Jim, your comment on water vapor and desiccants reminds me of a future
project I have in mind to build an oil-insulated transformer. Iâ??m still
trying to work out how I can build an enclosure that will prevent water
accumulation in the oil.
What I should do is periodically degas the oil with a vacuum pump, but Iâ??m
a poor college student.
On Fri, May 20, 2022 at 2:30 PM Lux, Jim <jim@xxxxxxxxxxxxx> wrote:
> On 5/20/22 11:08 AM, pupman.com@xxxxxxxxxxxxxxxxxxx wrote:
> > Hi Tedd,
> >
> > Cardboard tube works fine as a coilform by itself. As long as you bake
> > the moisture out, then seal it you're all set. Clear polyurethane
> > varnish from the hardware store works fine and requires no heat curing
> > or special processing to use. The amount of water paper can retain is
> > rather surprising. There was a capacitor factory here in Chicago that
> > you could smell from down the block. It was an odor similar to cheap
> > paper towels or wet paper bags and was just moisture being drawn from
> > the capacitor elements in vacuum chambers. Nothing went in soggy- it
> > was just moisture from the air they were removing from kraft paper and
> > cardboard insulation. Paper dielectrics were dried for days and there
> > was even one company that dried sections for something like 2 weeks
> > (nobody knows why) before impregnation. The cans would be soldered
> > shut under oil- that is they actually put large soldering irons into
> > the oil tanks and soldered under oil.
>
> I can guess why the two weeks - some "approved procedure" for a large
> customer (i.e. US Government, Phone company, or similar) said "Capacitor
> sections shall be baked out at X degrees and no more than Y inHg for no
> less than Z hours".
>
> We do that for spaceflight stuff, because gunk outgasses and then
> deposits elsewhere contaminating either the vacuum chamber or your
> optical equipment. Silicones are particularly troublesome - they're
> really sticky. But we do it for almost everything - coax cables, for
> instance, especially anything that might have "wicked" solvent or
> something. A bakeout of several days at 60C at <1E-4 Torr would be
> typical.
>
> As you note, paper is remarkably absorbent. And water vapor is quite
> "sticky" - if you're doing HV work, it doesn't take much water to
> greatly reduce the insulating capacity of the oil - in a big
> transformer, they'll have dessicant in the system, and it runs warm,
> which helps boil out the water (because sometimes, you can't "seal" the
> system - the changes in air pressure would make the enclosure buckle).
>
>
> >
> > Transformer manufacturers can bake parts at higher temperatures and
> > bake stuff for less than a day, even if it contains paper. Final
> > curing takes place in an oven again. These temperatues exceed what
> > most plastic parts or any dielectric in a capacitor can withstand.
> Exactly - baking out polyethylene at 70C would probably be a mistake.
> >
> > As for polyester resin as a choice, it sounds excellent. They even use
> > it in reconstituted mica capacitors, which are still by far the most
> > rugged, highest performing high voltage capacitors available.
> >
> You might look back in the archives - I believe Electrum, built by Greg
> Leyh, used form created by wrapping a cardboard tube with
> fiberglass/resin composite. Then pressure washed the cardboard out. I
> can't remember if the windings were embedded in the composite, too.
>
> https://en.wikipedia.org/wiki/Electrum_(sculpture)
>
> http://www.lod.org/gallery/electrum/electrum.html - If you go to the
> pictures page you can see the secondary being wound.
>
> I think Greg posted some details on how he built it on the list.
>
>
> >
> >
> > On Thu, 19 May 2022, Tedd Dillard wrote:
> >
> >> Is there any reason besides difficulty to use a cardboard tube or other
> >> material as an arbor and use fiberglass fabric and epoxy Resin to make a
> >> form?
> _______________________________________________
> Tesla mailing list -- tcml@xxxxxxxxxxxxxxx
> To unsubscribe send an email to tcml-leave@xxxxxxxxxxxxxxx
>
--
Joshua Thomas
My new email address is: joshuafthomas@xxxxxxxxx
Please update your information if you have not already done so.