Aluminum Electrolytics (was: M7891-DC 2a fuse on +12? fixed!)

From: CRC <technobug_at_comcast.net>
Date: Fri Feb 6 23:29:03 2004

On On Thu, 5 Feb 2004 15:47:34 -0800 (PST), Vintage Computer Festival
<vcf_at_siconic.com> wrote:

> One second? I've always been told to do this over the course of at
> least
> a couple hours in 5V increments.
>
> If I can be doing this in a much quicker time then I'd really like to
> know
> about it.
>
[...]
And on Thu, 5 Feb 2004 19:04:52 -0500 (EST), ohn Lawson
<jpl15_at_panix.com> also wrote:


> Then duck! IMHO opinion, and over 40+ years experience, to reform
> funky caps takes anywhere from a few minutes to over an hour,
> depending on
> many variables. If you think you've formed an electrolytic in one
> second,
> it was either tiny or good-to-begin-with.
>
> But the thing most folks forget is to somehow monitor the *current*
> being drawn during the Process of Reformation.
>
> Here endeth the Lesson.
>
>
>
>>
>> One second? I've always been told to do this over the course of at
>> least
>> a couple hours in 5V increments.
>>
>
>
> Again - it depends on the voltage/capacity of the filters involved,
> and
> just how "bad" gthey are. One simply cannot place an arbitrary
> voltage/time delta on all power supplies.
>
> The Ammeter (or it's analog) is here your friend.
>
Also at Thu, 5 Feb 2004 16:12:53 -0800, "vrs" <vrs_at_msn.com> commented:
>>>>

>>>> To safely re-energize a piece of equipment that is long in the tooth
>>>> and has been sitting around for some time, you have to bring the
>>>> voltage up relatively slowly to allow the electrolytic caps to
>>>> re-polarize. Everyone who is collecting should own a good-sized
> Variac.
>>>> You power the box up by bringing the voltage from naught to the
>>>> operational level with the Variac over a period of about a second or
>>
>> Then duck! IMHO opinion, and over 40+ years experience, to reform
>> funky caps takes anywhere from a few minutes to over an hour,
>> depending on
>> many variables. If you think you've formed an electrolytic in one
>> second,
>> it was either tiny or good-to-begin-with.
>
> When I read this, I took it to mean the Variac was a preferred method
> for
> turning on equipment *after* the caps had been reformed, as a matter of
> routine. (Otherwise it made no sense to me, based on what I know of
> reformation.) That may not be a good idea, but I couldn't dismiss it
> out of
> hand.
>

I reference <http://www.nichicon.com/english/lib/alminium.pdf>, and
excellent overview of aluminum capacitors...

The controlling parameter in reforming a capacitor is the amount of
heat generated. Generate too much and you blow things up. We have to
consider two cases: the capacitor is in circuit, or out of circuit.

In the latter case, the recommended method of reforming is by applying
the forming voltage through a limiting resistor and then letting the
capacitor form at voltage over a period of time with the limiting
resistor in circuit. As commented above, the current should be
monitored, but the critical term is the final leakage current. If it is
out of spec, there is nothing you can do... A capacitor manufacturer
used to be in town and they had to monitor old stock for mil spec
reasons. Their criteria for reforming old stock was a case temp rise of
no more than 5 degrees C and the leakage current had to be in spec in
less than 10 minutes.

An old electrolytic that takes a long time to reform has probably lost
electrolyte and should be considered suspect (its ESR is probably
miserable and capacitance below spec).

The case of in-circuit caps: With a lot of equipment, if you put it in
a brown-out condition, you can fry the silicon. Older switchers are
notorious for burning up when operated below their minimum voltage for
any length of time. In any case, when lighting equipment, the primary
concern is the front end caps. If the leakage current is substantial,
you will blow the fuse or take out the rectifier (valves don't have
this problem). By using a Variac to bring up the voltage over several
seconds, you mitigate the surge and allow initial reforming to take
place. If things go, you might as well replace the caps.

Depending on the power supply, caps down stream will either see the
operating voltage and have to "live" or, in the case where the power
supply current limits, will pull down the power supply until formed,
if . In the case of switchers, they have a tendency to *turn on* if
substantially over-rated (hey, remember we have been taught that it's
only current that counts ;<)).

So all is not cut and dried. You have to know what you are working
with... Hey, and I match your 40+ and raise you a few...
Received on Fri Feb 06 2004 - 23:29:03 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:41 BST