Dear Sirs,
I notice that a lot of efforts are done, for instance in preamps, in order to avoid couplind caps.
Are they so bad for sound ?
Opinions and suggestion of good film caps are very welcome.
Regards,
beppe
I notice that a lot of efforts are done, for instance in preamps, in order to avoid couplind caps.
Are they so bad for sound ?
Opinions and suggestion of good film caps are very welcome.
Regards,
beppe
Hi Beppe
By now you probably realise that asking questions rarely yields answers. And the answers are usually meaningless and conflicting unless you already have lots of first hand experience and understanding.
How bad are caps then? It seems very few systems exist with no caps in the signal path. If your main source is a cartridge, this automatically means replacing caps with transformers and trying to choose the lesser evil. Depending upon the system i use between 2 and 4 transformers in the signal path and find their distortions more musically acceptable than those of caps. Others will, no doubt, differ.
DC coupling is probably best but not easy to implement. The average textbook dc servo seems more harmful than the cap it replaces.
Of course, unless you use batteries there are always the caps in the PS. One way to minimise their flavour is to use small caps and large chokes, but this works well only in tube curcuits.
Would i say that removing a coupling cap is worth all the effort, expense and annoyances? By all means yes - listening to an entirely cap-free system is a great experience.
By now you probably realise that asking questions rarely yields answers. And the answers are usually meaningless and conflicting unless you already have lots of first hand experience and understanding.
How bad are caps then? It seems very few systems exist with no caps in the signal path. If your main source is a cartridge, this automatically means replacing caps with transformers and trying to choose the lesser evil. Depending upon the system i use between 2 and 4 transformers in the signal path and find their distortions more musically acceptable than those of caps. Others will, no doubt, differ.
DC coupling is probably best but not easy to implement. The average textbook dc servo seems more harmful than the cap it replaces.
Of course, unless you use batteries there are always the caps in the PS. One way to minimise their flavour is to use small caps and large chokes, but this works well only in tube curcuits.
Would i say that removing a coupling cap is worth all the effort, expense and annoyances? By all means yes - listening to an entirely cap-free system is a great experience.
analog_sa said:
Thanks a lot for a reply full of interesting points of discussion.
1) I am here first of all to gather valuable opinions from audio experts like you and I thank you for that.
I find the discussions always stimulating.
Nevertheless I understand that the answers usually do not come easy, beacuse a lot of parameters play.
2) Presently I listen to an humble power amp with no caps in the signal path. My opinion is that it is slightly let down by the preamp actually and it is not bad all in all.
My main/only source is a CD player.
Now that you talk about coupling transformers I read positive reviews of transformer coupled preamps with digital sources as well.
If they were not so expensive I would like to try one, one day.
3) Very interesting. Once I thought that transformers added great amount of distortion. But after having seen some really impressive distortion spectrum from a McIntosh amp I am not so sure.
Maybe the best transformers are the best thing for DC blocking.
My mind always go to Audio Tekne brand. Do you know it?
4) I see it very well.
5) Interesting.
6) Also with batteries some caps are needed in the PS to low down their impedance I think (at least with lead batteries).
7) In what sense? Could you tell me something about that experience ?
Thanks a lot.
Kind regards,
beppe
Hi Beppe
Sadly, this is not normally the case. The best interstage transformers cannot take any reasonable amount of dc current. I dare not let more than a single mV dc in the TVC after Thorsten's recommendations. If you want to use transformers to block dc you should look at balanced circuits where the dc component in the transformer primary is zero. Not easy to keep the balance though.
Or you may try building one of the servos from curldiy.com 🙂
Sadly, this is not normally the case. The best interstage transformers cannot take any reasonable amount of dc current. I dare not let more than a single mV dc in the TVC after Thorsten's recommendations. If you want to use transformers to block dc you should look at balanced circuits where the dc component in the transformer primary is zero. Not easy to keep the balance though.
Or you may try building one of the servos from curldiy.com 🙂
Coupling Capacitors - caps good or bad
.
1. Thank you very much Mr Beppe61. New interesting topic, again!
2. I do not mind using coupling caps. Always try to find best quality: Polypropylene.
- Comes from I often like to use only single ended power supply.
- They are easy and simple to make. +9 Volt Battery and 0 Volt=Ground
- Here talk about small transistors preamplifiers.
3. I do not mind building simple power amplifiers, too.
- With only single sided power supply.
- Here I must use output Electrolytic Capacitor.
- I use 4700uF Axial good quality caps for my 6-8 Ohm loudspeakers.
4. Regards from Mr Lineup
5. 😉
.
1. Thank you very much Mr Beppe61. New interesting topic, again!
2. I do not mind using coupling caps. Always try to find best quality: Polypropylene.
- Comes from I often like to use only single ended power supply.
- They are easy and simple to make. +9 Volt Battery and 0 Volt=Ground
- Here talk about small transistors preamplifiers.
3. I do not mind building simple power amplifiers, too.
- With only single sided power supply.
- Here I must use output Electrolytic Capacitor.
- I use 4700uF Axial good quality caps for my 6-8 Ohm loudspeakers.
4. Regards from Mr Lineup
5. 😉
beppe61:
The best way to know is to try it in a double blind test.
Prepare a test setup where a switch allows to include or by-pass 10 or more series-connected low-cost electrolytic capacitors in the wire coming from the preamplifier to the amplifier. If the switch produces any noise, then add an additional placebo switch capable of producing the same noise but without doing anything else.
Ask a friend to activate the switches randomly where you can't see him, and try to guess if the capacitors are or not bypassed.
I've read about a similar test with several dozens of low-cost op-amps connected in series, and the "audiophiles" failed nearly 50% of the times. Note that 50% is the probability of failing when guessing the flip of a coin.
The best way to know is to try it in a double blind test.
Prepare a test setup where a switch allows to include or by-pass 10 or more series-connected low-cost electrolytic capacitors in the wire coming from the preamplifier to the amplifier. If the switch produces any noise, then add an additional placebo switch capable of producing the same noise but without doing anything else.
Ask a friend to activate the switches randomly where you can't see him, and try to guess if the capacitors are or not bypassed.
I've read about a similar test with several dozens of low-cost op-amps connected in series, and the "audiophiles" failed nearly 50% of the times. Note that 50% is the probability of failing when guessing the flip of a coin.
Re: Coupling Capacitors - caps good or bad
Dear Mr. Lineup,
maybe some of my problems are just ramblings, but that is.
1)2) I prefer PP caps as well for signal coupling.
Batteries based PS are very interesting but a little complex to use.
I can understand that the noise is minimal in this case.
3) Interesting, Thanks for the info.
4) Kind regards,
beppe
lineup said:
Dear Mr. Lineup,
maybe some of my problems are just ramblings, but that is.
1)2) I prefer PP caps as well for signal coupling.
Batteries based PS are very interesting but a little complex to use.
I can understand that the noise is minimal in this case.
3) Interesting, Thanks for the info.
4) Kind regards,
beppe
Eva said:
Dear Miss.Eva,
1) Thank you for the very interesting procedure explanation.
But I would be very interested to know your opinion.
Are some caps better than others for coupling applications?
What do you use?
2) Very very interesting. Do you have any link to this article?
It should be a very interesting reading.
Thanks a lot for the extremely interesting information and advice.
Kind regards,
beppe
Dear Mr Beppe61,
Even some of the respectfull designers take this differently. Mr. John Curl seems to be allergic to caps (and also inductors).
His friend, Mr. Nelson Pass, uses caps, even in the output of his very updated power amps designs (ZV9), and one who build/heard it gives it good review about good sounding.
Some of the legendary amps like JLH have caps in the output node.
As our Mr. Lineup and Mrs. EVA mentioned, maybe you yourself should decided whether you should use or not using caps in your cct. But for myself, putting caps does more advantage than drawback (when solving problem like blocking DC offset). Trying to solve this kind of problem without caps will result in far more complex cct, usually not sounding better than putting 1 cap.
Kind Regards, lumanauw
Even some of the respectfull designers take this differently. Mr. John Curl seems to be allergic to caps (and also inductors).
His friend, Mr. Nelson Pass, uses caps, even in the output of his very updated power amps designs (ZV9), and one who build/heard it gives it good review about good sounding.
Some of the legendary amps like JLH have caps in the output node.
As our Mr. Lineup and Mrs. EVA mentioned, maybe you yourself should decided whether you should use or not using caps in your cct. But for myself, putting caps does more advantage than drawback (when solving problem like blocking DC offset). Trying to solve this kind of problem without caps will result in far more complex cct, usually not sounding better than putting 1 cap.
Kind Regards, lumanauw
lumanauw said:
Dear Mr. Lumanauw,
thank you so much for your kind and valuable reply.
1) Yes. I got a reply from Him telling what you state.
I do not know if He uses servos for offset nulling anyway.
2) Very interesting. Thanks.
Maybe the last word cauld be said using an oscilloscope and see for degradation of a let's say square wave?
3) To be honest and with your great and kind advice I have already decided for PP caps in the signal path.
4) We are at least in two then thinking the same.
Slightly off topic, I would like to inform you that I have very appreciated a your post in the thread about the Threshold NS10 line preamp (cap coupled by the way).
In that post you state that simpler circuits (less active devices in the signal path) have more chances to preserve the "harmonics content" of the musical signal.
Maybe I have not expressed your thought well but I have been impressed by this opinion, because this reflects in most ways my experience.
The most musical units I listened to were also very basic in their topology, at least concerning the signal amplification circuit.
The power supply instead were very well built and much more complex.
Could you correct my words and/or elaborate the concept?
It is a very interesting approach to the problem.
Thank you very much.
Kind regards,
beppe
Dear Mr.Beppe61,
1. First of all thank you very much for your kind words. I'm just nobody here, still a student here. You should pay attention more to what people like Mr. John Curl, Mr. Nelson Pass, Mr AKSA, Mr. MikeB, etc wrotes.
2. For me, 1 transistor is FAR MORE destructive than 1 capacitor.
If you use transistor, you put input signal to base, and take the output from emitor or collector. This output signal have degraded so much (from input signal), if you compared the same signal passing through 1 capacitor. In transistor story, you can aid this with feedback, but it is a story of paying back money by borrowing money from another guy.
1. First of all thank you very much for your kind words. I'm just nobody here, still a student here. You should pay attention more to what people like Mr. John Curl, Mr. Nelson Pass, Mr AKSA, Mr. MikeB, etc wrotes.
2. For me, 1 transistor is FAR MORE destructive than 1 capacitor.
If you use transistor, you put input signal to base, and take the output from emitor or collector. This output signal have degraded so much (from input signal), if you compared the same signal passing through 1 capacitor. In transistor story, you can aid this with feedback, but it is a story of paying back money by borrowing money from another guy.
Dear Mr. Lumanauw,
1) I did not want to flatter you. I was honest.
I found your opinion original and very valuable indeed.
Nevertheless I am extremely interested to top designers' thought as well, of course.
2) This is indeed a point: can the feedback cure the degradations caused to the signal by the active components?
This is "the question", after all.
Thank you so much.
Kind regards,
beppe
Bazukaz said:
Rod Elliott is a few times critisised for some minor details in a few of his many audio pages.
Still, most everyone can refer to his articles/projects
in most anything regarding audio designs and audio circuits.
- Elliott ESP http://sound.westhost.com/ - no doubt an institution in world of DIY Audio -
🙂
This is a part from his final 'Conclusion'. On capactitors.
lumanauw said:
Excellent point Lumanauw. Whatever one can say about Pass amps/preamps, neutral sounding is seldom a typical description. To me it seems that a fine mixture of harmonics and other colourations from both active and passive components contribute to a specific, attractive to many listeners sound.
The liberal, almost unnecessary sprinkling of electrolytics along the signal path in premium products such as the X0 may seem mysterious but i have no doubt they contribute to a particular, desired sonic signature.
In fact, i believe most of us are so used to capacitor distortion and colouration that the initial reaction to a capacitor-free signal path is almost a culture shock. The sound loses a diffuse quality of images, some midbass flab and a lot of electronic 'haze'.
I doubt that there is any question that eliminating capacitors from the signal path would be a good thing.
Before you begin designing a circuit, there are still questions to be resolved, however. There is a point that many people seem to find subtle, but it needs to be faced before going further. What does "in the signal path" mean?
If a capacitor is not in the signal path, then it doesn't matter what brand it is or what it costs. It need only meet the bare bones critera: capacitance and voltage rating. Beyond that, there's no point in worrying about it.
My philosophy is that nearly everything is in the signal path. A capacitor in the 12V supply used to drive relays is not, for instance, but if I can see signal--no matter how small--on a part on a scope, then I count it as being in the signal path. I specifically include the caps in the power supply and most anything in the circuit itself. The only exception I can think of offhand is something like a cap used to smooth the reference voltage for a current source...and even that needs to have good high frequency characteristics so that noise gets shunted to ground rather than showing up in your current source.
Note that Nelson Pass disagrees with me on this. His view, as I understand it, is that the signal path is the shortest trace from the input to the output. Anything off that direct line doesn't count. He specifically does not count the power supply.
I don't recall John Curl ever stating explicitly what he regards as the signal path, but I have gotten the sense that he feels everything counts.
As the old saying goes, you pays your money and you makes your choice.
If you feel that the power supply is not in the signal path, then you can use electrolytics without fear. If you think it does matter, then consider bypassing the electrolytics with some sort of film cap.
There's no question that caps color the sound, but by how much and is it better or worse than the alternatives. If this thread attracts much interest, it will quickly fill with dozens of posts touting level shifters, transformers, resistors (it is possible to level shift using resistors), etc. I'll stay out of it. I will simply note that after all these years, solid state equipment is still sold using phrases such as "tube-like sound" and--with the exception of bass extension/control--not the other way around. Look inside a piece of tube gear and what do you see? Scads of caps. Maybe one or two transformers. Something to think about.
One hint: Make sure you have sufficient current available to drive the cap.
Grey
Before you begin designing a circuit, there are still questions to be resolved, however. There is a point that many people seem to find subtle, but it needs to be faced before going further. What does "in the signal path" mean?
If a capacitor is not in the signal path, then it doesn't matter what brand it is or what it costs. It need only meet the bare bones critera: capacitance and voltage rating. Beyond that, there's no point in worrying about it.
My philosophy is that nearly everything is in the signal path. A capacitor in the 12V supply used to drive relays is not, for instance, but if I can see signal--no matter how small--on a part on a scope, then I count it as being in the signal path. I specifically include the caps in the power supply and most anything in the circuit itself. The only exception I can think of offhand is something like a cap used to smooth the reference voltage for a current source...and even that needs to have good high frequency characteristics so that noise gets shunted to ground rather than showing up in your current source.
Note that Nelson Pass disagrees with me on this. His view, as I understand it, is that the signal path is the shortest trace from the input to the output. Anything off that direct line doesn't count. He specifically does not count the power supply.
I don't recall John Curl ever stating explicitly what he regards as the signal path, but I have gotten the sense that he feels everything counts.
As the old saying goes, you pays your money and you makes your choice.
If you feel that the power supply is not in the signal path, then you can use electrolytics without fear. If you think it does matter, then consider bypassing the electrolytics with some sort of film cap.
There's no question that caps color the sound, but by how much and is it better or worse than the alternatives. If this thread attracts much interest, it will quickly fill with dozens of posts touting level shifters, transformers, resistors (it is possible to level shift using resistors), etc. I'll stay out of it. I will simply note that after all these years, solid state equipment is still sold using phrases such as "tube-like sound" and--with the exception of bass extension/control--not the other way around. Look inside a piece of tube gear and what do you see? Scads of caps. Maybe one or two transformers. Something to think about.
One hint: Make sure you have sufficient current available to drive the cap.
Grey
- Status
- Not open for further replies.