Hi all, I have a HP 5061A with a working physics package and run it in Cs mode every now and then. Recently it wouldn't start and the problem was that the +3500 volt A18 unit was not producing its "power good" signal from pin 4. I therefore had to fix it. The +3500 ( ion pump supply A18) and -2500 (electron multiplier supply A19) supplies are similar modules and are housed in sealed cans that are soldered shut. There is a soldered-on cap over a filler hole, possibly to fill the unit with inert gas. The circuits consist of a blocking oscillator running at about 1 KHz feeding a small power transistor that pulses the primary of a tuned transformer running at about 20 KHz. The secondary voltage is fed to a half wave doubler of appropriate polarity, the final capacitor of which is 0.01 uF. There is a voltage divider/bleeder consisting of a 200 Meg HV resistor and lower value resistors. The supply is free running with no voltage feedback. The bleeder time constant is 2 seconds so it should usually be safe to work on the HV section after about 10 seconds. The whole thing works from the +18.7 volt DC supply in the 5061A. To get to the circuitry I gently heated the can with a small gas torch until the solder melted and I was able to pull out the innards. I set it up on the bench and applied 18.7 volts but no output was present on pin 4. It soon became obvious that the problem was R4, a faulty 200 Meg resistor which was open. Replacing it cured the problem. Fortunately 200 Meg HV resistors are not too hard to find on the usual auction site. There are a couple of gotchas I need to mention: 1. If there is no output on pin 4 don't assume the oscillator section is not working. An open R4 will prevent any voltage reaching pin 4 and will also prevent any bleeder action. That 0.01 uF cap will hold the HV for a long time. Don't ask me how I know this! 2. The HV output and the oscillator 18.7 volt supply are both referenced to the chassis ground but connected separately. When testing the unit out of the 5061A you need to connect the negative 18.7 volt terminal (pin 2) to the chassis. I'm not sure why HP did this and the schematic was a bit confusing to me with different ground symbols in different parts of the circuit. The thing will run if you don't connect it this way but the voltage readings don't make sense. I no longer have access to a source of dry nitrogen but I don't think the lack of it will make much difference. Finally I have a question. My A18 unit produces a bit more than +3500 volts. It's about +3850. Is this a problem and will it harm the Cs tube? Regards to all, Morris