 Although the trend in modern electronics is toward lower- power circuitry, there is simply no getting around the fact that many of the latest electronic gadgets still require a "spritz" or two of high voltage to make them function properly. Unfortunately, many high-voltage circuits depend on relatively expensive and bulky step-up transformers to generate the "juice" that they require. That's because those circuits also have fairly hefty current requirements, as well as a "thirst" for good regulation.

Other circuits, in which extensive regulation and "vast" amounts of current are not required, rely heavily on high-voltage generating configurations that can be built around relatively inexpensive and readily available components.

In order to obtain the high voltages needed for the less demanding circuit configurations, a voltage-doubler is often used. Voltage doublers, which are sometimes used in radio-frequency-actuated circuits to obtain the control voltage, allow you to generate higher voltages than would otherwise not be possible with conventional power supplies. Voltage doublers are not generally used when a high degree of regulation is required or when the current drain is high. ## Voltage-Doubler Circuits.

As in conventional power-supply circuits, there are two basic voltage-doubler configurationshalf-wave and full-wave.

Figure 1 is an example of the half-wave voltage doubler (also referred to as a cascade voltage doubler), while Fig. 3 illustrates the full-wave version (also referred to as a conventional doubler). In both circuits, the direct-current (DC) output voltage is twice the peak alternating-current (AC) input voltage; or 2.8 times the root-mean-square of the AC input voltage.

That means if the circuit is fed from a 12.6-volt AC transformer, the DC output voltage VDC(0ut) would be:

The conventional doubler (Fig. 3) provides superior voltage regulation and less output ripple, but the cascade circuit (Fig. 1) can be used without a transformer. In addition, two or more cascade circuits can be connected in series to form voltage multiplier circuits with various multiplication factors.

The conventional doubler (Fig. 3) provides superior voltage regulation and less output ripple, but the cascade circuit (Fig. 1) can be used without a transformer. In addition, two or more cascade circuits can be connected in series to form voltage multiplier circuits with various multiplication factors.

## Half-Wave Doubler.

Refer to the half-wave voltage-doubler circuit shown in Fig. 1A, and assume thai Cl and C2 are both initially dis-charged. During the first half-cycle of the AC input, the upper input terminal of Ti's primary winding is positive with respect to the lower terminal (as illustrated in Fig. 1A), causing an oppositely polarized voltage to be induced in Ti's secondary winding. Under that condition, D1 begins to conduct, causing Cl to charge. At the same time, diode D2 is reverse biased, preventing its conduction, so C2 discharges through RL. The analysis is similar in the second half-cycle, except (as illustrated in Fig. 1B) that D2 conducts and C2 charges, while D1 is cut off and Cl discharges into RL.

The circuit is really a transformer- less voltage amplifier. While T1 can provide isolation, as well as increase the AC voltage initially going into the doubler, the amplification due to the doubling action would occur without it. When the polarity reverses, both the input voltage and the charge across Cl behave like two batteries connected in series, with their voltages combining to produce a DC output of about 36 volts peak. One problem, though, is that a half-wave doubler cannot be used with a current-hungry load.

## Full-Wave Doubler.

One way of increasing the circuits current capacity is to use a full wave doubler. A full-wave voltage doubler, unlike the half-wave version, is designed to take advantage of both positive and negative half-cycles of the input AC voltage. For the sake of greater clarity, the full-wave voltage doubler is shown redrawn in Fig. 3. The full-wave voltage-doubler circuit has better regulation than the half-wave version, is easier to filter, and produces nearly double the peak AC voltage (approximately 36 volts for the previous example) across RL. During the first half-cycle (see Fig. 3A), D2 is reverse biased and therefore cut off, while D1 is forward biased into conduction, so that the voltage across Cl (V,1) is approximately 17.766 volts DC. On the next half-cycle (see Fig. 3B), the polarization of the applied voltage is reversed, forward biasing  D2 into conduction, while reverse biasing D1 into cutoff. The load resistor (RL) is wired in parallel with the C1 /C2 series combination effectively creating a doubled level of about 36 volts DC.

Unlike the half-wave voltage doubler, the full-wave version has two capacitors across RL rather than one. Whereas Cl shown in Fig. 1 is cut off and unsupplied for half of every cycle, Cl and C2 in Fig. 3 are supplied on alternate half cycles. When the capacitor corresponding to the diode that's cut off discharges, it can only do so through the capacitor being supplied, slightly decreasing both its current and the maximum voltage it has reached. ## Voltage-Multiplication Circuits.

There are many variations of the voltage-doubler scheme. Figure 4 illustrates a voltage-multiplication configuration based on the circuit in Fig. 3 that can be used to generate a DC output voltage three times that of the AC input to the circuit. That circuit, a voltage tripler, operates in essentially the same manner as the doubler circuit of Fig. 3. The voltage quadrupler on the left is another example wiring diagram and the voltage is 4X the input.

By now a pattern should be beginning to emerge. Note the correlation between each circuit's voltage-multiplication factor and the number of diodes and capacitors in each circuit. Figures 6-8 show a few additional voltage-multiplication circuits. The voltage multipliers shown in Fig. 6 are the most straightforward. Note that the circuit in Fig. 6A is electrically identical to the one in Fig. 6B.

So keep that in mind it you should come across either format. The usefulness of the Fig. 6 circuits can be enhanced by adding voltage taps at each of the diode junctions.

For example, referring to Fig. 6B, voltage taps can be added at the D1 /D2, D2/D3, D3/D4, D4/D5, and D5/D6 junctions, for multiplication factors of x1, x2, x3, x4, and x5. Plus, another tap can be connected to the anode of D6 for a multiplication factor of x6, or VDc(Out)= 6(1.41)VAC (RMS INPUT)

Thus the circuit is able to provide six levels of DC voltage. Additional stages can be added to the circuit to generate multiplication factors of x10 or more.

Note, however, that as the voltage multiplication factor increases, the available current that can be drawn from the circuit decreases by a similar factor. For example, feeding a 12.6-volt, 1-amp AC source through a voltage doubler yields a DC output of approximately 36 volts at about 0.5-amps. Figure 7 shows an enhanced version of the Fig. 6 circuitknown as either a Cockcroft-Walton or Greinacher cascaded voltage doubler that offers better stabilization for moderate-current applications. A sewing needle can be used as an emitter for the voltage doubler shown in Fig. 8 to generate "corona wind," which sounds like a hissing noise. The circuit is capable of delivering 3.75 kV (kilovolts) DC when powered from 117-volt AC source, or 7.5 kV DC when fed from 240 volts AC.

The output of a cascaded voltage doubler should be terminated with no less than 200 megohms, and only then be allowed to extend beyond a protective plastic case, for safety. Voltages as high as 5 megavolts DC have been generated using cascaded voltage doublers, especially when operating in a pressurized atmosphere. The biggest advantage to using voltage doublers is that they use inexpensive low-voltage parts. Otherwise, if all the parts had to be of the high-voltage variety, you would have to use expensive and rather large capacitors. ## High-Voltage DC Generator.

A schematic diagram of a high-voltage DC generator is shown in Fig. 9. The circuit is built around a single hex inverting Schmitt trigger (101), a couple of transformers (T1 and T2), a transistor (Q1), 21 diodes, and several support components.

At the heart of the circuit is the hex schmitt trigger. One gate of the hex Schmitt trigger (IC1 is configured as a square-wave pulse generator. The output of IC1-a (a pulsating DC voltage) at pin 2 is fed to the inputs of ICI-b to ICI-f, which are connected in parallel to increase the available drive current. The pulsating output of the paralleled gates is fed to the base of Qi through R2, causing Q1 to toggle on and off in accordance with the oscillations of IC1-a. The collector of Q1 is connected in series with the primary winding of T1. The other end of T1 is connected directly to the positive terminal of the power supply. That produces a driving wave in the primary winding of T1 that is similar to a square wave.

The on/off action of Ql, caused by the pulsating signal applied to it, creates a rising and collapsing field in the primary winding of T1 (a small ferrite-core, step-up transformer). That causes a pulsating signal, of opposite polarity, to be induced in T1's secondary winding.

The pulsating DC output at the secondary winding of T1 (ranging from 800 to 1000 volts) is applied to a 10 stage voltage-multiplier circuit, consisting of D1 through D20, and C3 through C12. The multiplier circuit increases the voltage 10 times, producing an output of up to 10,000-volts DC-VDC(OUT)= 10(1.41)VAC(RMS INPUT)

The multiplier accomplishes its task by charging the capacitors (C3-C12) through the diodes (D1-D20); the output is a series addition of all the capacitors in the multiplier.

In order for the circuit to operate efficiently, the frequency of the squarewave, and therefore the signal applied to the multiplier, must be considered. The output frequency of the oscillator (IC1-a) is set via the combined values of R1, R5, and Cl (which with the values specified is approximately 15 kHz). Potentiometer R5 is used to fine tune the output frequency of the oscillator. The higher the frequency of the oscillator, the lower the capacitive reactance in the multiplier.

Light-emitting diode LED1 serves as an input-power indicator, while NE1 indicates an output at the secondary of Ti . A good way to get the maximum output of the multiplier is to connect an oscilloscope to its high-voltage output via a high- voltage probe and adjust potentiometer R5 for the maximum output. If you don't have the appropriate test gear, you can place the output wire of the multiplier about a half-inch away from a ground wire and draw a spark, while adjusting R5 for a maximum spark output.

## Caution:

The output of the multiplier can cause a strong electric shock. In addition, be aware that even after the multiplier has been turned off, there is still a charge stored in the capacitors, which, depending on the state of discharge, can be dangerous if contacted. That charge can be bled off by shorting the output of the circuit to ground. (In fact, it's a good idea to get in the habit of discharging all electronic circuits before handling or working on them.)

Also, 1C1 is a CMOS device and, as such, is static sensitive. It can handle a maximum input of 15 volts DC. Do not go beyond the 15-volt DC limit or the IC will "vaporize." Diode D21 is used to prevent reverse polarization of the input voltage source.

As far as the voltage multiplier goes, the diodes and the capacitors must be rated for at least twice the anticipated input voltage, So, if we have a 1000-volt input, all of the diodes and the capacitors must be, respectively, rated for at least 2000-PIV and 2000-WVDC (working volts DC) each. Because diodes with that voltage rating can be hard to find and expensive (if you can find them), pairs of series-connected 1-amp, 1000-PIV diodes were used to form 2000-Ply units.

Revised 2013 by Larry Gentleman