I ripped apart a broken 1950s-era Variac autotransformer to diagnose an internal short in the core, and found a layer of gooey insulating fabric mat between the coil/taps and the mounting plate:
The short seems to have been caused by some sort of deterioration over time/hole in the fabric at the site of one of the taps. I'd like to replace the insulating mat, and I'd like to know what a suitable current material/product would be for this. The 1955 document describing the cross section of the Variac says:
The coil and base are in contact over a large annular surface, with a glass-cloth insulator between them so good thermal conductivity is maintained.
As seen in the photo above this fabric material is drenched in a dark goo that looks and feels like molasses but I'm sure is not molasses (I did not attempt to taste it). The goal seems to be electrical insulation AND thermal conductivity for heat sinking. There is plenty of "fiberglass cloth" stuff around but most of it is described as being heat resistant. I assume the gooey stuff was intended to help with this but I don't know what this is even called?
What can I use to replace this arrangement?
Best Answer
The current "gooeyness" of the goo may or may not be meaningful. Many materials degrade over time and especially with heat, and this is especially true of old materials, where the long term durability of then-new materials may not have been as well understood as it is today. So the current properties of the material are not necessarily the ones you want to find in a replacement material, and in fact the electrical/chemical properties may have changed in a way that reduces its effectiveness--or even renders it harmful, for example if it has decayed to release corrosive byproducts, or the electrical resistivity has significantly decreased. At the very least, the current gooeyness may have contributed to increased creepage of the mechanical assembly, resulting in the internal short you've observed.
It's a bit hard to tell from your picture, but it actually looks to me like that glass fiber used to be attached to the coil, and the degradation of the material has allowed it to separate from the coil when you disassembled it. This may have been an adverse interaction between the resin used to bond the fiber mat and the varnish on the coil wires, or some other chemical compound that leeched out of the core or something. This is all a guess, you'd probably need a chemist well versed in the common materials of the day to say anything more definite.
Glass fiber materials are still relatively common in high-temperature applications, particularly high temperature wire, where mica or PTFE insulation is often clad in woven glass fiber. However, there are also a number of other options that may be more suitable. If heat transfer is particularly important, there are various electrically insulating/thermally conducting materials designed to convey heat will providing electrical isolation. These range from polyamide films coated in ceramic thermal transfer goo, to thermal epoxies, to squishy gum materials and foams, in a range of thicknesses and with/without reinforcing fibers, etc.
The preferred material depends on several primary factors:
Some of these factors are at cross purposes, for example having a thinner interface improves thermal conductance but doesn't accommodate uneven surfaces.
In your specific case, you have pretty uneven surfaces due to the coil winding, and you also have some taps that provide bumps where mechanical force will be concentrated, decreasing the material thickness in those locations and risking puncture in the worse case. A compliant thermal interface material at a suitable thickness will probably provide superior thermal performance to the original material, and you could reinforce the locations of the taps with some tougher material for local puncture protection.
After cleaning up the old goo as best you can (be sure to take personal protective precautions, as mentioned elsewhere it could be quite nasty) you'll need to ensure that the tap locations contain no sharp points and are as low profile as you can reasonably achieve. A more significant repair could involve replacing those tap wires with copper tape to provide a lower profile, but that's not necessarily required. Some polyamide (aka Kapton) tape and/or glass fiber tape at the tap locations will help protect the taps from punching through the compliant material.
You can find sheets of compliant thermal materials at the usual electronic distributers, so I would probably start there. These can usually be cut to size with scissors. The manufacturers of those materials will have many more options in their catalogs, but these may not be readily available for purchase in the small quantity you need, hence it's easier to start with what you can actually find for sale. You frankly don't need something THAT good in terms of thermal conductivity, given the original glass fiber. You'll need to look at the surface variation you need to accommodate to determine the material thickness and 'squishability' you need--be sure to read the manufacturer datasheets and available appnotes to understand what you need. Some options will be "phase change" materials; these will soften at high temperatures, allowing them to better comply to the mating surfaces, but this means that a heat cycle is part of the installation process. It's preferable to have some spring tension in the assembly to continue compressing the material as it heats and flows--this is quite common on CPU coolers, where the heatsink attachment includes spring tension by design. If that can't be arranged you'll need to tighten the assembly, soak it at or a bit above its maximum operating temperature while the material settles in, and then tighten it again before it cools.