As one astute observer said,
"If you can HEAR the difference between components in an audio chain but not MEASURE it, then you are measuring the wrong things."
Just what those things ARE is the realm of the never-ending quest.
Exactly
In audio first 1% harmonic distortion was considered 'inaudible', then 0.1%, then 0.01%, then 0.001%, and now it's generally accepted that the figure for inaudibility is ~0.0001%. That's an 80dB improvement over a six or so decades.
When digital recordings in the form of CDs arrived, wow & flutter was of the order of ~0.001% and considered immeasurable. Although some people heard 'digital-ness' in an instant and never liked the sound of CDs, it took decades for mainstream audio design engineers to accept that something was wrong in the 'perfect' world of digital. Jitter in the digital world is 'wow & flutter' in the timing circuits, and it's now generally considered that jitter needs to be in parts per billion to be inaudible.
However an electromagnetic wave is simply that. If it's measurable then regardless of fundamental or harmonics, the effects of the component can also be measured.
Now like every experiment and analysis it all comes down to understanding all the variables - including any RFI etc that the capacitor could be receiving due to it's construction vs a different construction.
Too many people measure two things and then give an statement based on those results. However as with all science - publish and be damned. This includes the results and the experiment.. so people can see if you've incorporated RFI or vibration, thermal etc.
In the end, until it's in your amp.. with all the characteristics and with the specific shielding, thermals etc you'll not know how it behaves.. and it will vary as your amp does. Therein lies the Achilles Heel to both sides of the argument - the component as tested is just that - as tested. Using outside the test configuration and expecting the same results is often akin to madness.. likewise testing in the amp to compare and not recording the other variables is also akin to madness.
Now like every experiment and analysis it all comes down to understanding all the variables - including any RFI etc that the capacitor could be receiving due to it's construction vs a different construction.
Too many people measure two things and then give an statement based on those results. However as with all science - publish and be damned. This includes the results and the experiment.. so people can see if you've incorporated RFI or vibration, thermal etc.
In the end, until it's in your amp.. with all the characteristics and with the specific shielding, thermals etc you'll not know how it behaves.. and it will vary as your amp does. Therein lies the Achilles Heel to both sides of the argument - the component as tested is just that - as tested. Using outside the test configuration and expecting the same results is often akin to madness.. likewise testing in the amp to compare and not recording the other variables is also akin to madness.
There actually might be a significant amount of distortion in the average good recording; but not all distortion is unpleasant.
Caused by something in the recording chain?
Or introduced by the actual source instrument?
Caused by the electronics. Whilst I take your point, I wouldn't consider the sound of an instrument, it's timbre etc., to be distortion.
There's also a number of other factors:
* resonance
* rise time
* derating
Overview of market areas and technology: https://www.wima.de/wp-content/uploads/media/WIMA-Application-Guide.pdf
More interesting for me - the data sheets for MKP10, FP1 and FP4 in one handy spot: Pulse capacitors - WIMA – Competence in Capacitors
I'm using a 1u MKP10 630V specifies 500V/usec, the FKP1 630V I'm also using in 0.22 size specifies 11000V/usec. It would be interesting to understand if using two in parallel increases harmonic noise based on the differing rise times introducing symmetrical variation of the wave form. Not a problem with audio I suspect but digital and square waves used in dacs/valve dacs may also cause differences - especially once you start derating (it suggests 1.38% derate oner +75degC for AC). Perhaps over-analysis..
In the end "sounding good" vs "perfect signal in signal out" is surjective. Burn-in is therefore surjective.. as has been said - it's almost a religious argument (I will not go into this to protect everyone's sanity). Personally I think if you're going to say it needs burn in - then it's fine to say to your ears it got better with time but don't attribute it to 'burn in' unless you can back that up by a peer-reviewed and repeatable experiment.
To me burn-in is what manufacturers do for high precision instrumentation to ensure that the component will not fail or deviate from the specification in the specified temperature, voltage and current when operated 24/7.
Now if the device needs an temperature controlled oven/cooling to maintain precision for example then I could understand it.. but I've not seen any amp manufacturers build in temperature control into their amps which leads me to the validity of 'burn in' and manufacturers specs they're selling. Tubes.. I can understand with gassing etc. Caps perhaps fixing themselves with self heal.. or internal heat melting/disapating the wax into the paper?
My point here is that when the amp leaves the factory that then needs 'burn-in' then you're getting an amp that simply hasn't been QC'd.. the manufacturer is guessing/gambling that the components will behave as per the intended design after a number of hours playing by the customer.
So yes - I can see where burn-in can occur.. but at the same time I feel without isolated component experiment that demonstrates a change over time, that it's highly likely there's more than just one parameter in effect. Especially with thermal characteristics with the rest of the components changing as the heat soaks through the amp..
@Jazz Man - I was listening to Vivaldi yesterday, thinking some of the stringed bass instruments sounds a little chubby at the bottom end. However that's the sound of long low tension strings being slow bowed connected to a ported wood box over long notes.
Listening to Rickie Lee Jones, the double bass being finger plucked had a different sound, taught and deep.
* resonance
* rise time
* derating
Overview of market areas and technology: https://www.wima.de/wp-content/uploads/media/WIMA-Application-Guide.pdf
More interesting for me - the data sheets for MKP10, FP1 and FP4 in one handy spot: Pulse capacitors - WIMA – Competence in Capacitors
I'm using a 1u MKP10 630V specifies 500V/usec, the FKP1 630V I'm also using in 0.22 size specifies 11000V/usec. It would be interesting to understand if using two in parallel increases harmonic noise based on the differing rise times introducing symmetrical variation of the wave form. Not a problem with audio I suspect but digital and square waves used in dacs/valve dacs may also cause differences - especially once you start derating (it suggests 1.38% derate oner +75degC for AC). Perhaps over-analysis..
In the end "sounding good" vs "perfect signal in signal out" is surjective. Burn-in is therefore surjective.. as has been said - it's almost a religious argument (I will not go into this to protect everyone's sanity). Personally I think if you're going to say it needs burn in - then it's fine to say to your ears it got better with time but don't attribute it to 'burn in' unless you can back that up by a peer-reviewed and repeatable experiment.
To me burn-in is what manufacturers do for high precision instrumentation to ensure that the component will not fail or deviate from the specification in the specified temperature, voltage and current when operated 24/7.
Now if the device needs an temperature controlled oven/cooling to maintain precision for example then I could understand it.. but I've not seen any amp manufacturers build in temperature control into their amps which leads me to the validity of 'burn in' and manufacturers specs they're selling. Tubes.. I can understand with gassing etc. Caps perhaps fixing themselves with self heal.. or internal heat melting/disapating the wax into the paper?
My point here is that when the amp leaves the factory that then needs 'burn-in' then you're getting an amp that simply hasn't been QC'd.. the manufacturer is guessing/gambling that the components will behave as per the intended design after a number of hours playing by the customer.
So yes - I can see where burn-in can occur.. but at the same time I feel without isolated component experiment that demonstrates a change over time, that it's highly likely there's more than just one parameter in effect. Especially with thermal characteristics with the rest of the components changing as the heat soaks through the amp..
@Jazz Man - I was listening to Vivaldi yesterday, thinking some of the stringed bass instruments sounds a little chubby at the bottom end. However that's the sound of long low tension strings being slow bowed connected to a ported wood box over long notes.
Listening to Rickie Lee Jones, the double bass being finger plucked had a different sound, taught and deep.
Last edited:
Yes! For over 30 years!
Solder joints needs 2 days to settle in. Many Audiohile customers of mine complained after receiving the repaired or/and modded components about harsh and uneven sound. I remember a customer who phoned me after 3 days (he had complained at the first day about the harsh mids) and told me that it was like magic cos now it sounds wonderful and he was very thankful..
Some come out of the closet after many years which can be hard I believe. A gay man once told me jokingly he wishes he had been born as a lesbian woman as the straight men then would like him.
This is however a man with balls of titanium stating his experiences regarding this subject. Not pro or anti but hats off sir!! May the electrons protect you against the flak you will endure.
Желаю удачи!
This is however a man with balls of titanium stating his experiences regarding this subject. Not pro or anti but hats off sir!! May the electrons protect you against the flak you will endure.
Желаю удачи!
Last edited:
And your religion is to attack everything that is not measurable like a little child. Forgetting the fact that your stupid equipment cannot even identify if the midrange driver of a 3-way speaker is paper or metal... whereas the village idiot can tell you in 1 min. (only with listening to it) FACT!
When all people would be like you we were still in the dark ages for sure.
Listen to the new opamps or D-amps, you can hear the progress in distortion figures very clearly.
