plan extensive experimentation you should obtain fewa Capacitor C4 is required, as in the con ven tiona l auto - spare Q3's. ignition system, to prevent excessively rapid vo ltage build Potentiometer R9 for the variable pulse generator project up. Nevertheless, that reactive voltage r eac hes severa l hun - can be any 2.5-megohm unit from the junk box. If you use dred volts, and we take advantage of toit light neon indicator one with a linear taper, though, you will find the control NE 1. Thus, each flash verifies the integrity of th e power - very touchy at the high end of the frequency range. The amplifier circuit. simplest resolution of that minor inconvenience is to use an If no arc is drawn, the positive pul se on the collector ordinary audio-taper potentiometer connected backward; that of Q3 is followed by a negative-g oing excurs ion. Trans ise- is, with the high end of the frequency range at the CCW tor Q3, designed for inductive lo ads, con tains a shunt (counter clockwise) end. diode which prevents that "backswing" f rom being ap - For reasons already mentioned, the time constant C3 R6X plied to the base through the base-collector junction. That determines dwell, or "on" time. As we have said, dwell is diode also protects Q3 if the power-supply l ea ds are acc i- not critical; but if the capacitor you use for C3 is a low- dentally reversed. quality part with an excessively high equivalent series resis- Automotive lamp II, as we said, limits surge curren ts tance (ESR), then dwell may turn out to be greater than occurring as resulta of various normal operating conditions, necessary to serve the needs of Ti. If in doubt, use tanta-a as well as accidents, such as the re versa l of power-supply lum type for C3. polarity. Also, it absorbs the energy of th e bac ksw ing. The Incandescent Lamp The Transformer We have emphasized the importance of Il, the current- Practically any 12-volt ignition coil h av ing a primary limiting lamp, and have specified a type 1156 auto bulb. resistance of around 1.5 ohms will wor k as T1 for the The merit of an incandescent bulb as a protective device high-voltage pulse generator, but there's a m inor cons id- lies in the dependence of its resistance upon the value and eration in the choice of traa ns former for the fence -charger duration of applied current. With colda resistance of only project. A common 12-volt 1- am p trans former w ith 115- about 1/2-ohm, the Type 1156 degrades performance only volt primary can be used h ere —h oo ked up bac kwar d of slightly; but in the case of a current surge or accidental course, so that the 115-volt windi ng serves as secon dary. short circuit, its resistance quickly rises to a "hot" value

- 26 - of around 6 ohms, sparing power amplifier Q3 from the devastating requirement of breaking an excessive current into an inductive load. Nevertheless, there is some leeway in the selection of II. For instance, in the lab-generator version where the load has a DC resistance of 1.5 ohms, a lower-resistance bulb will give a slightly better spark at high frequencies. The author has used Typea 1157 bulb here, connecting its two filaments in parallel, with satisfactory results. On the other hand, as we have indicated above, to prolong the life of Ti in the fence charger, you may elect lower-currenta or higher- resistance bulb. Try the smaller of the two 1157 filaments, alone before experimenting further. After the unit is built feel free to try others.

PARTS LIST FOR THE FENCE CHARGER SEMICONDUCTORS D1—No D1 in project: please ignore 02-1N914 silicon diode or similar This Is a top view of the circuit board. Note the ample Q1-2N3904 NPN silicon transistor or similar space provided between components. That is to prevent Q2-2N3906 PNP silicon transistor or similar arcing between the leads of high-voltage components. 03—MJE5742 amp,8 400-volt, NPN Darlington power transistor (see text) Circuit Construction All parts for either version of the project are available, CAPACITORS including the 2-piece plastic cabinet having provisions for C1-470-p,F, 16-WVDC electrolytic mounting at the end of an ordinary 1/2-in, water pipe or upon C2-10-µF, 16-WVDC electrolytic a standard camera tripod. You may choose to build either C3—For lab model: 2-p.F; for fence charger: 10-µF, both 16-WVDC electrolytic, axial (see text) version of the project in whatever kind of cabinet suits your C4-0.27-pf, 400-WVDC film needs. If you decide to use wire-wrap construction however, C5-1000-pF disc the ground bus and all connections in the power-amplifier C6-0.01-p.F disc circuit should be made with wire no smaller than #24 gauge. In the author's prototypes, power transistor Q3 stands off RESISTORS the circuit board; but if space limitations permit, a slight (All fixed resistors are , 4-watt, 5% ) R1, R7-100-ohm margin of safety is affordable by bolting downit flat so that R2—Selected (see text) the circuit board provides measurea of heat dissipation. R3, R8-10,000-ohm Omit R2 from circuit board and don't connect the supply R4-100,000-ohm conductor to the plus end of Ti until ready to fire up. R6—For lab model: 470-ohms; For fence charger: Also, leave the secondary leads unconnected for the fence 150-ohm charger. R9-2.5-megohm pot (see text) In planning chassis layout, keep high-voltage output con- ADDITIONAL PARTS AND MATERIALS ductors well away from the circuit board, especially in the T1—For lab model: Wells C1819 or similar ignition coil; version using an ignition coil as output transformer. A me- 1.6-ohm primary, 10,000-ohm secondary; For fence tallic or otherwise conductive cabinet must be connected to charger: 12-volt, 1-amp transformer (see text) the circuit common. Since 30a -kv pulse is capable of jump- NE1—Neon glow lamp; Type NE-23 or equivalent ing a 1-in, gap, however, you may have some difficulty 11-12-volt, 2-amp automotive bulb, Type 1156 or finding a feedthrough insulator big enough to handle the equivalent high-voltage conductor. One way to meet that requirement Cabinet or case: circuit board; solder lugs of various is to use a spark-plug wire, which may be passed through gauge, with internal teeth; cable to power supply, #14 the cabinet wall using only grommeta to prevent chafing. to #18-gauge zip cord or whatever suits, spacers, screws, Or the neck of the coil itself may be used as feedthrougha nuts, lockwashers, hookup wire, cable ties, solder, etc. device, as in the author's mode of construction. Additional parts for the lab model only: 1-in. to 2- 3/4-in. radiator-hose clamp to mount ignition coil; 7-mm spark- Lab Cabinet Loading plug wire, coil clip, coil nipple, alligator clip, alligator If you are using the author's recommended cabinet, situ- insulator; knob for R9; two banana plugs or other suit- ate the circuit board in the left end of its bottom. The board able terminations for cord to power supply. itself can be used as templatea for drilling the four mounting Additional parts for the fence charger only: two battery holes in the bottom of the cabinet. Mount the board.assembly clips, Mueller #46C or the like; 112-in. pipe, 1 1/2-in. large on four 7/16-in. metal spacers. The conductive coating in the nipple, coupling, etc., for grounding system. cabinet bottom may be grounded with a solder lug placed All parts except water pipe, caulk, hookup wire, and under one of the screws securing the board to spacer.a solder are available individually or in kit form from Maps For variable-frequency or lab model, situate the RATE con- and Zaps. 1132 Roseta Dr., Topanga, CA 90290. Please trol R9 in the clear-plastic front panel. Bring the power cable write for price list, sending self-addressed envelope with into the cabinet through the hole in the bottom rear, using a 450 postage. suitable grommet.

- 27 - The coil mounts on platforma toward the other end and The chimney referred to earlier provides the means for is secured with a hose clamp. Using the coil called out in connection to an external ground. A pipe nipple and cou- the parts list, some filing of the platform is required. The pling are required. First solder a length of hookup wire to coil case must be grounded or internal arcing may occur. the inside of the nipple. A hot iron (say 200 watts) is required Do not depend on casual contact between the coil case and for good wetting. Loosely engage the coupling to the nipple; the conductive coating. A grounding connection can be made and passing the wire up through the chimney, screw the by inserting an internal-tooth solder lug between the clamp nipple into the opening by turning the coupling. The nipple and coil case. At its base, the coil is stopped by its neck may engage the coupling as it engages the chimney. Al- passing through holea drilled in the end of the cabinet top. though the chimney hole is not threaded, the nipple will Hence, it's not likely to come loose with normal handling. nevertheless seat securely. Turn the coupling until it is tight At the free end of spark-plug wire install an alligator up against the bottom of the cabinet. If desired, apply super clip or other suitable connector. At the other end, first slide glue sparingly around top edge of the nipple, bonding it the coil nipple onto the wire, and then install the coil clip. permanently to the chimney. Now, if you later need to re- Important: To preclude arcing, solder the end of the wire move the coupling for any reason, the nipple will remain in to the clip. Push it into the coil neck and slip the nipple into place. Solder other end of the wire to common at the circuit place. When the top is installed later, the nipple provides board or at one end of the lugs on the transformer flanges. a tight seal. High-Voltage Attenuator Fence Charger Version Before proceeding with test and adjustment, you may wish Construction of the fence-charger version is somewhat to provide yourself with some means for measuring voltage simplified by less-stringent needs for insulation and by the pulses beyond the range of your oscilloscope. To that end, more conventional mounting means for Ti. Whatever chas- you can build 90-megohma attenuator, as shown in Fig. 2. sis layout scheme you employ, however, the Earth ground- When used with a standard 10-megohm probe, the device ing requirements described above also apply to this model: extends the vertical range of your scope by factora of ten. If you use conductivea cabinet, it must be connected to the The attenuator consists of nine 10-megohm resistors con- circuit Earth, and so must the case of T1. Don't forget that nected in series. A length of spark-plug wire provides sup- a means must be provided to connect that common to an port for the resistor array and also serves to introduce distrib- external ground. uted capacitance for AC equalization. To preclude arcing, In the author's model of the fence charger, Ti is mounted each end should extend an inch or two beyond the terminal. in the cabinet bottom. To ensure gooda connection to the Once you have commissioned your pulse generator or transformer case, first scrape any varnish or wax from the fence charger, you can fine tune the attenuator by adjusting mounting flanges. Then mount with 1/2-in. metal standoffs the bus-wire gimmicks at either end of the spark-plug wire. and 8-32 hardware. Use two or three solder lugs as required That is most easily done by generating high-voltagea pulse for various grounding connections. within the range of your oscilloscope (say 1600 volts peak), Mount ceramica feedthrough insulator in middle of the measuring with only the 10-megohm probe; then, trimming platform for fence connection. The underside of the platform the length of the gimmicks to give the same defection with comprises recessa which, in an outdoor installation, keeps the probe connected to the 90-meg attenuator (setting the the output end of the insulator clean and dry. sensitivity 10 times higher, of course).

"GIMMICK" (2) NINE 10-MEG 1-W RESISTORS IN SERIES

3/8 IN. MIN 7-MM SPARK-PLUG WIRE

HI-TENSION LEAD FROM PULSE GEN

RETURN LEAD

CONVENTIONAL 10X, Fig. 2—This is a high-voltage attenuator. Used with 10x probe, extent.it 10-MEG PROBE the scope range by factora of 10. Spark-plug wire serves to support resistors and provides distributed capacitance for AC compensation.

- 28 - Selecting R2 2. Turn power off and ground the one which drew the We had advised you during construction to omit one con- smaller arc. Connect the other to the output feedthrough. nection to the primary of Ti so that you can now select R2 3. Reconnect scope, apply power, observe polarity of without energizing the power amplifier. Using clip leads, output pulse. If you get positivea pulse, reverse the primary first connect typical value shown in parts list. Then connect connections. A negative pulse jumps a longer gap from a your 'scope to the junction of R6 and R7, and apply power. small object (the fence wire) to largera one (the victim) than For the lab pulse-generator version, now set the RATE con- does positivea pulse (believe it or not). trol to maximum frequency and select valuea for R2 which If you wish to view the current pulse, temporarily hook gives repetitiona rate of about 20 Hz. For the fence-charger 0.1 to 0.2-ohm resistor in series with negative power-supply model, select a value which gives the desired rate, but no lead, and connect a 'scope across it (being careful to avoid higher than 60 times per minute. Remember that the slower ground loops, as can arise though test connections or via the the rate, the longer between recharging. power-line safety ground). With fence-charger option, if Now turn the supply off and add the missing wire to the possibje, stimulate 1-mile wire by connecting 0.015-µF, power-amplifier circuit. In the author's lab-generator chas- 2000-WVDC capacitor across its output. A rising waveform sis layout, it is necessary to first loosen the coil in order characteristic of inductor charging should be obtained—the to free the circuit board. If you plan to test the unit with' abrupt drop at its trailing edge of course representing the the circuit board loose, be sure to temporarily replace the cutoff of Q3 and the generation of the high-voltage pulse. lugs grounding the coil case and cabinet. Place cardboarda With the lab-generator version, dwell is not critical thanks sheet under the circuit board to insulate it from accidental to the relatively low inductance of the typical ignition-coil contact with the cabinet coating, etc. The unit is now ready primary. In the fence-charger option, however, primary infor performancea test. ductance will probably be much higher and will vary considerably depending upon your choice of transformer.a Fig. 3 shows the current waveform typical of such primary.a If it ends too soon, that is before the field has reached its steady- state value (A), then maximum output capability cannot be attained. If it ends too late (B), then average current con- sumption is higher than necessary. To get optimum results (C), adjust the width by changing R6 as needed.

A 8

The fence charger model has a nipple for mounting on a length of water pipe. It effectively grounds the system. Fig. 3—For the fence-charger option, this is currenta pulse seen across the small resistor in series with the supply: Testing In A the pulse width is too short; In (B) The pulse width is Connect the high-voltage output to the 90-meg probe or too long; and in (C) the pulse width is correct. whatever instrument you wish to use to observe the high- voltage pulse. Turn the power supply on and gradually in- If you know the exact value of the small resistor, given crease the voltage (adjusting the lab-generator rate as de- the peak voltage appearing across it you can now calculate sired), synchronizing the 'scope to display the largest excur- peak current (I = E/R). A typical value is 4 to amps.6 sion. (When you don't know exactly what to expect, it's easy to be fooled into syncing on the backswing or some Buttoning Up other minor lobe.) Reinstall the circuit board, remembering to replace the The unit should start working at supplya voltage of 1.5 lugs which ground the cabinet, pipe coupling, Ti, case, to volts,3 but it will shut itself down down if you vary the etc., and to secure the coil. Test the unit once more, then voltage too abruptly. If that happens, just turn the power off assemble the cabinet. and then back on. If you're using the author's recommended cabinet with At a 12-volt input you should get pulsea of about 20 to the pulse-generator option, leave the high-voltage cable and 30 Kv pk from the lab generator or 3.5 to 5 Kv pk from the nipple connected to the coil, passing the other end through fence charger. In the latter version, proceed as follows to the hole in the cabinet top as you bring the top into place. decide which secondary lead should be grounded: Slide the front panel up into the cabinet top. Now, close 1. Turn power off and disconnect scope from both ends. the cabinet by swinging the left side down. Moderate force Turn power back on, and using an insulated tool (to avoid is required to push the coil nipple into the hole. Make sure getting zapped), bring each end in turn to the transformer tongues in tf... cabinet top engage the mating slots in the case, leaving the opposite end free. One will probably draw bottom, and hold it together with one hand while installing a small arc and the other won't. (Continued on page 32)

- 29 - ir is the most important ingre- BY JOHN IOVINE dient to our survival. Think about Ait; you may survive a few days without water, a little longer without food, but, deprived of air, your survival time can be measured in minutes. The quality of the air surrounding many cities has become so poor that many local news stations provide an air-quality report along with the weather forecast. Air pollution is so commonplace now that words have Negative-Ion been created to describe it. The word "smog" for example, a contraction of the words "smoke" and 'fog." As smogif wasn't enough, today there are new pollution concerns, not the least of Generator which are the increasing CO 2 level, the green-house effect, the depletion in Make the very air you breathe healthier the ozone layer, and acid rain. with an easy-to-build high voltage project. Research. Long before there was any talk or concern about air pollution and such, some scientists and experimen- To summarize a few points from his negative ions. Greater than 19% of the ters noticed that the ionization of even book, negative ions help elevate reports described no effect, and a few clean air can improve its quality. Clean mood, enhance physical performance (less than 1%) detailed some detrimen- air (principally composed of 78% nitro- and training, and sterilize harmful air- tal effect. Since the preponderance of gen and 21% oxygen) typicallyis full of borne bacteria. An abundance of the evidence supports the beneficial positive and negative ions in approxi- positive ions on the other hand can be effects of negative ions, feltI that build- mately a 5-to-4 ratio. What researchers held responsible for a number of low ing an ion generator was a worthwhile found was that when this ratio changes grade medical problems, such as fa- project. A summary of some of the ben- one way or the other hasit an effect on tigue, headaches, and anxiety. eficial effects reported by some re- biological systems. There are detractors to this point of searchers are listed in the boxed text This idea was popularized by Fred view. So before startedI to design a entitled "The Positive Effects of Negative Soyka who, in the 1970's, wrote a book negative-ion generator, didI some re- Ions." is It by no means an exhaustive list, titled "The Ion Effect." Mr. Soyka studied search to find out itif would be worth- it's justa sampling of the scientific bene- natural occurrences of negative and while. surveyedI approximately 100 fits noted. But thisif theis case wouldit positive ionized air. His findings and in- world-wide scientific reports on the be to our benefit to improve the quality quiries demonstrated that negatively effects of negative ions from 1973 of air that we breathe with a negative ionized air had substantial health ben- through the present. canI report that ion generator. efits. out of my survey approximately 80% of Despite the numerous scientific re- the citing's note the beneficial effects of ports supporting the health benefits of WARNING!! This article deals with and involves subject matter and the use of materials and substances that may be hazardous to health and life. Do not attempt to implement or use the information contained herein unless you are experienced and skilled with respect to such subject matter, materials and substances. Furthermore, the information contained in this article is being provided solely to readers for educational purposes. Nothing contained herein sug- gests that the negative-ion generator described herein has any health benefits whatsoever. Neither the publisher nor the author make any representations as for the completeness or the accuracy of the information contained herein and disclaim any liability for damages or injuries, whether caused by or arising from the lack of completeness, inaccuracies of the information, misinterpretations of the directions, misapplication of the information or otherwise.

- 30 - DISCHARGE POINT The Positive Effects of Negative Di 1 IMD5210 Ions 14 PL1 C3 • Learning enhancement in normal and .002 I learning-disabled children. The task used T2 C4 to test the children was dichotica listening 12V: 8KV .002 I test. R2 1K • Negative Ions can be used to decrease amounts of radon in a building atmo- Q1 R1 sphere. U1 TIP120 15K 02 • In one animal study 1279 calves were 555 2N3055 broken into two groups, one of 649 head R3 and the other of 630 head, negative air 2.2K ionization was used to test for prophylac-a 12V Cl C2 tic effectiveness against respiratory dis- R4 1000 T.047 47011 eases. The results were remarkable: In the treated group (649 head) 45 calves became sick and died.3 In the control group Fig. 1. This is the simple schemaec for the Negative-Ion Generator. Note that it is (630 head) 621 became sick and 33 died. • A 40-50% reduction of microbial air pol- basically a timer that induces a high voltage in the windings of T2. lution in dental clinics. • A test using college students showed ionized air, no manufacturer of nega- research papers supporting this article improved performance on a visual vigilance task. tive-ion generators can make any are listed in the text entitled "Bibliogra- • In 1983 it was reported that chickens health-benefit claims without running phy" so you can do the research on raised in negativelya ionized atmosphere afoul of the FDA. For that reason alsoI your own and make your own decision. showed improved anabolic processes. The will make no such claims. Instead, the chickens raised in negatively ionized air had an overall greater weight than controla The Ion Generator. The design of the group fed the same quality and quantity of Negative-Ion Generator is fairly feed. The meat of the treated group had PARTS LIST FOR THE straightforward (see Fig. 1). The circuit is higher protein and essential amino-acid NEGATIVE-ION GENERATOR a high voltage generator. containsIt a content. In addition, higher concentrations standard 555 timer that's used to gener- of vitamins E and A were found in their livers. SEMICONDUCTORS ate square-wave pulses. The pulses are U1- 555 timer, integrated circuit applied to the base of the TIP120 NPN QI—TIP120 NPN Darlington transistor Darlington transistor. The Darlington Q2-2N3055 power transistor Bibliography provides sufficient current to the base D1-10,000-volt, 10-mA silicon rectifier of the 2N3055 power transistor to turn it diode (see text) on. Each time that happens, current Negative Air Ion Effects on Learning Dis- BR1-4-amp, 50-PIV bridge rectifier abled and Normal Achieving Children, L.L. flows through the high voltage auto- Morton and J.R. Kershner, University of RESISTORS transformer, T2. The high voltage lead of Windsor, Faculty of Education, Ontario, (All resistors are /14 -watt 5% units.) the transformer is connected to a 10 Canada, 1990 RI-15,000-ohm R3-2200-ohm kilovolt high voltage diode. Notice the R2-1000-ohm R4-470-ohm Effect of Negative Ion Generators in Sicka polarity of the diode. It is biased to M.J. Finnegan, C.A. Pickering, CAPACITORS place a negative charge on 03 and ES. Gill, I. Aston, and D. Froese, Depart- C1 1.000 p.f, ment of Thoracic Medicine, Wythenshawe - - 15-WVDC electrolytic 04, leaving the discharge point nega- Hospital, Manchester, England, 1987 C2-0.047-µF, polyester-film lively charged. The voltage at the dis- C3, C4—.002-p,F, 6,000-WVDC, charge point negatively charges the Aeroionization in Prophylaxis & Treatment ceramic-disc of Respiratory Disease in Calves„ TI. Sol- air forced past byit the fan. ogub, N.F. Borzenko, V.P. Zemlyanskiy, and ADDITIONAL PARTS AND MATERIALS The author's prototype was built on K.F. Plakhotnyy, Russia 1984 TI-12-volt, I.2-amp power transformer sections of perfboard using point-to- Effect of Ionization on Microbial Air Pollu- 12-12 volt to kilovolt8 autotransformer point wiring. is It a suitable method that tion in the Dental Clinic, J. Gabby, O. FAN1-12-volt DC fan you can use in your own ion generator Bergerson, N. Levi, S. Brenner, and I. Eli, Enclosure, line cord, switch, TO-3 provided you follow some precautions: Research Institute for Environment Health, socket and heat sink, perfboard, wire, Make sure you place 03, 04, D1, and Sackler School of Medicine, Tel Aviv, lsreal, solder, etc. 1990 the discharge point (which we'll de- The following are available from Images scribe momentarily) on a piece of per- Effects of Ionized Air on the Performance of Company (P.O. Box 140742 Staten a Vigilance Task, G.C. Brown and R.E. Kirk, fboard all their own. The junctions Island, NY 10314-0024; Tel. Systems research Laboratories, Inc., between those components should be 718/698-8305): D1 ($1.50), C3 and C4 Brooks AFB, TX, 1987 at least a centimeter apart. Both this ($1.50 ea.), and T2 ($17.95). Include little high voltage board and the auto- "Negative Ions," JE. Wright, Muscle & Fit- an additional $2.50 for shipping and ness Magazine, January 1991 handling. NY residents must add transformer should also be kept at least Effect of Artificial Air Ionization on Broilers, appropiate sales tax. 1 centimeter away from the perfboard P. Stoianov, G. Petkov, and B.D. Baikov, Vet containing the other components, the Med Nauki, Bulgaria, 1983 NOTE: The above prices are subject to fan, and the power transformer. change. In passing, youif chose not to buy D1

- 31 - NEGATIVE ION GENERATOR from the supplier mentioned in the Parts List an exact replacement may be difficult to locate. However, even though hasit a lower current rating, an ECG-518 should work fine, although this substitution has not been tried. The discharge point should be "pointy" to enhance the ionization of the air. You can use sewinga needle, for example. An alternate discharge point can be fashioned from smalla piece of No. 22 stranded wire. Strip off about Y2 inch from one end of the wire and sep- As you can see, there is some space left between components. Note that in the prototype arate the fine copper strands so that two capcitors have been used to emulate CI. they are more or less evenly dispersed. When the wire connectedis to the high negative voltage, the end of each and out through an opening hole at any arcing or discharge from the high strand will behave as dischargea point. the top of the enclosure. voltage transformer or high voltage ca- You can use any enclosure large Any screen or covering on the fan- pacitors, cut the power immediately. enough to hold all the components. I'd outlet hole should be non-metallic or Let the project sit for awhile to let the recommend using plastica enclosure if plastic in nature. Using a metal screen capacitors discharge, and, without one available.is You should place fewa would severely cut the efficiency of the touching the project if possible, coat air holes in the side or bottom of your generator because the negative ions the faulty area with a little "No Arc" enclosure for the fan to draw air in. The that come into contact with the metal spray (available from Radio Shack). Al- fan should be situated in The enclosure screen would be neutralized. low the material to dry before testing to pull air in past the discharge point(s) When testing the circuit, youif see the unit again.

(Continued from page 29) If you need one pulse at time,a or bursts of pulses, con- the cabinet hardware with the other. Turn the five bottom nect pushbuttona or momentary switch in series with one screws snug, but not tight. of the power-supply leads. Cabinet assembly of the fence-charger version is easier If you have trouble getting lower output voltages, but not cause you don't have to cope with the coil neck or connec- higher, the spark-plug cable may have pulled loose. When tions to the front-panel potentiometer. For outdoor use, how- that happens, high voltage settings give what appears to be ever, you will have to caulk seams against the weather. normal performance because the spark path is completed Silicone rubber is good for that purpose because it can later by jumping within the neck of the coil; while at the lowest be peeled off if servicing becomes necessary. Acrylic rubber voltage settings it appears not to be working at all. If that makes bettera seal, but because it sticks more tenaciously, difficulty is encountered, pull the cable out, inspect the sol- it makes later disassembly more difficult. der joints, then simply push it back into the coil. Carefully apply a very thin bead first along the inside Using the author's cabinet and construction techniques, edges of the opening in the top front, and install the front the fence-charger ground connection is made through the panel. Then, again very carefully and sparingly, apply a pipe fittings sticking out of the bottom end of the chimney. bead along the slot in the cabinet bottom; and finally, as- An Earth-ground means is provided by an ordinary 1/2-in. semble the top and bottom. Depending on your skill in the water pipe. The length should be chosen to permit the pipe application, there may be some squishing around the seams. to be driven at least 3 ft deep, but the deeper the better, Surplus material around the outside can be peeled off later, depending on estimated conductivity of the soil; with enough after the sealant has set. pipe rising above ground to place the unit at comfortablea viewing level. Thus a pipe of at least 7 ft is required. A Installation and Operation more effective ground can be had by adding salt to the soil. For maximum safety, you should, if feasible, connect one Temporarily screw a pipe cap onto top end so as to side of the lab-generator power supply to Earth ground. If protect the threads during hammering operation. Pound it not, then be sure to provide returna path for the spark to into the ground, remove the cap, and screw the fence- one of the power-supply leads. Set the RATE control to get charger assembly onto the end. Connect the fence and bat- the desired rate, and the power-supply voltage to get the tery to the unit. desired output potential. If the output is not excessively loaded, The neon lamp flashes with each pulse to assure you that the small in-circuit neon lamp flashes with each pulse. The everything is working okay, except in absolute darkness, auto lamp may glow dimly when the rate is set near its upper since a few photons of light are necessary to prime the limit, but otherwise it should never light during normal op- neon. That apparent drawback, however, has the definite eration. It does light brightly to warn you when the power earmarks of an advantage because when it's pitch black the leads are reversed or if there is an internal short. unit cannot call itself to the attention of an interloper. • - 32 -