Aha, but there are other big advantages of a choke. (1) It stores energy (an esthetically pleasing feat), and (2) it reduces the peak charging current of the subsequent capacitor using the stored energy. It also reduces line harmonic production due to the lower peak charging current.
This all (in a properly designed power supply) reduces garbage on the output of the power supply.
The penalty of course is the cost and size of the choke over a resistor, and the requirement for a higher secondary voltage feeding the choke to compensate for it's voltage drop.
TINSTAAFL
I’ve been trying to work out my views on this, and reasons for them. I’ve gone between PP and SE over the years, and finally ended up with SE. I know why and I’ll try and explain it. I’m a professional musician and whatever you say anecdotally about musicians e.g. they have lousy systems, listen on the kitchen radio etc., they do for the most part have well developed aural skills, particularly classical and jazz musicians (I was both). Some aren’t fussy about their systems because they don’t need to be – they play live enough and additionally have music going round in their heads a lot. Musicians on Desert Island Discs sometimes ask to take the scores, not recordings. Some musicians, however, are very critical of their sound systems.
Frequently we read that people who like SE do so for “subjective” reasons, and this has nothing to do with measurement. But aural skills CAN be measured. It’s not a level playing field. Try this aural training test for instance (it’s been on audio forums recently) and see how you score:
Tonedeaf Test: Test your musical skills in 6 minutes!
So what does this measure? As far as I can see it measures musical memory and the ability to recognise musical intervals and notes. In other words, the ability not just to “hear” music but to “recognise” the timbre and notes within it. Musicians with good aural skills score over 80. I know my own aural ability is pretty good, and it had to be as a jazz bassist who relied entirely on a good ear in my career. I did pretty well in the Royal Academy of Music for aural training and was put in the top level together with Sir Simon Rattle. I scored 86.1 on the test above.
Does this mean anything in terms of how I listen to an audio system? Well, I can only give my view on this which is that I’ve always listened critically for two things –
1) Clarity of inner parts in complex textures
2) Timbre and tone of acoustic instruments
Reasons are understandable enough – I want to hear all the notes and I want the instrumental sounds to be as natural and faithful as possible. My aural training and skills mean that my musical memory for acoustic sounds is good – I retain the timbre of instruments well in my memory. And I can also recognise individual parts and what notes they’re playing. This is no different from any other professional musician with good aural skills.
So how does this work with PP and SE? I preferred PP for quite a while because the textures were clearer – more audible parts in complex music. I then reached a plateau as my SE builds got better where I couldn’t choose easily - PP had better textures but SE had better timbre and tone on a number of acoustic sounds, particularly the voice where the difference in the naturalness and harmonic structure of the sound was clearly audible. As I further optimised my SE builds the clarity improved and since the timbre was always better, this became the preference.
The PP and SE systems I compared were obviously not the same, but they were quite similar. They both had 4P1L drivers and Lundahl LL1660 interstages and 2a3 or 300b outputs with O-Netics OPTs. My most recent SE build has PSE 4P1L outputs and is a step up in clarity. My conclusions were that an optimised PP design (diff. pairs over constant current sinks without coupling caps or cathode bypass caps) is probably a cleaner sound unless the SE circuit is heavily optimised (good iron, no coupling caps or cathode bypass caps). But the gains in timbre and tone with an optimised SE amp were sufficiently persuasive to tip the balance. This, as I see it, is all to do with the harmonic spectrum and absence of crossover distortion which have been discussed a number of times.
The reasons above may apply to other professional musicians – I don’t know too much about that, and I’m only presuming that they also attach importance to timbre and clarity of parts. The same may be equally true of non-musicians who listen critically for the same things. Beyond that, I do know that others listen for quite different things – soundstage, bass slam, dynamics and so forth. How they evaluate PP and SE on those parameters I don’t know – I don’t listen critically for these elements as others do. It also depends a lot on whether the music is electronic or acoustic. One thing I know for certain is that the listener is part of the feedback system in audio. And I think we can be more imaginative in evaluating the role of the listener than just declaring that “it’s all subjective”.
Frequently we read that people who like SE do so for “subjective” reasons, and this has nothing to do with measurement. But aural skills CAN be measured. It’s not a level playing field. Try this aural training test for instance (it’s been on audio forums recently) and see how you score:
Tonedeaf Test: Test your musical skills in 6 minutes!
So what does this measure? As far as I can see it measures musical memory and the ability to recognise musical intervals and notes. In other words, the ability not just to “hear” music but to “recognise” the timbre and notes within it. Musicians with good aural skills score over 80. I know my own aural ability is pretty good, and it had to be as a jazz bassist who relied entirely on a good ear in my career. I did pretty well in the Royal Academy of Music for aural training and was put in the top level together with Sir Simon Rattle. I scored 86.1 on the test above.
Does this mean anything in terms of how I listen to an audio system? Well, I can only give my view on this which is that I’ve always listened critically for two things –
1) Clarity of inner parts in complex textures
2) Timbre and tone of acoustic instruments
Reasons are understandable enough – I want to hear all the notes and I want the instrumental sounds to be as natural and faithful as possible. My aural training and skills mean that my musical memory for acoustic sounds is good – I retain the timbre of instruments well in my memory. And I can also recognise individual parts and what notes they’re playing. This is no different from any other professional musician with good aural skills.
So how does this work with PP and SE? I preferred PP for quite a while because the textures were clearer – more audible parts in complex music. I then reached a plateau as my SE builds got better where I couldn’t choose easily - PP had better textures but SE had better timbre and tone on a number of acoustic sounds, particularly the voice where the difference in the naturalness and harmonic structure of the sound was clearly audible. As I further optimised my SE builds the clarity improved and since the timbre was always better, this became the preference.
The PP and SE systems I compared were obviously not the same, but they were quite similar. They both had 4P1L drivers and Lundahl LL1660 interstages and 2a3 or 300b outputs with O-Netics OPTs. My most recent SE build has PSE 4P1L outputs and is a step up in clarity. My conclusions were that an optimised PP design (diff. pairs over constant current sinks without coupling caps or cathode bypass caps) is probably a cleaner sound unless the SE circuit is heavily optimised (good iron, no coupling caps or cathode bypass caps). But the gains in timbre and tone with an optimised SE amp were sufficiently persuasive to tip the balance. This, as I see it, is all to do with the harmonic spectrum and absence of crossover distortion which have been discussed a number of times.
The reasons above may apply to other professional musicians – I don’t know too much about that, and I’m only presuming that they also attach importance to timbre and clarity of parts. The same may be equally true of non-musicians who listen critically for the same things. Beyond that, I do know that others listen for quite different things – soundstage, bass slam, dynamics and so forth. How they evaluate PP and SE on those parameters I don’t know – I don’t listen critically for these elements as others do. It also depends a lot on whether the music is electronic or acoustic. One thing I know for certain is that the listener is part of the feedback system in audio. And I think we can be more imaginative in evaluating the role of the listener than just declaring that “it’s all subjective”.
That's not bad at all. Over on Audio Asylum DIY/Tube where there are quite a few engineers there were a few scores in the 60s and a lot in the 70s.
Audio Asylum Thread Printer
On the Music Lane forum where there are quite a few professional musicians scores were in the 80s with a couple of 90s.
Audio Asylum Thread Printer
Last edited:
Andy,
You make a number of interesting points and observations. Regarding supposed 'golden eared' listeners, I've long thought that such listeners have probably learned to simply be more observant of various auditory phenomena and psycho-acoustic effects, not that they have better measurable hearing acuity than the general populace, which they typically don't. I suspect that most of we audiophiles have simply learned to be more observant of reproduced sound than the general populace. Which could then provide one possible explanation for why, over time, our audio systems, which once gave us pleasure, often eventually come to disappoint us. Although, it could simply be that people with obsessive-compulsive type behavior are naturally drawn to being audiophiles
I suppose, there two primary objective advantages for PP operation over the same given circuit implemented as SE - aside from the superior PP power output efficiency for power amps. The first advantage is that PP operation greatly cancels the even-order distortion products, while somewhat increasing the odd-order products in exchange. The other primary advantage is that PP circuits produce less dynamic loading of the power supply, effectively, increasing supply rejection.
Subjectively, the difference in relative production of even-order versus odd-order distortion products is thought to produce exactly the general sound character you describe. Just in case you may have missed it, the below link leads to a Nelson Pass authored paper from 2008 partly addresses this very subject. In it, Nelson reports that (I'm paraphrasing here) in his experience, some listeners prefer the warmth and smoothness of an even-order dominated THD spectrum, while others like the crispness, as you say, of an odd-order dominated spectrum. If I correctly recall, Nelson concludes that the best sound seems to occur when H2 and H3 are of about equal magnitude while H4 and higher are greatly absent.
http://firstwatt.com/pdf/art_dist_fdbk.pdf
You make a number of interesting points and observations. Regarding supposed 'golden eared' listeners, I've long thought that such listeners have probably learned to simply be more observant of various auditory phenomena and psycho-acoustic effects, not that they have better measurable hearing acuity than the general populace, which they typically don't. I suspect that most of we audiophiles have simply learned to be more observant of reproduced sound than the general populace. Which could then provide one possible explanation for why, over time, our audio systems, which once gave us pleasure, often eventually come to disappoint us. Although, it could simply be that people with obsessive-compulsive type behavior are naturally drawn to being audiophiles
I suppose, there two primary objective advantages for PP operation over the same given circuit implemented as SE - aside from the superior PP power output efficiency for power amps. The first advantage is that PP operation greatly cancels the even-order distortion products, while somewhat increasing the odd-order products in exchange. The other primary advantage is that PP circuits produce less dynamic loading of the power supply, effectively, increasing supply rejection.
Subjectively, the difference in relative production of even-order versus odd-order distortion products is thought to produce exactly the general sound character you describe. Just in case you may have missed it, the below link leads to a Nelson Pass authored paper from 2008 partly addresses this very subject. In it, Nelson reports that (I'm paraphrasing here) in his experience, some listeners prefer the warmth and smoothness of an even-order dominated THD spectrum, while others like the crispness, as you say, of an odd-order dominated spectrum. If I correctly recall, Nelson concludes that the best sound seems to occur when H2 and H3 are of about equal magnitude while H4 and higher are greatly absent.
http://firstwatt.com/pdf/art_dist_fdbk.pdf
Last edited:
Andy,
You make a number of interesting points and observations. Regarding supposed 'golden eared' listeners, I've long thought that such listeners have probably learned to simply be more observant of various auditory phenomena and psycho-acoustic effects, not that they have better measurable hearing acuity than the general populace, which they typically don't. I suspect that most of we audiophiles have simply learned to be more observant of reproduced sound than the general populace. Which could then provide one possible explanation for why, over time, our audio systems, which once gave us pleasure, often eventually come to disappoint us. Although, it could simply be that people with obsessive-compulsive type behavior are naturally drawn to being audiophiles
I suppose, there two primary objective advantages for PP operation over the same given circuit implemented as SE - aside from the superior PP power output efficiency for power amps. The first advantage is that PP operation greatly cancels the even-order distortion products, while somewhat increasing the odd-order products in exchange. The other primary advantage is that PP circuits produce less dynamic loading of the power supply, effectively, increasing supply rejection.
Subjectively, the difference in relative production of even-order versus odd-order distortion products is thought to produce exactly the general sound character you describe. Just in case you may have missed it, the below link leads to a Nelson Pass authored paper from 2008 partly addresses this very subject. In it, Nelson reports that (I'm paraphrasing here) in his experience, some listeners prefer the warmth and smoothness of an even-order dominated THD spectrum, while others like the crispness, as you say, of an odd-order dominated spectrum. If I correctly recall, Nelson concludes that the best sound seems to occur when H2 and H3 are of about equal magnitude while H4 and higher are greatly absent.
http://firstwatt.com/pdf/art_dist_fdbk.pdf
I was a professional musician for the first part of my career but a performance psychologist for the last 25 years, largely working with and researching professional musicians. One interesting bit of research by Shuter-Dyson into hemispheres showed that amateur or non-musicians had more right-brained responses to music - textures, sound itself, pictures and so on. Professional musicians became increasingly left-brained through their careers, with more critical and analytic responses. Shame really to lose that euphonic enjoyment of sound itself! But professional musicians as I was saying do listen critically for timbre, intervals and so on and they score higher on aural tests for such things.
Much as you describe! The question I'd ask, though, is what audiophiles are critically listening for? I'm not sure it's timbre and individual musical parts. Agree that it's not about hearing acuity itself but how we are processing it.
Last edited:
It would be great if there was a test to find those who could actually hear better than others, besides the ordinary audiology testing. I would assume that having outstanding hearing acuity would be a prerequisite for being a gifted listener, but since there's no reliable test, as far as I know, then listening preferences are entirely subjective. Even in the field of wine tasting, there doesn't appear to be any reliable double blind testing to differentiate those who have extraordinary sensory ability and those who do not. You'd think with all that money at stake in the wine business, there'd be some kind of real science behind the testing, but that doesn't appear to be the case.
I know for me, my acuity is awful, so I don't pretend to be a very good listener.
Thanks for your observations. The big question is whether the PP amps ran in class A, as the SE amps did; that would take crossover distortion out of the equation.
In a properly designed AB amplifier, the crossover distortion is somewhere between negligible and zero. I don't know where the idea that Class A has any bearing on this originated. Take a look at the distortion residuals of some of these examples.
Audio Research Reference 75 power amplifier Measurements | Stereophile.com
Audio Research Reference 150 power amplifier Measurements | Stereophile.com
Conrad-Johnson LP125M monoblock power amplifier Measurements | Stereophile.com
Music Reference RM-200 Mk.II power amplifier Measurements | Stereophile.com
http://www.stereophile.com/content/ypsilon-aelius-monoblock-power-amplifier-measurements
So,
Different power supplies should sound the same if they are "competent" in design....DC is DC...
Interesting way of just not trying anything else.
Why build SE at all? You know you don't like it. You say you are trying to find something better, however you stick to the same ideas.
Like Einstein said trying the same thing over and over expecting a different outcome is madness!
Regards
M. Gregg
So,
Different power supplies should sound the same if they are "competent" in design....DC is DC...
Interesting way of just not trying anything else.
Why build SE at all? You know you don't like it. You say you are trying to find something better, however you stick to the same ideas.
Like Einstein said trying the same thing over and over expecting a different outcome is madness!
Regards
M. Gregg
Listening to my latest SE as we speak, just finished and working on the bench, sounds OK on the bench so far - will report when I plug it into the main system.
However I see little point in crippling it with a marginal power supply so its got lots of capacitance after a C-L stage. As you say - if you do it right its just DC with enough capacity to prevent sag. Thats all thats needed and its a simple engineering and modeling exercise.
Shoog
If you have a choke on the power supply,
Try putting a 0.1uF polypropylene 1000V across it just for fun, if it makes a difference..whats wrong?
You can easily take it off again..
Regards
M. Gregg
Last edited:
Why don't we have a "like" button on this message board?!?! Thanks for posting that! I did only average on the test, but I'm going to take it again when I'm fresh to see if it improves.
Wow, sorry for your opinion. I think it's valuable. I only scored average, but I will take it again. If you really did score 100, then you are a very rare person!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.