breaking in audio capacitors

[tomchr]

Originally Posted by kstagger

Case in point, I once tried out some Russian Teflon caps in a pair of single-ended monoblocks. At first I thought I could tell a real difference, but as the amplifiers broke in, the effect became smaller and smaller. Psychoacoustics? "dialectric relaxation?" I dunno

In your case, my money is on psychoacoustics.

Dielectric absorption (I'm assuming that's what you are referring to) is a real effect that can be quantified, measured, and verified through a simple experiment. Charge the capacitor to a specified voltage. Discharge it to 0V. Remove the discharging current and measure the voltage across the capacitor. The voltage will start off at 0 V and increase slightly. This slight increase in voltage versus the voltage the capacitor was changed to is a measurement of the dielectric absorption of the capacitor. Whether it has an effect on the sound quality of an amp or not, I don't know. I highly doubt it as the time scale for the voltage build-up is actually fairly long (seconds).

~Tom

[Triodethom]

My view given the cost of the big teflon caps designing for more direct connection multi-level power supplies that use the least cap in the signal path make good sense . It the historic pasts power supplies cost a lot and coupling caps where cheap reducing the number voltages and how well regulated they where . Now coupling caps can go 100s of dollars each and a well regulated power supply is far less to build.

[HollowState]

Originally Posted by tomchr

Whether it has an effect on the sound quality of an amp or not, I don't know.

Believe it, it does. It's not necessarily the absorption, but the speed with which it gives up it's energy. Plastic dielectrics are faster then paper in this regard. How do I know this? After reading an old artical in Audio Magazine by Walter Jung years ago called "Picking Capacitors", I replaced all the paper capacitors in my Ampex 351 clone electronics. Wow, what a surprise. The high frequency response was very exaggerated and peaked. In fact I had to readjust the equalization on both playback and record by a lot. Right to the end of the adjustment range in fact. Now this was capacitor replacement over multiple stages. Replacing one or maybe even two might not be that radical. But you will hear a difference if you listen carefully to an amplifier that has been "upgraded" with them. I'm not a golden ear, so if I can hear a change there really is a difference.

[Tony]

Originally Posted by monolith61

Hi there

Buy a second hand reciever on ebay, do not connect an antenna to the tuner to obtain noise and roll on the volume when there is a speaker in place. Check the level, replace speaker with the cap in series with a 100R 10W resistor. Let it cook for a while ..

grz, //WDC

MONOLITH MAGNETICS

so then what happens to the cap?, any measurements taken before and after the so called break-in?

comments like "i heard a difference" mean nothing to me, i wasn't there to hear what was heard to matter.......i do not take anecdotes as evidence.....

[tomchr]

Originally Posted by HollowState

The high frequency response was very exaggerated and peaked. In fact I had to readjust the equalization on both playback and record by a lot.

Did you measure this change?

I'm not trying to argue that capacitors don't color the sound. But I firmly believe that capacitor break-in is a hoax along the same lines as myrtlewood speaker stands. For capacitor selection, pick a type that as close to an idea capacitor (i.e. no ESR, no ESL, just capacitance) in the frequency range of interest. My favorite is polypropylene. Polyprop caps come close to the ideal cap in most cases.

~Tom

[Tony]

i also have a preference for polypropylene caps.... the generic ones are just as good as expensive branded ones....

[HollowState]

Originally Posted by tomchr

Did you measure this change?

No I did not because it was very audioable and I knew something was way off. And I don't believe in passsive parts break in either. Although I do believe that any given amplifier does sound better when fully warmed up. I also like polypropylene capacitors along with polycarbonate and teflons. I build with them. But I don't go replacing everything with them when working on vintage equipment because you will change the sonic character of the device. And this is especially true with music instrument amplifiers. More then once I've restored a guitar amp back to normal because someone upgraded it with "better sounding" parts.

[TubeHead Johnny]

I don't know about caps, but devices like tubes and speakers definitely sound better after a break in period.

[6J6]

Well with NOS / Older electrolytics, especially high voltage ones, it is good practice to install them in-circuit and slowly bring up the voltage with a variac to allow for reforming to occur gradually, and to avoid dielectric punch through from surge/ripple currents.

If the PIO is high voltage, the same logic could be applied if you want it to, even though it is non-polarised. Certainly wouldn't hurt.

Regards,
6J6.

[tomchr]

Originally Posted by HollowState

No I did not because it was very audioable and I knew something was way off.

I would argue that unless you have measurements that show the peaking of the HF response you describe, the root cause of the change in perceived sound could be anything. Including time of day, listener fatigue, "I did something so it must sound different" psychoacoustic effect, etc.
A capacitor is a capacitor. A 1 uF electrolytic cap will cause the same frequency response as a 1 uF polypropylene. But they cause different harmonic distortion, hence, different sound.

Originally Posted by HollowState

Although I do believe that any given amplifier does sound better when fully warmed up.

For all the tube amps I have built, the THD while fully warmed up has been significantly lower than it was when the amp was first turned on. In many cases, I've seen upward of 10 dB improvement from cold to warm.

Sand amps shouldn't change characteristics as they warm up, however. I would consider it a design flaw if warm-up in excess of 10~15 seconds was needed for good performance.

Originally Posted by HollowState

More then once I've restored a guitar amp back to normal because someone upgraded it with "better sounding" parts.

~Tom