What actually Fails inside of Paper Capacitors

Here's a simpler way to look at it.

Paper capacitors are an old type of capacitor. ELementary Electronics about 1971 showed how to make them yourself, some paper, some foil, some liquid if I recall properly. I think even then I knew they were hideous, that piece of foil was relatively large, must have inductance. We see this, paper capacitors were often marked with "+" or "-" so you knew which end to connect to ground. They weren't polarized, but they might pick up unwanted signals if you connected the wrong end to ground, it was the nature of how they were made.

They were also fairly big for the capacitance offered.

But for a time, they were what was available. When you started posting, I got curious and apparently bypass value ceramic capacitors took some time to become feasible, which has to explain all the old radios with paper capacitors. When I got into the hobby in 1971, nobody would suggest paper capacitors for bypass capacitors (and probably not anything else). "Keep leads short", but that didn't work so well when the paper capacitor was the equivalent of a long lead.

So it doesn't matter why paper capacitors fail, there's lots of reason to replace them anyway because they are an antiquated form of capacitor. The switch to ceramic bypass capacitors happened because you could get a more reliable package, but it was also smaller, and lower inductance.

Michael

No and yes

Not until there is a fault.

Not so much the foil as the paper dielectric.

First and foremost, the paper was NOT acid free archival paper. Mainly because everyone assumed the wax would seal the moisture out. It doesn't because it is hygroscopic. So what happens over time is moisture gets into the capacitor paper, mixes with the acid content eats the metal foil and causes conductive paths in the paper. At first the leakage goes up, until some point when the capacitor fails shorted.

--
Jeff-1.0 
wa6fwi 
http://www.foxsmercantile.com

I'd like to see that E.E. article. That would be sort of fun to try, just for the heck of it.

E.E. does have PDF file reprints of all their issues on

But finding the right issue might be tough...

Guess I'll begin by downloading all the issues from 71. That will take most of the day on dialup....

1) Cellouse is marvelously hydrgroscopic. It sucks up humidity like a sponge. Wet cellouse is conductive. There's no reason to overthink this. Also, don't assume the paper the manufacturers used was stored or wound in perfectly dry conditions. That's nearly impossible. I have no doubt caps made under humid conditions failed more quickly than caps made during dry times.

2) Wax is NOT a perfect seal against moisture. I'm sure plastics are better but they aren't perfect, either. Ever see a tupperware container sweating from the inside? That moisture diffused it's way through the polyethelyne in maybe a few days or even hours.

Paper caps from the old days that HAD to last a long time were sealed in soldered metal cans in oil. Too expensive for consumer gear! The engineers who designed this stuff KNEW the paper caps would be failing in a few years. So what? Electronics changes fast and the buyer would almost certainly buy something newer before deteriorating caps killed the device.

Using good caps in consumer gear would be a waste of money. Cheap caps were considered good enough.

As far as what kind of wax was used, some sources say beeswax, some say microcrystalline wax. I don't much care.

There is that. But, let's start with the wax: A mixture of beeswax (small a mount), paraffin, and petroleum jelly for handling. Does anyone here know h ow old pure beeswax candles were made? Beeswax was dissolved in warm water, and the wicks were dipped multiple times into the solution, picking up a b it of additional wax each time. the keyword is *DISSOLVED*. Even paraffin w ax holds water, and beeswax is directly water-soluble.

Now, the paper: Rather thin waxed paper purchased from the lowest bidder. A nd the wax, as above, was subject to moisture damage leading to paper damag e.

And, of course, the foil. As above, without the wax.

And, there are multiple sites dedicated to reproducing these beasts for aud iophile/phool purposes. Go figure.

Peter Wieck Melrose Park, PA

t.

amount), paraffin, and petroleum jelly for handling. Does anyone here know how old pure beeswax candles were made? Beeswax was dissolved in warm wate r, and the wicks were dipped multiple times into the solution, picking up a bit of additional wax each time. the keyword is *DISSOLVED*. Even paraffin wax holds water, and beeswax is directly water-soluble.

And the wax, as above, was subject to moisture damage leading to paper dam age.

udiophile/phool purposes. Go figure. Huh, Well doesn't a paper cap add that certain 'crispness' to the sound? :^)

George H.

I know a woman who does historical re-enactments. One of her things is making candles. You are right, the wicks are dipped in pure beeswax, (Many times), and each time they get thicker.

I am not sure if you're saying the beeswax is dissolved *IN* the water, or if you mean the container holding the wax is placed in warm water. But to clearify this, there is NO WATER *IN* with the beeswax.

Melting the container of wax with the container in some warm water would likely work. Although she melts her container over a bed of hot coals, which were created by burning wood inside a shallow hole in the ground, surrounded by rocks. There is an iron grill across the rocks, and she has a tin pan between the wax container and the grill (to keep wax from dripping into the fire).

It's a simple but time consuming process. She says the main thing is to keep the temperature right, so the wax stays melted, but dont overheat. That's all accomplished by the spacing between the coals and the wax pot, as well as the coals themselves.

At least now I know what wax is used, (or combination of). The petroleum jelly (vaseline) part seems kind of bizarre, and why that helps with "handling" makes no sense to me. But I'll take your word for it.

So, now we know how water molecules can migrate thru the wax. But that dont explain how they get inside of those old plastic coated caps, (called bumblebees), or similar old caps. Those are sealed real well, unless they develop a crack.

As in "burnt to a.... "?

Peter Wieck Melrose Park, PA

They are NOT "sealed real well." Nor are the old "Micamold" paper dielectric capacitors.

I just re-capped a Collins R-390A receiver. There are 28 paper dielectric capacitors in the one I worked on. About

20 of them were the Sprague bumble bee types, and the rest were Sprague Vitamine-Q.

Go ahead and say it, "Collins! Top of the line! Mil-Spec!" They were still crap capacitors

--
Jeff-1.0 
wa6fwi 
http://www.foxsmercantile.com

l amount),

d pure

additional

r,

In the systems I have observed, the wax is mixed into the warm water, and f or the most part floats on top - Little actually dissolves, and what does m ay be partially oxidized material. But there is definitely some that does. What this method does is allow a very thin layer of wax (as little as two i nches) will allow full dipping once the wicks have their first coat. The ol d farmer that demonstrated this to me explained that it was much safer than putting a separate container in water, or, especially, over a fire.

It does make sense if you consider the need to make a stable material that is easily spread, and spread very thin, and does not go soft at hand-temper atures (which are generally much lower than the typical tube radio. Paraffi n makes the hard base, but does not spread well and needs to be quite hot t o melt fully. Beeswax gives the spreading properties, but gets soft, so is used in small quantities. Petroleum jelly (thick, pure oil) prevents the ma terials from clouding and also reduces sticking to the processing machinery , operating at high speed, mostly. Keep in mind that both the beeswax and t he jelly are minor admixtures,

This is an easy one. A 'bumblebee' cap seal relies on the interface between two dissimilar materials with vastly different coefficients of expansion. Plastic and metal. Over time, micro-cracks will develop at the leads. Keep in mind that these capacitors live in a tough environment, and can go from freezing to very hot quickly and repeatedly.

The failure mechanisms are common, well understood, and time-sensitive - fa r shorter than "today". Unless you dabble in instrument amps, meaning that your caps serve as inductors, resistors, and occasionally as capacitors, yo u have no business leaving these beasts inside anything expected to functio n.

Peter Wieck Melrose Park, PA

Those oil filled bathtub caps failed, as well. The paper broke down, just like waxed caps, plus the rubber plugs would harden, crack and leak oil.

--
Never piss off an Engineer! 

They don't get mad. 

They don't get even. 

They go for over unity! ;-)

Yes they do and we are seeing the failures now. But it's been a pretty good run for a paper dielectric cap. Much better than the wax covered caps.

Who are you kidding? Paper caps were already failing after ten years of service. The Sprague "bumble bees" and Black Beauty capacitors were already known failures by the '60s.

--
Jeff-1.0 
wa6fwi 
http://www.foxsmercantile.com

OP also consider in many tube applications the coupling cap has plate voltage on one side like 250V and feeds a 100k grid resistor on the other side.

It doesn't take much leakage to upset the grid voltage.

m