Our solid-state Tesla SOLID STATE coil can produce nrwvaer TESLA COIL output of about 100,000 volts, DUANE A BYLUND ‘ I ‘sla coils have been around for almost 100 years and, with the exception of vacuum-tube driven coils, not much has changed from the way Nikola Tesla invented them. This article describes a new lype of ‘Tesla coil; a true solid-state Tesia coil. One thing that makes our Tesla coil unusual is that the coupling to the secondary coil is by a direct electrical connection rather than by magnetic fields. Direct coupling is not new to ‘Tesla coils but it is seldom seen. ‘The solid-state Tesla coil ts by no means as spectacular as capacitive- discharge Tesla coils but it gives just as good, or better, performance as a vacuum-tube Tesla coil. Sparks as long as 8 inches are possible with a power-line consumption of 2amps at 120 volts (see Fig. 1), and the output reaches a peak of about 100,000 volts. Although the average power in- put to the device is around 250 to 300 watts, the peak input power to the Tesla secondary coil {s about 800 walts. The Tesla coil is an exccllent teaching tool, as many interesting things may be learned with the aid of this device. Circuit description ‘The schematic for the solid-state Tesla coil is shown in Fig. 2. The sec- ondary of the Tesla coil, when directly driven bya.solid-state driver, appears likea series RLC circuit. That's due to the self-capacitance of the coil with respect to ground. The capacitance is normally very small with the induc: tance being fairly large. At the reso: nant frequency, the inductive reac- tance cancels the capacitive reac- tance. The effective impedance Is limited by such losses as the DC re- sistance of the coil, AC skin effect of 1661 usaWaLe3s &RADIO-ELECTRONICS the wire, eddy currents induced in nearby objects by the field of the coil, and so on. Series RLC circuits have rela- tively low impedances when oper- ated at the resonant frequency. The coil used in this project, when operated at its resonant fre- quency, looks like a 450-ohm re- sistive load to the solid-state driver. Series RLC circuits pro- duce high voltages on the induc- tor and capacitor at the resonant frequency. The bigh voltage is due to a high current flowing through a high reactance (re- member that the inductance is large and the capacitance is small, creating large reactances in each component at a given fre- quency}. That is what produces the corona discharge at the end of the secondary coil ‘The heart of the driver is IC1 the SG3524 pulse-width modu- lator. The duty cycle is fixed at about 45% for best efficiency. The frequency Is controlled by the re- sistance on pin 6 and the capaci- tance on pin 7. With the values shown, the frequency has a range from 200 to 240 kHz. A flip- flop inside the chip divides that by 2so that the effective output of the driver has a range from 100 to. 120 kHz. The outputs on pins 12 and 13, are 180 degrees out of phase with FIG. 1—THE SOLID-STATE TESLA COIL, can produce sparks as long as 8 inches. The output reaches a peak of about 100,000 volts. Warring! This’ article deals with and Involves subject rater and the use of materials and substances that may be hhazardous to health and fe. Do nol at No implementer use the information Conlained herein, unless you are experi- fenced and aklled with respect to such subject malter, materials, and sub: stances. Nether the publisher nor the au: ‘hor make any representation 3s for the ‘completeness or accuracy of tha informa tion contained hevein, and disclaim any libiity for damages oF Injuries, whether ‘caused by oF avising rom the fack of com- pleteness, inaccuracies of tho inform tion, misrepresentations ofthe eirectons, ‘misapplication ofthe information, orother- each other, and drive the gates of MOSFET’s QI and 92, which, in turn, drive the primary of trans- former TI. Transformer T1 drives the bases of switching-tran- sistors Q3 and Q4. The compo- nents in the base circuitry are used to increase the switch speed of the transistors. Tran: sistors Q3 and 94 switch the line voltage across the primary of T2, which increases the voltage and drives the end of the secondary coil directly. Note that the line voltage delivered to T2 is half- wave rectified by D1. That is im- portant to the operation of the Tesla coil because a pulsating voltage is needed to produce the best effects. When the device is plugged into a wall receptacle it will be in its standby mode. That is, the 21- volt power supply will be opera- tional and the FET's will be driv- ing the primary of Tl. The standby mode produces enough power to “tune” the driver to the coil’s resonant frequency before full power is applied. (Remember that the resonant frequency can be affected by nearby objects.) ‘The current supplied to the sec- ondary coil is indicated by LED1 ‘Tuning is accomplished by ad- Jjusting the frequency via Ri and observing LEDI. When reso- nance Is achieved, the secondary coil will have a low impedance which will produce maximum current, lighting the LED. Di- odes D3—D6 limit the forward and reverse voltages on LEDL when in the high-power mode. (Note that you must use an LED that lights at 1.5 volts—some LED’s, including most green ones, need 2.1 volts or higher. When the device is switched into the operating mode (or the high-power mode), half-wave line-voltage pulses will be applied to the primary of T2. As the hall- wave voltage Increases. the cur rent in the secondary coil in- creases and the energy stored in the inductance and capacitance of the secondary coil will in- crease. During this time there is, no corona from the secondary coil {if the coil is constructed as shown in this article). Sometime before the half-wave line voltage reaches its peak, the corona will appear on the Secondary coil, which will dissipate the stored energy very quickly. During the remainder of the half-wave line voltage. the coil will produce cor ona but the energy level will not be as great as the initial dis- charge. The coil will produce six- ty individual corona discharges every second, although you'll see a continuous discharge. 5 -1. eS USE THIS FOIL PATTERN, shown half-size, to etch your own PC board.FIG. 2—SCHEMATIC FOR THE SOLID-STATE TESLA COIL. The secondary ofthe Tesia coil appears like a series RLC circuit due to the s ‘ground. All resistors are Vewalt, 5%, un- less otherwise indicated. Rt—1000 chins, 10-turn ‘potentiometer RR—8500 ohms RS, 42900 ohms, Ye watt 5, RE—2200 ons FR7—330 obs, 1 watt AB, R8—0.56 chms, 2 watts, ‘flameproot ‘R10, R11—3300 ohms Capacitors G1—0.001 iF, 50 volts, 5%, polyester AOE Sd wos pepeser C3, C410 nF, 85 volts, tantalum C5990 uh, 95 volts, eioctrotytic 68, G72 HF 200 volts, nonpolar fim-type ‘©8,CO—0.02 nF 1000 ots, core ‘Sbmivonductors: 3G1—-SG3528 pulse-width modulator ‘Dt—MATSI dodo (02-Dé—IN4934 diode. (D7, D8—aN4936 diode (09—notusod ‘O10-D17—1N4008 diode Qi, OP—SK9IS power MOSFET PARTS LIST 03, O4—SNEB7S oF SK9140 NPN ansistor LEDI—red LED. Soo text ‘Other components: FI—S.amp, 250-0, fastblow hse BRI—VMO8 bridge rectifier, Varo ‘Ti—hand-made transformer (the onre ts TOK # PCOEER25 6-2 and the bobbin is TOK # BEER-25.5-116CP) ‘T2—hand-made transformer (the ‘oo is TDK # PC30EC70Z and ‘the bobbin is TOK # BEC-70-$18) ‘TS—hand-made transformer (the ‘core is TOK # POSOEERDS5Z and the bobbin is TOK ¥ BEER-25.5-1IBCP) ‘T4—TISVACIISYAC contentapped ‘wanstormor (Tad F-132P) SI—SPST hey switch Miscellancous: enciosure, slumi- ‘um angle bracket, high-voltage wire (to connect main unit to Tesla ‘secondary), 30-gaugemagnet wire for Testa secondary and Lt and L2, BA-gauge magne! wire for L3 and U4, 18-gauge stranded hook-up wire Yor LS and LB, 15-gauge mag- ‘capacitance of the coil with respect to inet wire tor 12 primary, 26-gau Sete a gauge magnet wire Yor both wit ings of TS, brass rod, discharge ‘ball, hardware, A Iineoord, etc. Note: TOK ferrite cores and bob= bins are available from MH&W International, 18 Leighton Place, Mahwah, NJ 07430, (201) ‘891-8800. The following items ‘are available from Corona Coll, PO Box 474, Riverton, UT 84065: TH-$15.00 PC board—St5.00 © Aluminum angle bracket (heatsink and PC-board mount)—$5.00 A 124-page book by the author, ‘Modern Tesla Col Theory, is available for $16. Please add $15 S&H for the Tesla ‘Seoondary, and 10% S&H for all ‘other items. lest waewaLdasRADIO-ELECTRONICS eanana “7 PLUBUAGK qr Sap SFo0r ay wine ot Testa ssecoNDaRY FIG, —PARTS-PLACEMENT DIAGRAM. It's best to play it safe and use the PC board for this project; we've provided the foil pattern if you would I Construction Most of the construction is fair- ly simple if the printed circuit board is used. A parts-placement diagram is shown in Fig. 3, and we've provided the foil pattern if, you would like to etch your own board. Figure 4 shows the com pleted prototype board housed in its aluminum enclosure, The mast difficult item to con- struct will be the Tesla secondary coil, followed by T) and T2. The to etch your own board. secondary coil may take an hour or so to make if you prepare ahead of time. Préparation in: cludes making some device that will easily rotate the coil form while winding the wire. The au- thor used a small lathe and it took about 15 minutes of actual winding time and 30 minutes to get set up. Do not deviate at all from the following parameters of the sec- ondary coil! Any deviation will FIG, 4—HERE’S THE AUTHOR'S COMPLETED PROTOTYPE housed in its aluminum enclosure. I's important that the case be properly grounded. change the characteristics of the coil and it may not operate with the driver unless modifications in the driver are made. Any change in physical dimensions or wire size will alter the resonant, frequency and effective imped: ance of the coil. Any change to the discharge electrode will effect the maximum energy obtainable. ‘he coil form for the secondary winding 1s a standard 5-gallon plastic container 10 inches in di- ameter at the bottom, 12 inches in diameter at the top. and 14 inches long. The bottom of the container becomes the top of the coil. To make winding easier you should drill a hole about an inch in diameter through the center of the bottom of the container. A similar hole should be drilled through a removable lid and then the complete coil form ean be ro- tated easily on a dowel. Start the secondary winding | inch from the small-diameter end and close-wind 30-gauge magnet wire for a total length of 10 inches. It does not matter what direction the wire is wound in. When winding the original coil for this article, shellac was used to lubricate the wire as it was wound and also to act as a sea- lant afterwards. It was difficult to wind the coil because the coil form was very slick and had a slight taper to it and, as a result, the wire kept slipping, It may be easier to spray the container with adhesive before winding the wire to make it stay in place. A couple of coats of shellac should be ap- plied to the finished winding. You also must put 3 or 4 beads of sil- icone sealant around the end of the winding at the top of the coil to keep corona discharges aw: from the area. If corona dis charges appear along the coil at the top it will imit the maximum, energy and destroy the coil form, ‘The discharge ball. or elec- trode, is a brass-plated metal doorknob, 1-inch in diameter. that can be found in hardware stores (see Fig. 5), The ball 1s mounted on a 4-inch brass rod: you can drill and tap the ends of the brass rod with a 6-32 tap (or whatever matches the threading, on the doorknob) to make mounting easier. The brass rod is connected to the coil form by two pieces of plastic, one on each side of the coil form, over the '4-ineh,PLasTic supPORT OW INSIDE OF COIL FOR FIG. 5—THE DISCHARGE BALL Is a brass-plated metal doorknob, -inch in diameter. The balls mounted on a 4-inch brass rod that's beon tapped to make mounting the bale hole. A 6-32 screw passes through the pieces of plastic and into the brass rod to hold the sembly together. The wire is sol- dered to a lug held in place by the 32 screw. ‘A banana jack is used to make connections at the bottom of th coil. Locate the jack about %- inch from the edge of the wire on the coil. Silicone should be used to insulate the connections be- tween the magnet wire and the brass rod and banana jack. The finished coil, when built exactly as we've shown, will have a reso- nant frequency of about 110 KH. ‘Transformer TI is made with a ferrite core and bobbin from TDK (see the parts list). Coils Ll and L2 are wound first with 30-gauge magnet wire, 16 turns each. making one layer on the bobbin. The two windings are bifilar wound, as shown in Fig. 6-2: LI starts on pin 3 and L2 starts on pin 4. Wind both in a coun- terclockwise direction while look- ing at the top of the bobbin. ‘Terminate L1 on pin 1 and termi- nate L2 on pin 2. Put a layer of cellophane tape on top of the winding to insulate it from L3 and La. Coils L3 and L4 are made with 5 turns each of 24-gauge magnet wire and are also bifilar wound, on top of Li and L2, and in the same direction. Coil L3 starts on pin 6 and LA starts on pin 8. Ter- minate L3 on pin 5 and terminate L4 on pin 7. This completes the transformer until it is mounted on the PC board. Put the two core halves through the bobbin and put tape around them to hold them in place. As shown in Fig, 6. LS and L6 are wound after the trans- former is mounted on the board: L5 and L6 are wound with 18- gauge stranded hook-up wire with one turn each. Solder the collector (Q4) end of L6 to the PC board. Go one turn in a coun- terclockwise direction around the core of T1 and then terminate the other end of L6 at the primary of T2. Solder the collector (Q3) end of LS to the PC board and go in a clockwise direction around the core of TI for one turn, termi nating the winding at the cath- ode of D1. ‘Transformer T2 is also made from a ferrite core and bobbin from TDK (again, see the parts list). The primary is 10 turns of 15-gauge magnet wire, although asmaller gauge, say 18, can prob- ably be used. It does not matter what direction the wire is wound in but the turns should be equal- ly spaced across one layer of the bobbin. Put several layers of cel- lophane tape on top of the pri- mary to insulate it from the secondary and to provide a smooth surface on which to wind the secondary. The secondary is made with 280 turns (the exact number is not critical) of 26- gauge hook-up wire, The direc- tion is unimportant. You can use magnet wire if you desire but you should put cellophane tape be- tween each layer. The low-voltage end of the secondary is the one that is the closest to the primary winding. When the windings are complete, put the core halves through the bobbin and hold them in place with tape wrapped around them. ‘Transformer T3 is made with the same core and bobbin as Tl Both windings are bifilar with 18- gauge magnet wire for as many turns as possible. The start of both windings are polarized as indicated by a dot in the sche- matic diagram (Fig. 2). The pins on the bobbin are not used and should therefore be cut off, and the 18-gauge wires are then sol- dered directly to the PC board as indicated. An aluminum angle bracket is used when mounting switching- transistors Q3 and Q4. The bracket provides the physical support between the PC board and enclosure and also provides good heat sinking for the tran- sistors. The transistors should leet uBeWaLdasHADIO-ELECTRONICS, FIG 6—TRANSFORMER T1 IS MADE by winding colts L1 first windings, coils L3 and L4 are wound on top of Li. layer of cellophane tape on top oft ‘and L2 Coils LS and L6 are wound after the transformer is mounted on the board (b). See text for detailed instructions, be insulated from the aluminum insulating hardware is normally included when you purchase the transistors. Use the PC board as a template for drilling holes for the srans‘btors in the aluminum. bracket. The angle bracket is mp unted to the enclosure by dril- lung holes and taping them with a 6-32 tap Thermal conductive FIG 7—SEEN HERE IS THE DISCHARGE ‘rom the ball electrode into the compound is used between the transistors and angle bracket and between the angle bracket and the enclosure. A banana jack is mounted in the back of tlie enclosure to make connections between the Tesla secondary coil and the high-volt- age ferrite transformer. The out- put voltage from the ferrite transformer may reach 5000 volts peak with no load so it is wise to use extra insulation for the banana jack. Mounta piece of plastic, 1/-inch square. to the back of the enclosure over a 1- inch square hole, and mount the banana jack in the center of the plastic, That will space the ba- nana jack at least ‘-inch from the metal enclosure. The prototype used a 10-turn potentiometer for Rl to make fire- quency adjustments easier and this allowed the use of a 10-turn dial to mark the frequency set- tings for different purposes. You can use a regular potentiometer but the 10-turn unit is superior. ‘An enclosure was fabricated out of ¥eineh aluminum with a plexiglass top, but any metal en- closure would be suitable. Just be absolutely sure that you ground the metal enclosure. SS m 72 Pritahy 8 COLLECTOR Operation Warning: The power output from the Tesla coil is dangerous! Make sure no one comes in con- tact with the output voltage di- rectly from the driver. Make sure nobody tampers with the unit, and keep it out of reach of chil- dren. Make sure you use a key FIG. 8—THE SPARKS WILL JUMP even farther if a grounded electrode Is placed near the discharge ballswitch to turn power on and off to prevent someone from getting in- Jured, and keep the key in a safe place! Caution: All components on the secondary of TI are not iso- ted from the power line. Use aution when measuring values in this area. You must isolate an oscilloscope from ground if mea- suring in this area. Make sure you use a three-prong power cord and that the case of the driver is well grounded. Also, make sure you plug the unit into a well- ‘grounded electrical outlet. Double check all wiring to make sure it is correct. Make sure the operate switch is in the standby position (line voltage disconnected from D1). Using a digital voltmeter isolated from ground, measure the voltage across C3 and C4. Ifeverything is, working correctly in the low-volt- age circuitry, there should be about 2.5 volts across those pacitors. If that voltage is not present you should check the 21 volt power supply. Make sure that 5 volts is on pin 16 of ICI. If the oscillator is working correctly you should have about 3.6 volts on pin 6 of ICL Connect the Tesla secondary coil to the driver with a 3-foot insulated wire (it is a good idea to keep at least 3 feet from the sec- ondary coil). You should always unplug the driver when you are making connections between the driver and secondary coil to be absolutely safe. The wire con- jecting the coil and driver car- ies a dangerous amount of power so be certain the wire Is well insulated, Ina dimly lit room. you should be able to adjust the tune control to set the driver at the coil’s resonant frequency. Ob- serve the LED and watch for one place in the tuning control's ad- justment where the LED glows brighter than anywhere else. Never apply full power to the driv- er unless you can obtain re nanee first. Damage to the driver will most likely oceur if reso- nance is not maintained. ‘Once you obtain resonance you can switch to the full-power mode: the LED will glow very brightly. With no objects around the coil you should observe a snappy brush discharge 5 to 6 inches in length emanating from the discharge electrode (see Fig. 7). It might be somewhat louder than you would expect. Very slight adjustments in the tune control may improve the dis- charge. You should be able to get 7-inch streamers with a grounded electrode above the coil (See Fig. 8). Be aware that any change of the physical surround- Ings around the coil will change its resonant frequency and the tune control will need to be ad- justed to maintain resonance. When operating the Tesla coil, be aware of the temperature of the enclosure where the aluminum angle bracket is mounted. Shut off the power if the area gets too warm, The prototype was oper ated for 2 full minutes, and you could just start to feel some warmth on the enclosure. How- ever, you should operate the Tesla coil only for short periods of time. Once you have a working unit you can start to experiment with different things. Try removing the discharge ball and use a point instead. Try changing the dis- tance of the ball electrode from the coil. Try holding an incandes: cent lamp’a short distance from the coll—but be very careful. Dif- ferent lamps will produce dif- ferent discharges. RE r------ FCC NO-CODE I amateur ravioucense | ‘The FOC recently passed Docket 90: '55 whict forthe frst bme allows a naw ‘codeless entry ham radio license of technician grade. Privileges 30 MHz land above — All modes! (See PLE. ar ticle in April 1991 issue) Get all the no-code license details, Study & testing information plus a one. year subscription to one of ham radio's longest running specialty made publ: Cations that wll teach you all about the new modes you willbe able to operate! FSTY SSTV FAX TTY PACKET AMTOR OSCAR FM REPEATERS MICROWAVE AND LOTS MORE! ‘The SPEC-COM Journal P.O. Box 1002, Dubuque, IA 52004 (918) 587-8791 BS vie ‘CIRCLE 108 ON FREE INFORWATION CARD EARN YOUR B.S.E.E. DEGREE THROUGH HOME STUDY (ue New sa Hy nt Progra oe ‘ears gran rat or presous Sehoong snd fice ete rogues scam rogram ad (gues tor cused Clecrene Techneans ‘Tira his Speci Progfam yu an oul at {he ove en of yur eoctones background fogeier snd cam you B.S EE Degree. Up ines "Advarce Rasy! Many fash in 12 trons ores, Shes ad grates nt 50 Soto and hrougnout ha Ws. Estbishes Over 40ers ote or tee Daserpive Ut ole) SS one n bop Ci lasti 20 cron Too 1661 H3AWSLd3S CIRCLE 58 ON FREE INFORMATION CARD