[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Potting compound suggestions



Original poster: "Jim Lux by way of Terry Fritz <teslalist-at-qwest-dot-net>" <jimlux-at-earthlink-dot-net>


>        I'm looking for some suggestions for relatively cheap potting 
> compounds for a couple circuits I'm putting together. I'm looking to 
> minimize corona loss and sparkover while, at the same time, increasing 
> the safety factor (insulation) for both circuits.

A worthy goal...

You've got some tradeoffs to assess.. reworkability and messiness..
If you really think you're going to be assembling and disassembling 
multiple times (i.e. the repair of a blown diode idea), then you should 
consider immersing in a dielectric rather than potting.  Potting is 
permanent, and rework of potted stuff is a last gasp measure.  Mineral oil 
is probably the simplest immersion dielectric: cheap, fairly easy to clean 
off (you let the assembly drip dry).  Pressurized gases (nitrogen, CO2) are 
also easy, if your stuff will fit in an appropriate container  (i.e. a 
piece of pipe).  In both cases, sealing is the challenge.

For actual potting, bear in mind that it's not all that hot at truly 
suppressing corona, unless you do it under a vacuum, and carefully lay 
things out so you don't get voids when the potting material is poured in.

Indeed, some forms of RTV are not suitable for electronics.  There are 
electronics suitable RTVs available, check the web sites.  (Insulcast, 
Electro-Lite, Polycast, GE, etc.)

You also need to consider thermal properties.  There are some potting 
materials that have relatively high thermal conductivities, and others that 
make fine thermal insulators.

Potting in wax is a method that has been historically used, but there is 
significant shrinkage as it cools, so voids are a real problem.

If all you want to do is suppress corona, then "corona dope" type 
approaches and good construction practices might be 
sufficient.  Exceedingly high voltage air insulated CW  multiplier stacks 
can be made with very low corona loss by careful arrangement and field 
control (i.e. you put a big electrode on the top to establish the over all 
field down the stack.  In the case you cite of building a voltage divider, 
you'd need to have good field control anyway, so air might work just as 
well as anything else.



>  Let me describe each circuit to give you an idea what I'm dealing with.
>
>         The first circuit is a six-stage Cockcroft-Walton voltage 
> multiplier circuit that takes 15 kV square wave input at 20 mA with a 
> frequency of around 20 kHz from a solid-state neon sign transformer. I 
> may scale this circuit back to have it output around 60 kV instead of 
> around 90 kV.
>
>         I know as the number of stages in a multiplier increases, the 
> actual output voltage can change significantly according to the load. 
> Therefore, I'd like to be able to measure the output voltage using a 
> combination voltage divider and digital multimeter (with about 10 Meg 
> internal resistance). I've got some 200 Meg and 10 Meg  2% high voltage 
> resistors I want to stack to form the divider, but again I want to 
> minimize corona loss and keep the exposed high voltage to a minimum.
>
>         I want to pot each circuit separately and I'm trying to balance 
> off several factors. Number one is cost, I want to keep things as cheap 
> as possible. I realize that submersing the circuits in mineral oil is one 
> relatively cheap way to go, but it's also pretty messy and I'd like to 
> steer clear of that method unless I have no other choice. Second, from 
> what I've read, some epoxy-based potting compounds generate significant 
> amounts of heat during curing. Also, I've heard some types of RTV release 
> corrosive chemicals while curing. So I'm looking for something that won't 
> melt the capacitors or digest the diodes in my multiplier. Third is 
> reversibility ( is that a word?). If I blow a diode in the multiplier, it 
> would be nice to be able to un-pot the circuit to replace the failed 
> component(s). By unpot, I mean anything from melt under low heat (if 
> potting in paraffin) to gently chip off pieces of the potting compound. 
> However, being able to unpot the circuit is the lowest priority.
>
>         Does anyone know of any potting compounds that meet one or more 
> of these requirements?
>
>
>         Thanks in advance,
>
>         Rick Richter