Head-Fi.org › Forums › Equipment Forums › Sound Science › Capacitor burn in, how long does it really take?
New Posts  All Forums:Forum Nav:

Capacitor burn in, how long does it really take?

post #1 of 51
Thread Starter 
The concept of burn in is always fairly controversial, of some things more than others. But one component in the system chain that most people seem to acknowledge as requiring burn-in is capacitors, in particular certain types of boutique capacitors such as Black Gates, V-Caps, and so forth.

The disputes tend to arise as to exactly how long capacitors take to form their dialectric and be "burned in". Is there a relationship between burn-in length and the amount of capacitance in any given cap? Do different types of capacitors form their dialectric at different rates - such as electrolytic, ceramic, paper-foil-in-oil, and so on?
post #2 of 51
Capacitors form up within a second or two. Then they degrade over time. Degradation depends on how hard they're run, how much heat they're exposed to and what they're made of.

The only capacitors that hold up over the long run are ceramic discs, mica caps and the air-gapped tuning capacitors that you find in radio tuning sections. Everything else goes bad eventually.

If you want your caps to last, minimize their exposure to heat. In other words, turn your rig off when you're not using it.

If you think caps sound "better" as they age, that's OK. For me, the newer and fewer hours the better.
post #3 of 51
Quote:
Originally Posted by Covenant View Post
The concept of burn in is always fairly controversial, of some things more than others. But one component in the system chain that most people seem to acknowledge as requiring burn-in is capacitors, in particular certain types of boutique capacitors such as Black Gates, V-Caps, and so forth.

The disputes tend to arise as to exactly how long capacitors take to form their dialectric and be "burned in". Is there a relationship between burn-in length and the amount of capacitance in any given cap? Do different types of capacitors form their dialectric at different rates - such as electrolytic, ceramic, paper-foil-in-oil, and so on?
This is an instance where the ignorant take something from one context and erroneously try to apply it to other, completely unrelated contexts.

The "forming" of a capacitor's dielectric has only to do with aluminum electrolytic capacitors.

In a film cap, or a ceramic cap, or a paper and foil cap, the dielectric is already formed. It's the film, or ceramic, or paper.

In an aluminum electrolytic capacitor, the dielectric is an aluminum oxide layer on the aluminum foil. This oxide layer is formed by way of anodizing. After the capacitor is made, a "forming voltage" is applied to it, which builds up the aluminum oxide layer, hence the term "forming."

Anyone who tells you that the dielectric in a film or similar type cap needs "forming" is talking out of their ass.

k
post #4 of 51
Thread Starter 
Quote:
Originally Posted by Koyaan I. Sqatsi View Post
This is an instance where the ignorant take something from one context and erroneously try to apply it to other, completely unrelated contexts.
Thank you for the clarification. I freely admit to ignorance of this topic, which is why I asked the question

Quote:
In an aluminum electrolytic capacitor, the dielectric is an aluminum oxide layer on the aluminum foil. This oxide layer is formed by way of anodizing. After the capacitor is made, a "forming voltage" is applied to it, which builds up the aluminum oxide layer, hence the term "forming."
Does the forming of that oxide layer continue as further voltage is applied, or is it a once-off process that occurs when the capacitor is manufactured?
post #5 of 51
Quote:
Originally Posted by Covenant View Post
Thank you for the clarification. I freely admit to ignorance of this topic, which is why I asked the question
Well, there's ignorance, and then there's ignorance. And ignorance in and of itself is not a bad thing nor was I ever intending to impugn you.

We're all ignorant of some things or other, such as you were of capacitors. That's not a bad thing because you KNOW you're ignorant on the subject of capacitors just as I KNOW I'm ignorant on the subject of... well, actually I'm not ignorant of anything but that's beside the point.

The problem comes about when people don't know their limitations and think they know more than they do and go on to make assumptions about other things based on that limited knowledge.

Quote:
Does the forming of that oxide layer continue as further voltage is applied, or is it a once-off process that occurs when the capacitor is manufactured?
Well, since the anodizing process requires current, and since aluminum oxide is an insulator, for a given forming voltage, as the oxide layer gets thicker, current diminishes until the forming voltage is no longer able to shove any more current through the capacitor.

It's ultimately the forming voltage that determines the thickness of the oxide layer.

k
post #6 of 51
Thread Starter 
Quote:
Originally Posted by Koyaan I. Sqatsi View Post
We're all ignorant of some things or other, such as you were of capacitors. That's not a bad thing because you KNOW you're ignorant on the subject of capacitors just as I KNOW I'm ignorant on the subject of... well, actually I'm not ignorant of anything but that's beside the point.
He's modest too, folks

Quote:
Well, since the anodizing process requires current, and since aluminum oxide is an insulator, for a given forming voltage, as the oxide layer gets thicker, current diminishes until the forming voltage is no longer able to shove any more current through the capacitor.
Does that whole process occur during manufacture, though? Or are eletrolytics sometimes shipped/sold with the dialectric unformed, or partially formed?

What I'm getting at is trying to figure out just how much truth there is to capacitor burn in beliefs, and the extent to which they're true.
post #7 of 51
wet Al electro's are a special case - long storage without bias causes degradation of the Al Oxide layer and an increase in leakage current

decades old equipment that has been in storage should be powered up slowly with a Variac to ramp up the V across the Al electro supply caps and allow the Oxide layer to "reform" - the excess leakage current could overheat the Al caps that could otherwise have been saved


a relatively unexplored dielectric property is that even good plastic films absorb water at fractional % levels - and water has a dielectric constant of ~80 and is a highly polar molecule

careful measurements of even polystyrene shows ~10x increased dielectric loss with water absorption

I think dielectric properties of the PCB substrate, capacitor films and plastic packaging in electronic circuits all change with equilibrium moisture content - and that some of the moisture can be "cooked" out by long run times of hot running Class A or tube circuits

I wouldn't bet that the effects would be audible in well designed, properly controlled/blinded listening tests though
post #8 of 51
Quote:
Originally Posted by Covenant View Post
He's modest too, folks
Actually I am. It's just that I have this terrible weakness for smartass comments.

Quote:
Does that whole process occur during manufacture, though? Or are eletrolytics sometimes shipped/sold with the dialectric unformed, or partially formed?
Far as I'm aware, electrolytics are supplied fully formed from the factory.

Quote:
What I'm getting at is trying to figure out just how much truth there is to capacitor burn in beliefs, and the extent to which they're true.
There seems to be about as much truth in it as there is in cable burn in.

k
post #9 of 51
Quote:
Originally Posted by Covenant View Post
What I'm getting at is trying to figure out just how much truth there is to capacitor burn in beliefs, and the extent to which they're true.
You might notice that products that make scientifically verifiable claims always cost less.

Since test equipment and scientific knowledge and training come at a price, you might expect there to be more overhead to cover hiring engineers and setting up a test lab.

Strangely, it seems that the less scientific a product is, the more it costs. Those costs are often wildly more than the cost of manufacturing the product.

Any guesses as to why that is?
post #10 of 51
technical

In Blackgate technical papers, the idling process takes an average of 30 hours. (bottom of page)
post #11 of 51
Quote:
Originally Posted by Happy Camper View Post
technical

In Blackgate technical papers, the idling process takes an average of 30 hours. (bottom of page)
Not that there's nothing to be said for Black Gate capacitors, but that "technical paper" is more marketing literature than technical literature.

k
post #12 of 51
Quote:
Originally Posted by Covenant View Post
He's modest too, folks
Nah... Just say it how you feel it... He won't mind. He's a cocky SOB. Ain't that right Steve?
post #13 of 51
Quote:
Originally Posted by IPodPJ View Post
Nah... Just say it how you feel it... He won't mind. He's a cocky SOB. Ain't that right Steve?
Cocky SOB? What'choo talkin' 'bout, Willis?





k
post #14 of 51
In court, it is the manufacturer's expert advice and stands as fact in it's applications. You are interjecting opinion as to their claim.

How often do you change oil in your car K?
post #15 of 51
Thread Starter 
Quote:
Originally Posted by Happy Camper View Post
In Blackgate technical papers, the idling process takes an average of 30 hours. (bottom of page)
Can you explain to the plebeians like me what that means?

Quote:
Originally Posted by Koyaan I. Sqatsi
Not that there's nothing to be said for Black Gate capacitors
What is there to be said?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Capacitor burn in, how long does it really take?