[ARC5] Reforming Electrolytics attached as PDF

[Bob kb8tq]

Hi

What ever you want to call it, sitting at a fixed bias voltage and running that way for a number of 
hours / days *does* reduce the leakage on both aluminum electrolytic and tantalum capacitors. 
If you dig into the test methods / spec sheets, you eventually will run into a “pre bias” process 
before the caps are tested for the specified leakage current.  If you put bias on one that has been
on the shelf for a while, it’s leakage will indeed be a bit high at turn on and you can watch it drop 
for quite a long time as the insulation “fixes” its self. 

Bob

> On Mar 19, 2018, at 2:46 PM, Richard Knoppow <1oldlens1 at ix.netcom.com> wrote:
> 
>   It should be understood that there is both series and parallel resistance in a capacitor. ESR may be less important than parallel resistance, which is also called leakage.
>   All electrolytic capacitors have leakage and it is the leakage that ruins them when it becomes too low. ESR, also called Dissipation Factor, is a measure of the Q of a capacitor while parallel resistance or leakage measures the amount of DC is passes.
>   Ageing without voltage may reduce the thickness of the dielectric layer. This will lower the effective voltage rating of the capacitor and may raise its capacitance, but may not affect its leakage. A high leakage capacitor of any kind may measure OK for both value and ESR so both kinds of measurement are necessary. The leakage is measured by testing for current through the cap. An ideal capacitor should not pass any DC current. Practical electrolytics always have some. The old capacitor checkers of the sort with a magic eye indicator were leakage testers. They usually have a variable DC source for electrolytics and a means, often the eye tube, for indicating leakage current.
>    I don't think reforming will help leakage.
>    A typical impedance bridge does NOT measure leakage although it may give a value for parallel resistance. These two are confusing because they use the same name for different things. An impedance bridge may indicate either series or parallel resistance depending on its configuration and range but, unless it of the kind that has a DC source for the capactor it will not indicate leakage resistance.
>    Leakage may not be apparent with a low test voltage because it may be due to break down of the dielectric at some voltage. However, it may show up even with a conventional ohm meter when it reaches high value. I have bad electrolytics that show bad on a Hewlett-Packard 410-B VTVM ohm range and also show bad on a General Radio Megohm meter and also on an old Eico capacitor tester. They are just bad.
>    Storage in a capacitor is another way of measuring leakage. ESR will not affect the rate at which a charge is lost, that is due to parallel resistance. It is partly a property of the electrolytic and partly of the insulation of the capacitor body. Good non-electrolytic caps can hold a charge for years.
>    There are other properties of capacitors such as dielectric absorption AKA voltage recovery, which can be critical in some applications and unimportant in others. This is the effect of a capacitor not fully discharging when shorted. After the terminals are opened again some voltage, often considerable appears across them. This can be very important in capacitors intended for timing and low frequency applications. Mica caps have a lot of absorption, paper is low in absorption and plastic has very little. It is of no importance in the usual applications for electrolytic caps.
> 
> 
> 
> On 3/19/2018 10:43 AM, Kenneth G. Gordon wrote:
>> On 19 Mar 2018 at 10:06, Bob kb8tq wrote:
>>>     His method is to reform the electrolytic VERY SLOWLY, with a B+ source limited to
>>>     FIVE MA., while monitoring the Cap voltage.
>> This is EXACTLY what that military document I posted says: 5 mA, no more, over a long
>> period of time.
>> Some of us use 1 mA, which may actually be too low. Testing is required.
>> I also have and use an ESR meter.
>> Ken W7EKB
> 
> -- 
> Richard Knoppow
> 1oldlens1 at ix.netcom.com
> WB6KBL
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html