5
\$\begingroup\$

Can someone recommend best-practices for potting a passive DC high voltage (<20 kV) circuit? I'd like to eliminate a micro-discharge problem we are seeing.

Details of our use-case follow:

I have a a DC high voltage system (<20 kV) which passes though a low pass (RC) filter to eliminate ripple and other noise before passing into a vacuum chamber. I can monitor the AC component to see its effectiveness

I can get good results from the filtering component; however we potted our most recent build and are seeing micro-discharges: these appear as 1V+ dips in the output voltage. The number of these per unit time increase substantially above about 15 kV, until at 20 kV they appear every 20 ms or so.

I am wondering if there is any surface-prep we can do to the resistor/capacitor leads and or if someone can recommend a specific potting compound that would work better.

We did a thorough mix and vacuum degas the compound before pouring it into the enclosure. I didn't use a primer however. We used Krayden DC170 compound for the potting

\$\endgroup\$
3
\$\begingroup\$

Increase the diameter of the conductors. This will decrease the electric field strength at the surface. It sounds like you only need to halve the field strength, so only need to double the radius of curvature.

Where you use connecting, use thicker wire. Where that wire is a component lead, then you could wrap some wire or foil around it, or slip a metal tube over it. If you have soldered joints, make sure you have not left any sharp edges or points where you've cut wires. Use some extra solder and let it ball up under its own surface tension.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks for the guidance. I'm going to build another one and employ your feedback. I'll mark this answered if we get improved performance. \$\endgroup\$ Nov 12 '20 at 13:15
2
\$\begingroup\$

Another thing you need to look out for is air bubbles when you apply the potting, if any air bubbles are present it will drastically reduce the dielectric resistance of he arrangement as a whole.

since this is very important to know the effectiveness of your current setup you need a diagram showing the distances between ground and high voltage, this way other people can give better advice, seeing the datasheet for the compound you mentioned it has 14kV per mm so if you have 2mm of clearance it should be ok.

However, if there are pin to plate geometries(you got a sharp pin with high voltage facing a ground plate) you should derate this dielectric strength by about 30%... again without looking at the geometry and the distances it is hard to give concrete advice. Try to make your pins less sharp and more like spheres.

\$\endgroup\$
1
  • \$\begingroup\$ To remove bubbles, assemblies can be placed into a vacuum chamber while the potting compound cures. \$\endgroup\$
    – rdtsc
    Nov 17 '20 at 13:37
2
\$\begingroup\$

In our second build we had much better performance.

  • clean all surfaces
  • smooth points and edges in BOTH the ground planes and on high voltage parts (see below)
  • used primer
  • vacuum degas of potting compound

One thing we did was expended some effort make smooth all of the ground plane surfaces, since arcs begin at the negative terminal, and we are @ +20 kV. The ground plane smoothness might be just as important as the HV terminals.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.