[sdiy] A little on capacitors from people smarter than me

[cheater cheater]

Hi guys,
I was following a recent thread elsewhere on capacitors in the audio
chain and distortion. I'm sure a lot of people here know all about
this, but I think it's a relevant topic here especially for beginners
like me. A lot of the stuff I see (especially modulars) uses
cheap/old/random caps, or people go with the first part they can find
in their box, etc. It can and does make a big difference in audio
quality. I thought I'd forward some materials and interesting opinions
for anyone who might be interested.

One particular thing that's pointed out is that if the (electrolyte)
capacitors were shelved for a few years it's good to reform the oxide
layer on the electrodes by slowly bringing up voltage across the
capacitor (in the right direction, of course). This lowers leakage
current. If after reforming the capacitor still has high leakage
current it's broken. In that case reforming also turns out to be a
good way to screen capacitors for deficiencies.

The whole process of reforming and testing is easy to set up and can
be automated, which is important for people who want to manufacture,
or just don't want to lose time.

Out of curiosity - how many people here have been applying this?

Cheers
D.

------------------------

http://waltjung.org/PDFs/Picking_Capacitors_1.pdf
http://waltjung.org/PDFs/Picking_Capacitors_2.pdf
http://electronicdesign.com/article/analog-and-mixed-signal/bob-s-mailbox15653.aspx
http://www.scribd.com/doc/2610442/Capacitor-Sound

on capacitor reforming:
http://www.vcomp.co.uk/tech_tips/reform_caps/reform_caps.htm

-------------------------

Please see the technical paper that I wrote entitled "Ceramic
Capacitors" back in later 1980 or so, basically in response to the
Jung/Marsh article in Audio Magazine. My paper is on the last page
(page 8) of my 990 data package:

http://www.johnhardyco.com/pdf/990-2007.pdf

In it I refer to Jung's use of X7R ceramics in his tests. I also make
reference to AES preprint #1719(D1) entitled "Topology Considerations
for RIAA Phono Preamplifiers" where Jung expresses a favorable view of
the COG/NP0 dielectric. I contacted him about the COG/NP0 dielectrics
after his Audio article, and he must have taken a closer look (first
look?) at the COG/NPO parts for the AES preprint.

John Hardy

--------------------------

(...) from what I can tell, the oxide
layer is reformed by current flow, but only in one direction.  In the
wrong direction, it actually erodes it, which is supposedly why the
release magic smoke so quickly when the polarity is reversed.

So the idea, is that if they've been sitting for a while, you should
slowly bring up the current so that the oxide has time to reform.  If
current doesn't fall to a steady state, then the cap is fubar, and
must be replaced.

I say if one is worried about reforming the thing in the first place,
just replace it, but hey.

Cheers,
Christopher Deckard

----------------------------

> If I remember to, I'll bring a recap job up on the variac, but more
> often than not, I just throw the switch, and let 'er rip.

I generally take that approach. After the 1977 blackout in NY City,
when power came back on at ABC's network headquarters, hundreds of
pieces of equipment failed to come back up and it was almost
exclusively power supply capacitors.

So, when I'm working on a piece of equipment, I figure that if the
caps are weak, hitting them with the startup surge will identify the
weak ones and then I can replace them.

I may be more circumspect with ancient tube gear, but I don't work on
that, I think I have only one tube antique here, and I'd probably
power it up carefully after changing the power transformer as it
hasn't been on for > 25 years. I do have the replacement power
transformer in my round toit pile.

Cheers,

Richard

Richard L. Hess

---------------------