[Texbychoice]

In the process of replacing all the electrolytic capacitors in my Onkyo M505.  Precautionary due to age.  Four new 18,000 uF power supply caps installed.  Caps on each channel amplifier board will be next.  Noticed a fuller and more open sound immediately.  Having read various writings, pro and con, about capacitor "burn-in" wonder if anyone here has first hand experience?     

[Speedskater]

With new power supply capacitors, it should only take a matter of a few minutes. And it's about background hum and noise.
* * * * * * * *
While other in circuit electrolytic capacitors might impact bass response.

[Folsom]

I've never heard an electrolytic that wasn't 90% of it's changes within 20hrs.

Well except BlackGates, N/NX types and depending on location.

Minutes? 

[rollo]

I've never heard an electrolytic that wasn't 90% of it's changes within 20hrs.

Well except BlackGates, N/NX types and depending on location.

Minutes? 
[/quo

I am with you on this. However I disagree somewhat as to time frame. I have yet to meet a cap that did not take less than 100 hours to settle in.


charles

[Texbychoice]

Interesting that Speedskater comment about bass response on the money.  Replaced all Electrolytics on the amplifier boards with Nichicon audio grade.  All the old caps removed were Nichicon.  The M-505 is considered direct coupled so did not know what to expect.  Initial listening impression, maybe a small bit improvement in vocals. Maybe a bit more detail.  Running a separately amplified Sub so quite surprised at the Bass difference.  Tight and clean with depth.  That alone made the effort worthwhile.

[avahifi]

Do any if you guys actually use a capacitor meter to measure the before and after values of the parts you are replacing?

Do you install matched pairs, channel to channels?

Capacitor values are not the numbers printed on them, they are the values they actually measure.

A 5% tolerance 10uF cap can be anything between 9.5 and 10.5.  If you end up with a 9.5 in one channel and 10.5 in the other, you now have a 10% difference between the channels. You will probably hear that depending upon where in the circuit the mismatch occurs.

Changing parts without measuring is like running tire tests without a pressure gauge and balancer.

[Texbychoice]

As a matter of fact, every capacitor removed was measured.  All were within 10% of marked value.  All originals were Nichicon 85 Deg C rated. 

All new caps, Nichicon 85 Deg C 20% tolerance, were measured and within 5% of marked value.  Measured values of each new cap matched between the channels.  A large mismatch might be noticeable.  I heard a significant difference in bass, yet the removed components measured withing spec.  Certainly something happening during in circuit operation.  Maybe the new caps will settle in and return to the old cap sound?  Fun hobby, isn't it?

However, the circuit designer of a mass produced high quality product has to account for normal part tolerance across all active and passive components.  An audio amplifier handles a pretty small bandwidth and well designed should not be so sensitive that small component variations cause problems. You can go down the rabbit hole rapidly by hand matching every measurable characteristic of every component.  You step into the realm of 2nd, 3rd, 4th order effects and beyond.  More power to anyone willing to foot the bill for that level of effort.  Absolutely support individual choice, whatever floats your boat.  After 100 hours of use, everything will have changed anyway.  Commentary from the perspective of performing worst case design analyses on far more complex things than audio amplifiers.       

[Speedskater]

Tolerance problems can be important in EQ sections and feedback loops. But signal coupling capacitors and big power supply electrolytics will be very tolerance forgiving.