Rohde & Schwarz UHP BN1950 High Voltage Tester Restoration.

[thejazzcat]

Hi,

I'm trying to restore this old unit from R&S to test the insulation of an output power transformer.

First thing I noticed was a low voltage at the multiplier output, about 900 VDC. I checked the capacitors in this section and found that most were leaking at low voltage and their capacity has dropped to around 0.33 µF. I then replaced these capacitors (the next ones I could find were 0.33 µF at 630 V) and the 4 selenium rectifiers for 1N5408 diodes. Now I get about 2.25 K VDC at the output of the multiplier. One thing solved.

The device measures the resistance in the 50 V range correctly, almost exactly! But when I start increasing the voltage range, the resistance begins to decrease.

I have recorded some measured values that I get in megaohm with different resistance values. R(M) = resistance to be tested in megaohms. You can find them in the attached pictures. The second table was recorded to document something that happens when I try to measure the output voltage at the + and - connections. The first column corresponds to the measured value that I receive on a Simpson 467 DMM (which offers the BN1950 an input impedance of 10M), the second with the same DMM and an additional 5 kV probe (input impedance of 100 M). With the probe I can check all the voltages up to 2 KVDC.

I have been working in guitar tube amplifiers for a while, but my electronics experience is limited. This circuit seems so simple, but I don't know exactly what to do to fix this behavior or to understand the reason for the lower readings at higher voltages. 

All resistors measured to specs, the multiplier ones have drifted in vale from 10M to 10.5 (I guess no problem here). The ones for the voltaje selector measure within specs. The meter was tested getting full deflection at a 100uA, consistent with the specs.

I attached the corresponding schematic for reference.

If anyone can give me some suggestions, that would be truly appreciated.

All the best.

Carlos