[TowerTalk] [Fwd: Re: Lubricating air variable capacitors]

Steve steveac at charter.net
Wed Oct 5 22:47:35 EDT 2005


Barrie,

I cleaned the capacitor with Simple Green followed by the dishwasher per 
your advice. Took the cap back my friend's shop today. The 1-2 microamps 
of leakage at 1KV before the cleaning process was reduced to about one 
tenth of a microamp afterward.

Then that I noticed some dark matter on the end of one of the 
insulators, I'd call it carbon tracks but that may not be the right 
term. I cleaned that area with a pencil eraser, removing the "carbon 
tracks".

Tested again and found zero leakage current at 1KV.

The capacitance still tests higher than manufacturer specs . . . perhaps 
my friend's Sencore LC-102 needs calibration :-\

Thanks everyone for the advice.

73,

Steve K8JQ

Barrie Smith wrote:
> 
> 
>> The LC-102 can also generate as much as a thousand volts for measurement
>> of leakage current. So I put 1000VDC (the capacitor is rated for 3500
>> volts) across the air variable and measured leakage current thru it. The
>> value of leakage current varied ranged between 1 and 2 microamps as I
>> turned the rotor.  Leakage current seemed to be slowly increasing as
>> time passed. What is typical leakage current for this type of capacitor?
>> What is the leakage path? Can the leadage be reduced/eliminated? Or,
>> more to the point, should I be concerned with this level of leakage
>> current? As a comparison, I had three other air variables (including two
>> big transmitting air variables of the type used in Johnson KW
>> Matchboxes) that measured zero leakage under the same test conditions.
>>
>> Advice?
> 
> 
> As I mentioned in a previous post:  Is the capacitor clean?  Are any of 
> the plates bent, or have nicks?
> 
> Spray it with "Simple Green", let it set for a few minutes, and then run 
> it through the dishwasher a couple of times.  Let it dry completely and 
> test it again.
> 
> 1 to 2 microamps is way too much leakage for a 3500 volt cap.
> 
> Leakage should be measures with the plates fully engaged.
> 
> Leakage will increase considerably if there is dirt and grime on the 
> plates or the end-caps.
> 
> 73, Barrie, W7ALW
> 
> 
> 


More information about the TowerTalk mailing list