Some capacitor leakage measurements

===================

ear.

d

I wanted to try 'em in a high Q LC BP filter. Low dissipation factor is what I read. (at some point leakage will look like dissipation... I think.?*) The fact that you can't buy 'em implies there is no real market. I always wonder what's the market for the big 10 uF film panasonic caps I like. (mentioned above.)

George H.

*so what's the analog of charge in mechanical mass- spring system? I'm thinking the mass.

Well, yeah; Keithley electrometers used a lot of 500V rated Mallory SX polystyrene caps, for low-voltage parts of their innards. I assumed that they were the good solution at that time (but you can't reflow-solder polystyrene parts very well).

They're not high production items, so you skip the distributor and call the manufacturer up. Ask about surplus or overruns.

makers are

bycap electrocube arizona capacitor Electronic Concepts, but they suck as a company

I don't have any spec sheets in front of me, but I don't recall teflon having impressive megohms/uF values. If you need high temp go teflon. For anything else, other dielectrics tend to be better.

The whole shakeup over polystrene and why only one company still makes them has to do with the problem of only the highest quality plastic films can be used to make capacitors, and capacitor makers are the smallest user of plastic films. One of the above companies actually has their own plastic film production line as nobody else can make film to a usable yield. One factory I went to had a mountain of pallets of plastic film that was rejected by their QC. It was probably good film, but not good enough for capacitors.

I suspect that even if they are better, they will set the specification at the limit of what they can measure in a reasonable time period.

Thanks for the interesting story.

I had read somewhere that nobody was making the polystyrene film anymore and that capacitor makers who sold them were just using up old stocks of film, but I had no way of knowing if all that was true.

For those in Australia, there is a company (Rockby) selling new-old-stock Evox-Rifa polystyrene capacitors, but only a few very odd values, and sometimes discounted to very cheap prices (a few cents). I bought a lifetime stock of a few values and have used a few from time to time, for peak-hold circuits etc. Not all of their polystyrene capacitors are new old stock, I took a chance by ordering one lot that did not have a manufacturer's name and part number, and those seem to be from China.

No, a nanoamp or two will not flatten their battery, so going to any effort to find a capacitor that leaks less than picoamperes is not sensible in this situation.

Good point.

Yes, I noticed that when I was looking into phase-stable cables for a VNA. Also I suspect that other dielectrics may be better for microwave PCBs in measuring instruments where stability over temperature is important. I guess you could also ovenize everything at a higher temperature, where the TC is lower.

Anything that requires long assured lifetime, like automotive, industrial and medical. Check out the service life of electrolytics in elevated temperatures - even with lifetime temperature scaling, would you accept those for your cars brake or steering system ?

--
mikko

Teflon (PTFE) is notoriously microphonic.

Years ago I made a femtoampere-class electrometer, and when testing it the output swinged back and forth at a leisurely rate. The swinging happened synchronously with the movements of the Teflon-insulated input coaxial cable.

--

-TV

I can't see why they'd need a less leaky capacitor there than the much larger ones they use for the ringers in phones.

--
  Jasen.

I noticed that too, in pH meter inputs. Teflon had low leakage but high triboelectric charge generation.

Ever seen demo set ups for ESD generation? They place a metal plate on top of a teflon block, then press (or slide) it, then pull it up. Kilovolts get generated (big sparks). Teflon seen to be very (negative charge) triboelectric, see

I'm sure many have seen this, but for those that haven't it I interesting.

Mikek

Electronic Design is pitiful. I received the printed mag yesterday; I have no idea why. The entire issue isn't much thicker than a business card. The few articles are pitiful, mostly blather about silly software. The online version is unbearable.

--

John Larkin         Highland Technology, Inc 

Science teaches us to doubt. 

  Claude Bernard

I think you have mentioned this before. It seems with the internet revolution, most print media has gone downhill, even my local newspaper has shrunk in pages and in physical dimensions. But that's a different subject, I presented the link to show info about Teflon, that I thought was interesting. I do think I originally saw the write up in a different source, but when I searched, this is the source I found. Sorry it set you off! I suggest you unsubscribe and don't read the online version of Electronic Design. Mikek

It's a discussion group. We need stuff to discuss, ideally on the topic of electronics.

I have no idea why they have started mailing me the silly skinny mag. I don't look at their online version any more. It's annoying and useless.

I do need to keep current on what's happening. The mags (ED, EDN, EE Times, Electronics Weekly, All About Circuits) are barely worth a quick glance, and some are so loaded up with web tricks that they are unbearable. Why does anyone need 35 cookies?

Once a month roughly, I cruise the various semiconductor web sites and look at the new product listings. Digikey and Mouser have new product pages too.

Any other ideas on how to keep up to date? One would think that such a giant industry as ours would support at least one decent mag.

The optics mags are still fat and good, full of ads and articles and product releases. As are the microwave mags, but I don't do RF (ie, narrowband frequency domain) much, so only a fraction of their content is useful to us.

Repeated rant: the DC specs of RF parts are usually terrible.

--

John Larkin         Highland Technology, Inc 

Science teaches us to doubt. 

  Claude Bernard