From TI Literature (SNAA031A):
"3.6.1 Parts Selection
In listening tests at TI's sound room evaluating different circuit components used in the LM4702 demo amplifier, there was one part whose negative effect on audible signal quality was undeniable. A DC blocking capacitor on the input of the LM4702 degraded sound quality. In multiple listening tests, with different participants and at various locations around the country, the negative effects of even the best film and foil polystyrene DC blocking input capacitors in the audio signal path was confirmed. It is therefore recommended that DC blocking capacitors not be used in the signal path for mid to high-end audio equipment. Where DC offset from another signal source may be a problem then the use of a DC servo circuit that keeps DC offset from appearing at the output of the amplifier is recommended."
...Didn't say if it showed up on an AP though, so still some hope for a comfortable/satisfying explanation.
"3.6.1 Parts Selection
In listening tests at TI's sound room evaluating different circuit components used in the LM4702 demo amplifier, there was one part whose negative effect on audible signal quality was undeniable. A DC blocking capacitor on the input of the LM4702 degraded sound quality. In multiple listening tests, with different participants and at various locations around the country, the negative effects of even the best film and foil polystyrene DC blocking input capacitors in the audio signal path was confirmed. It is therefore recommended that DC blocking capacitors not be used in the signal path for mid to high-end audio equipment. Where DC offset from another signal source may be a problem then the use of a DC servo circuit that keeps DC offset from appearing at the output of the amplifier is recommended."
...Didn't say if it showed up on an AP though, so still some hope for a comfortable/satisfying explanation.
If the ESR is linear over frequency it will not cause distortion. New polymer elco's are surprisingly linear. Mouser has 72 10 uF electrolytics with sub 100 mOhm ESR. A zero squared.
TI has great engineers. Their result is quite consistent with most critical listeners throughout time and place.
There is a science documentary on math statistics which goes into The Wisdom of the Crowd. How the average from the crowd is very very accurate though a few in the crowd are far off and wrong.
See Netflix
The distortion due to DA (Dielectric Absorption) will not show in sine wave tests. The symmetrical waveform will always average to zero in a THD instrument. IMO. Asymmetrical tests are needed and then you see the affect. A fast sample-hold instrument of transient waveforms (music) shows difference and how the differences are correlated to the sound. I have seen such data by others using equipment from UC Berkeley which reinforces my opinion.
THx-RNMarsh
Last edited:
The big IF. No the esr is not linear as it is composed of the film and bulk metal, term contact etc all in series. However, as the film losses gets lower/better the esr does also. Lower esr is always preferred. But, note at what freq the esr test is being made. Is that freq applicable to your use?
Normally, the self resonant freq is where the esr is the only remaining residual component. Large value of C often have a poorly defined self res freq due to the many contributing losses vs freq whereas better films are quit well defined. 100 miliOhm would not be low enough, though in some apps. For a comparison, the patented Multi-Cap made by REL-CAP of PP is in the micro-Ohms. Essentially only the bulk metal R between the two leads.
Esl is much higher with larger values, esp electrolytics. but, that another issue. The reason the self res freq is important is because above that freq it isnt behaving like a pure capacitor any more. The phase changes to 90 degrees inductive.
Why bother with all that. Just go direct-coupled and servo it, if you have to.
THx-RNMarsh
Last edited:
I think there are other things higher up on the Pareto.
For example room acoustics and speakers. The cap thing has been thrashed to death as has the science behind asymmetrical excitation signals. I use a servo to maintain the amp output DC offset at 0 V, but still have a DC blocking cap on the input. I’ve measured distortion (QA401) across all sorts of power levels with and without and can not detect any difference.
For example room acoustics and speakers. The cap thing has been thrashed to death as has the science behind asymmetrical excitation signals. I use a servo to maintain the amp output DC offset at 0 V, but still have a DC blocking cap on the input. I’ve measured distortion (QA401) across all sorts of power levels with and without and can not detect any difference.
TI do not have a listening room so we can possibly write this off as something that the app engineers came up with to give the punters what they wanted. Of course this will not stop people leaping on it as proof of their viewpoint
John, amplifier design has moved on from the 1970’s which is where your comfort zone is.
3rd rate high offset, high drift JFET input circuits, limited power bandwidth, feedback that changes with frequency, low loop gain (to deal with the lack of output inductor) etc may have been de riguer 30 years ago but no longer cut the mustard.
And this is precisely where attention is needed.
Modern amps and improved circuit techniques- be they VFA or CFA - deal gracefully with all these issues. Please do further study before making wild statements.
3rd rate high offset, high drift JFET input circuits, limited power bandwidth, feedback that changes with frequency, low loop gain (to deal with the lack of output inductor) etc may have been de riguer 30 years ago but no longer cut the mustard.
And this is precisely where attention is needed.
Modern amps and improved circuit techniques- be they VFA or CFA - deal gracefully with all these issues. Please do further study before making wild statements.
I'll have to remember to be very careful about anything you say in relation to things you do sell...
Bonsai, you are an insulting person. I don't know why. I make 'World Class' electronics, even today.
Those of you, like PMA, who rely on double blind testing, might accept even Mylar coupling caps, but I don't, primarily because we could hear one in one of my designs, 40 years ago. Most here are at the design level that I was at 40 years ago, when I was selecting Mylar caps for cost effective audio designs, but I was proven wrong. Then, after realizing how large and expensive GOOD caps were, I also went to servoing instead, but Richard was first. Most mid-fi designers, including those who design most of Parasound's products use coupling caps instead of servos. It is easier, but I don't.
By the way, I don't use an output inductor, because it can contribute to the sound as well. I used to use 1-2uH air wound chokes on outputs. No more! Again, it is a 'cheat', just like coupling caps, it makes design easier.
Fet input IC's for servos are CHEAP! Less than a dollar. What is the problem? A GOOD coupling cap will cost $10 or more.
Those of you, like PMA, who rely on double blind testing, might accept even Mylar coupling caps, but I don't, primarily because we could hear one in one of my designs, 40 years ago. Most here are at the design level that I was at 40 years ago, when I was selecting Mylar caps for cost effective audio designs, but I was proven wrong. Then, after realizing how large and expensive GOOD caps were, I also went to servoing instead, but Richard was first. Most mid-fi designers, including those who design most of Parasound's products use coupling caps instead of servos. It is easier, but I don't.
By the way, I don't use an output inductor, because it can contribute to the sound as well. I used to use 1-2uH air wound chokes on outputs. No more! Again, it is a 'cheat', just like coupling caps, it makes design easier.
Fet input IC's for servos are CHEAP! Less than a dollar. What is the problem? A GOOD coupling cap will cost $10 or more.
Last edited:
TI do not have a listening room so we can possibly write this off as something that the app engineers came up with to give the punters what they wanted. Of course this will not stop people leaping on it as proof of their viewpoint
A listening room is nice to have, but I first noticed the difference between MP3 and CD from that terrible VW stock sound system while driving; I subsequently cross-checked and confirmed this with a low-end Altec computer sound system. The MP3 is so horrible I never dared to try it with my "real" Hifi.
Of course, the diff between caps are more subtle, but still within reach of an OK pair of headphones. Mind you, I'm not trying to defend TI, never did any serious business with them.
Doing the same, I do not worry too much about film caps sonic impacts. But do not like electrolytics in the signal path when it can be avoided: i can hear the difference.
In the same time, i can sleep by night if there is an electrolytic somewhere, it is not day and night ;-)
The final quality of an amp depends on the care that is given to a multitude of small details, isn't-it ? It is an asymptotic curve.
Last edited:
Very possible. Obsolete part now and I don't recall they ever set the world alight other than easy for DIY build. Never had the pleasure of hearing an amp with them in.
- Status
- Not open for further replies.