Test Setup.

To measure the output voltage of a power supply, refer to the test setup shown in Figure 2. A resistive divider is used to attenuate the high voltage by a known amount to a level where it can be measured conveniently by a standard DC voltmeter. RT represents the end-to-end resistance of the divider. It is also the load presented to the supply. If measurements are to be made at no-load conditions, RT should be at least 10 times the normal load resistance value. R1 represents the resistor across which the measured voltage is developed. It is in the low voltage return path for safety purposes. The ratio of R1 to RT is the amount of attenuation. If that ratio is made 1/1000, a voltmeter will read directly in kV.

Designing a high voltage attenuator is a straightforward task.  However, it may not be an easy task because of the high voltages involved. RT must be constructed from a large number of smaller resistors.  These resistors are connected in series with one another to reduce the voltage drop across each individual resistor to a level it can safely handle, both from a voltage coefficient and dissipation point of view. Resistors must be spaced far enough apart, if air forms the insulation medium, or encapsulated, or immersed in dielectric oil, to avoid arcing. R1 must be small in value so that the input resistance of the voltmeter does not load it. All resistors must have low, or matched, temperature and voltage coefficients to avoid changes in attenuation with temperature or attenuation that drifts with changing voltage levels.

To simplify this task, Glassman has available three standard high voltage attenuators; RD 50A covers voltages to 50 kV, RD 100A to 100 kV and RD 200A to 200 kV. All provide an attenuation ratio of 1/10000 with low temperature and voltage coefficients. These attenuators provide an additional circuit with a DC blocking capacitor which allows measurement of the ripple voltage using an AC RMS voltmeter or oscilloscope.

Test Method.

Once a suitable attenuator is available, measurements of output voltage are simple and straightforward. If the supply has a metered output, or a remote monitoring terminals, those indications can be compared to the measured results. For measurements of output voltage at other than no-load, refer to Figure 3. The added load resistor is constructed and connected exactly as described above under "Changing Load."

Figure 2

[ Contents | Page 1 | Page 2 | Page 3 ]
[ Page 4 | Page 5 | Page 6 | Page 7 | Page 8 ]