High-Voltage Amplifier Circuits by Dennis L Feucht Some linear amplifiers require an output voltage range that is large relative to the rest of the system. Some applications are: Power pulse amplifiers, motor drivers, piezoelectric transducer drivers, oscilloscope deflection-plate drivers, on-chip programming supplies, analog computers, EL panel and display supplies, and audio power amplifiers. Whether it is 12 V or 1.2 kV, special considerations arise when transistor breakdown voltage becomes a …

Some linear amplifiers require an output voltage range that is large relative to the rest of the system. Some applications are: Power pulse amplifiers, motor drivers, piezoelectric transducer drivers, oscilloscope deflection-plate drivers, on-chip programming supplies, analog computers, EL panel and display supplies, and audio power amplifiers. Whether it is 12 V or 1.2 kV, special considerations arise when transistor breakdown voltage becomes a critical design parameter. This article presents circuit alternatives for high-voltage (HV) amplifiers, and their relative merits. Translator Stage Usually near the amplifier input is a stage of amplification that performs the function of voltage translation — of accommodating HV at its output node. Two possible circuits are the common-emitter (CE) and common-base (CB) transistor configurations. (Although BJTs will be used here, FETs can also be applied, taking into account relevant differences between them.) The CE translator is shown below in simplified form. If a negative supply is available, the emitter resistor ( R E ) can be returned to it, allowing the input voltage at the base to extend to ground. Without R E , the collector voltage of Q1 can extend down to near-ground, to the onset of collector saturation. With even a small ( >33 ?) R E , the input circuit is more linear and less loop gain is required to stabilize it. With any sizeable value of R E , the input resistance is relatively high, which is required for a voltage-input amplifier. Ideally, a negative supply of -5 V to -12 V is much greater than the 0.7 V of the base-emitter junction, and allows stable biasing of the transistor at a given current. An alternative to the CE is the CB, shown above. Its input resistance is R E + r e , where r e is the dynamic emitter resistance of Q1. The base is returned to a convenient supply voltage, which can also be synthesized with a resistive divider. (Then, the equivalent base resistance, R B , adds to the input resistance as R B /(? + 1). The CB alternative is inherently faster than the CE because the Miller effect is eliminated. However, the collector of Q1 is limited in range on the bottom end by the base voltage, and cannot swing RE Q1 NPN +5V RE Q1 NPN
down to near-ground. The lower the base voltage is made, the analogous CE situation with a grounded R E is approached. The base voltage needs to be large enough to allow R E to dominate input resistance, for both bias and incremental gain stability. Because transistor breakdown voltage is an issue, the CB has the clear advantage over the CE. The lower the base resistance, the higher the voltage that can be sustained at the collector. (A CB also has a higher dynamic collector resistance, allowing larger load resistors without sacrificing gain.) Next Stage What goes up must come down — sometimes. Unless the HV amplifier is like an oscilloscope deflection amplifier, for which the output is usually at an elevated dc value, the high voltage that the translator stage can accommodate must also allow for low voltage at the low end of the output-voltage range. Assume that the design needs to accommodate at least ground in the output range, if not lower. Two circuit alternatives get back down to ground. The common-base, shown below, forms a complementary cascode stage with a CE translator. R C is kept small, to maximize the upper end of the output voltage range at the collector. R B , diode D1 and the current source provide a base voltage that….

Download High-Voltage Amplifier Circuits.Pdf