With all this talk about HV stuff, I think it's a good time to introduce my HV curve-tracer/test box thing. I have been wrapping up the final design and features recently, and it's looking real good, with high utility for all sorts of HV device testing. It's basically a low power, variable HV engine, using a 5 kV RMS "static-neutralizing" transformer, with various operating modes. A built in HV probe circuit, and special receivers for DUT return current and DC leakage, provide output signals to a scope and a DVM.
The HV engine has these main functions, selected by a SP8T HV switch:
The AC is variable from ~0 to +/- 7500 V peak
The DC ones are variable from ~0 to 15 kV peak
The maximum steady-state current from any output is around 2 mA (shorted).
The HV probe drives a 1 meg scope input, with 1000:1 ratio, so 1V/kV.
The DUT return receiver absorbs the cold-end current, and drives another 1 meg scope input, with various ranges:
1V/mA, 2 kohm burden
1V/uA, 1 megohm parallel 200 pF burden
LOG (bipolar compression), about 4 mA/400 mV reference level, 1 kohm plus diode parallel 1.5 meg burden
The DC leakage receiver drives a DVM, and has internally selected termination Rs to accommodate DVMs with input R of 1 meg, 10 meg, or 1E9 and up.
It is especially useful for capacitor leakage testing, with low-pass filtering to reduce ripple current and interference. The ranges are:
1 V/mA, tau 1 sec, 2 kohm burden
1 V/uA, tau 1 sec, 1.1 meg burden
1V/uA, tau 11 sec, 1.1 meg burden
The receivers and other ports are protected from faults with spark gaps and other items. No equipment output signal can exceed 30 V.
My test setup with a 5 digit DVM, can resolve to 10 pA - but only after tens of minutes to allow settling of the dielectric absorption, and with maximum filtering, and with line voltage fairly stable (the HV is not regulated). This was on a .01 uf 5 kV ceramic cap - the rectangular type often found in old scopes - stressed at about 6500 V. I'm finding these caps are quite remarkable in performance, despite their age and being used. This particular unit settled at less than 100 pA with 6500 V, so well above E13 ohms, which I'd expect just from a good insulator, without any capacitance included. Of course, this is only at room temperature, with no other stresses applied, but it was quite surprising. I'll be reporting on the caps and other items separately - it's been a lot of fun checking things out, with interesting results.
There are also some other items and functions, jumper-selected on the working deck. These include 4600 pF of HV filter capacitance, clamp circuits to make +/- 2 kV and +/- 200 V, and a high R (160 meg) for DC leakage feed and other very low current uses. The 2 kV clamp makes a 4 kV pp trapezoidal wave, for amplitude calibration of the HV probe/scope combo, and the 200 V one makes a 400 V pp - at one-tenth the risetime - "square" wave, to adjust the probe and cable compensation.
Anyway, it is moving along and working nicely, so I'll soon be able to put out some info on it. I'll report more as I finalize and draft up documents from my notes. I think some will find the system and design concepts interesting and useful.
Now, this whole deal may be considered OT here, since it isn't about Tek scopes per se, but I think the principles are very useful for checking out many of the problematic HV type components used in our scopes. So Dennis, if you don't want this here, please speak up ASAP, and I will refrain from further discussion.