[sdiy] How do old electrolytics go bad?

[Bob Weigel]

That's right I recall you explaining this long ago I think regarding the 
construction.  Something I never studied for some reason along the way.  
Thanks for the refresher. -Bob

Roy J. Tellason wrote:

>On Friday 06 February 2009 05:40:34 pm Bob Weigel wrote:
>  
>
>>I typically see old electro's drift higher and higher..even to 10X the
>>original value and then they'll plummet often to dead or they'll even
>>short sometimes.  This in observing Fender and other amps of various
>>ages largely.
>>
>>The *source* of this phenomena, I assume, is that the electrolyte
>>gradually degrades and in doing so some functional distances may be
>>altered..and of course closer plates mean higher capacitance.  Shrinkage
>>of the dielectric material.   And then as it gets too close it punctures
>>and leaks sometimes.  Or something totally disconnects in other cases.
>>Not sure but rambling possibilities anyway :-)
>>
>>Anyway that phenomena is common whatever the case, let me assure you.  -Bob
>>    
>>
>
>The actual electrolyte isn't the stuff that's between the plates,  which is 
>conductive BTW,  but a thin layer of oxide (?) on one plate,  I forget if 
>it's the positive or the negative one.  This is applied when the cap is made,  
>and the thickness of the layer determines both the capacitance and the 
>working voltage.
>
>As time goes by,  and with no voltage applied,  the oxide will start to 
>dissolve,  giving a thinner layer,  higher capacitance,  and lower working 
>voltage than what's on the label.  This happens even with brand new caps that 
>have been on the shelf for a while.
>
>The solution is to apply current-limited voltage to the cap,  over time,  
>which re-forms the oxide layer.  This is an especially common practice with 
>folks who are building antique-style gear or trying to restore old radios,  
>where the required parts may simply not be available any more.
>
>Google "electrolytic re-forming" for more info on this.
>
>The other thing is that most electrolytics have fairly wide tolerances,  most 
>often either -20%/+80%,  or -0%/+100%...!
>
>I got a cap checker a bunch of years ago,  a B&K unit.  What got tossed out 
>was a bunch of caps that were all reading *way* low that I'd salvaged out of 
>assorted gear over the years.  Stuff that read on the high side never 
>bothered me all that much.
>
>  
>
>>Ingo Debus wrote:
>>    
>>
>>>Hi,
>>>
>>>we all know that electrolytic capacitors often go bad when they are
>>>old. In the past I had replaced some that had a significantly lower
>>>than rated capacitance. But now I got two that show a higher than
>>>rated value on my DMM: one is rated 100uF/3V and shows 270uF on the
>>>meter, the other is rated 16uF/16V and shows 23uF. Of course I
>>>desoldered them from the circuit before measuring.
>>>Is this normal? Or did they have the higher values from the
>>>beginning? Or are they leaking and thus fooling my meter?
>>>
>>>I estimate these caps are about 40 years old.
>>>
>>>Ingo
>>>_______________________________________________
>>>Synth-diy mailing list
>>>Synth-diy at dropmix.xs4all.nl
>>>http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>>      
>>>
>>_______________________________________________
>>Synth-diy mailing list
>>Synth-diy at dropmix.xs4all.nl
>>http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>    
>>
>
>
>
>  
>