Re: Polystyrene caps: was Re: Polyester Film Capacitors

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Dr. Duncan Cadd by way of Terry Fritz <twftesla-at-uswest-dot-net>" <dunckx-at-freeuk-dot-com>

Hi Terry, Jim, Ed, Bart, John, Ted, All!

Thanks for all the excellent feedback on polystyrene caps!  I've made
a composite edit of your various postings and commented in between.


>Original poster: "Terry Fritz" <twftesla-at-uswest-dot-net>
>
>Oops!  The dissipation is 0.0007 not 0.07 so it only runs 7 times
hotter :-)

Yes, this is what prompted my thoughts, and as Ed says (further down)
it may even be better than this.


>Original poster: "Jim Lux by way of Terry Fritz
<twftesla-at-uswest-dot-net>" ><jimlux-at-earthlink-dot-net>
>
>Fruengel talks a bit about polystyrene pulse caps..  I suspect that
for
>modern manufacturing, polypropylene is preferred for some reason.
Maybe PP
>has better solvent resistance when cleaning PC boards with caps on
them than
>PS... PS seems to dissolve in just about anything hydrocarbon based.
Maybe
>it's cheaper, or easier to metallize?

You make (as ever) some good points.  I know (having worked with
polystyrene copolymers for a few years) that pure PS is very brittle,
so film winding may prefer PP for this reason.  Your comments re
solubility also make sense.  Perhaps this has more to do with polymer
chemistry than was initially apparent to me!  Shame!  Back to the
chemistry textbooks Dr. Cadd!

>Original poster: "Ed Phillips by way of Terry Fritz
<twftesla-at-uswest-dot-net>" ><evp-at-pacbell-dot-net>
>
>Polystyrene is an excellent capacitor dielectric, with very low loss
>tangent and very low dielectric absorption.  Used to use them for
analog
>differentiators and integrators, long, long, long ago.

You were into analogue computers?  Not the Nike missile program in the
50s? My, but you've seen some interesting things in your time!  Please
tell me more off list!

>Problem is that
>it also has a relatively low dielectric constant

Not significantly less than PP I think . . .

>and softening temperature.

Now there you may have hit it.  The glass transition temperature Tg is
around 90-100C if memory serves.  There will be some interesting
dielectric changes going on at Tg - one of the classic ways of
measuring Tg in fact!  Now why didn't I think of that?!  It's _my_
subject . . . dohhhhhhhhhh ;-)  I recall reading that PS caps are
limited to applications where the temperature rise is expected to get
to less than 70C.  I wonder how high PP goes?

>There are lots of 200 V polystyrend capacitors up to at
>least 0.01 mfd.  The ones I have are of the extended foil type, ideal
>for high current.  Unfortunately, the leads are about #28.  Doubt if
>anyone is going to the trouble of making really big ones when there
are
>dielectrics which work OK and are a lot easier to handle.

I do actually have a couple of 1000V extended foil PS caps, but only
1nF and the same lead size.  No maker's mark though, which is a pity.
Got 'em at a hambeurs in Antwerpen, Belgium.

>Where did you get those numbers, Terry?????  According to my
"Reference
>Data for Radio Engineers" the dissipation factor of polystyrene at 1
MHz
>is "<0.0001".  Only goes up to 0.00033 at 300 MHz!  Don't have
>polyprolylene in this handbook, but believe it is similar in
properties
>to polyethylene.  Looked up polypropylene capacitor characteristics
and
>find only that the dissipation factor at 1 KHz is "<0.1%", or much
>poorer than polystyrene.  Looked at a number of different
manufacturers
>and all list the same number.

This is interesting.  I've dug out my "Handbook of Chemistry &
Physics" 42nd ed. 1960-61 and it lists PS at K=2,4-2,65, power factor
1Mc/s 0,0001-0,0004, no data for PP but polyethylene which may be
similar it gives K=2,3 and power factor <0,0005.  Only PTFE is in the
same league and its K=2.  (Also very expensive!)  "Reference Data for
Radio Engineers" Federal Telephone and Radio Company 3rd ed 1955 gives
PS at 1Mc/s dissipation factor 0,00007 K=2,56; PE it gives K=2,26,
dissipation factor <0,0002.  On paper at least PP and PS should be
comparable, if not the balance being a bit in favour of PS.  I'm
beginning to think that the reason we don't see more beefy PS caps is
for the reasons Jim and Ed cite.  But wait . . .

>Original poster: "Barton B. Anderson by way of Terry Fritz
<twftesla-at-uswest-dot-net>" ><tesla123-at-pacbell-dot-net>
>
>Hi Terry,
>
>I run CSI 60kv polystyrene pulse caps and have yet to warm them
>in pig usage. I don't run at long lengths, but they've seen lots of
usage
>over the past 3 years.
>
>Dimensions are 3" x 17" with large 3/8" bolt/ 5/8" nuts (the good
stuff)
>at each end. When first picked up, I've always been curious to see
>these get warm, but they don't. I check for this during all runs.

Wow!  So somebody _does_ still make these things :-)  Who are CSI?  If
they have a web presence maybe they also have data sheets . . .

>Original poster: "tesla by way of Terry Fritz <twftesla-at-uswest-dot-net>"
><tesla-at-paradise-dot-net.nz>
>
>Team
>I've sucessfully made small polystyene rolled caps (for the PSU
>decoupling filter)
>
>Made 1nF about 40kV in oil using 65mm plumbing pipe and fittings.
>Poly was about 0.5mm thick and used 3 layers to achieve required
>dielectric strength. No problems to date running at 12kvpk
>Would suggest the MMC path for main cap but polystyrene should be
>ok for rolled caps if you really want to punish yourself
>Best
>Ted Linney in NZ

Neat.  I was thinking of doing something vaguely masochistic and maybe
polymeric using the thinners (monomeric styrene) which you can get
from glass fibre suppliers.  Be nice to use some of my professional
ability in making a cap for one of my coils.

> Original poster: "John by way of Terry Fritz <twftesla-at-uswest-dot-net>"
<zavisa-at-home-dot-com>
>
> Also, have any of you used polystyrene for your caps?  It's seems so
> much cheaper to use about 93mils (a sheet of 1/32 and a sheet of
1/16)
> to get the required capacitance to match your transformer than going
the

Seems it can be done!  There may yet be mileage in this :-)

It seems that much depends on just how the polymer is prepared,
casting in vacuo gives the best results and films are significantly
worse, I suspect because of the need to plasticise films to reduce
brittleness for the rolling machinery.  I can, on reflection, think of
a number of reasons why the quality of the product might be critically
related to the type of process used to make it (but then I can also
criticise PP on the same basis.)  All good stuff!  Thanks!

Dunckx