Re: local regs
Koinichiwa,
Now if you use remote sensing (BTW, you CAN do something suitably similar with many a regulator circuit) more extreme circuits make sense.
I commented to that effect in the original thread about turning the Jung Reg into a PCB to replace the 78XX/79XX series of regulators. If you do that the basic limitations of the implementation of the 78XX/79XX device in the generic euiment being modified will not allow the potential of the regulator to be realised.
However, as remarked to ALW, if you start down that "complex reg with remote sensing" route and throw major money at PSU (usually in a new design, modding existing gear with 3-Pin regs in place often lacks the space anyway), not go for a Shunt Reg?
The actual advantages of a shunt reg are quite dramatic and the output impedance, noise and bandwidth can be made equal to the Jung reg.
As ALW remarks, the AC BEHAVIOUR is critical. And series regs have a very poor AC behaviour, as they CANNOT SINK CURRENT. A push-pull reg or a shunt reg CAN sink as much current as it supplies or more.
Sayonara
Koinichiwa,
janneman said:
Now if you use remote sensing (BTW, you CAN do something suitably similar with many a regulator circuit) more extreme circuits make sense.
I commented to that effect in the original thread about turning the Jung Reg into a PCB to replace the 78XX/79XX series of regulators. If you do that the basic limitations of the implementation of the 78XX/79XX device in the generic euiment being modified will not allow the potential of the regulator to be realised.
However, as remarked to ALW, if you start down that "complex reg with remote sensing" route and throw major money at PSU (usually in a new design, modding existing gear with 3-Pin regs in place often lacks the space anyway), not go for a Shunt Reg?
The actual advantages of a shunt reg are quite dramatic and the output impedance, noise and bandwidth can be made equal to the Jung reg.
As ALW remarks, the AC BEHAVIOUR is critical. And series regs have a very poor AC behaviour, as they CANNOT SINK CURRENT. A push-pull reg or a shunt reg CAN sink as much current as it supplies or more.
Sayonara
full circle
Sayonara,
We come full circle here. One of my reasons to try the opamp supply is that it can both sink & source current, which a reg with a pass transistor cannot. The other reason being lower parts count etc.
By the way, the remote sensing is not as complex (in execution) as you imply. Simply run two extra wires with the load connections, that's all.
I must believe you on your word on the shunt reg, having no experience with that.
Jan Didden
Sayonara,
We come full circle here. One of my reasons to try the opamp supply is that it can both sink & source current, which a reg with a pass transistor cannot. The other reason being lower parts count etc.
By the way, the remote sensing is not as complex (in execution) as you imply. Simply run two extra wires with the load connections, that's all.
I must believe you on your word on the shunt reg, having no experience with that.
Jan Didden
lateral career moves
'The actual advantages of a shunt reg are quite dramatic and the output impedance, noise and bandwidth can be made equal to the Jung reg."
Which will require a high speed opamp, remote sensing and just about as many parts for the same performance.
Why does a power supply for an audio circuit need to sink and source current? The power supply terminals of the devices you are regulating are not going to sink and source current. Preloading the regulator with resistor would allow for this, but I really can't think of a situation with a regulator for line level circuits where the load is going to sink and source currents, except when creating an active ground terminal. The signal currents should be small enough for preloading to handle this case as well. What desired feature next, cruise control?
'The actual advantages of a shunt reg are quite dramatic and the output impedance, noise and bandwidth can be made equal to the Jung reg."
Which will require a high speed opamp, remote sensing and just about as many parts for the same performance.
Why does a power supply for an audio circuit need to sink and source current? The power supply terminals of the devices you are regulating are not going to sink and source current. Preloading the regulator with resistor would allow for this, but I really can't think of a situation with a regulator for line level circuits where the load is going to sink and source currents, except when creating an active ground terminal. The signal currents should be small enough for preloading to handle this case as well. What desired feature next, cruise control?
Re: lateral career moves
Koinichiwa,
Of course. I did not say that shunt reg with remote sensing etc would be less complex (actually, adding the cascaded current source for feeding the shunt makes it more complex), I merely remarked that it is "better", IMHO.
That said, limited power shunt regulators can be made VERY SMALL and located directly at the supplied pin. Their AC behaviour is IMHO better than series regs.
A low measured output impedance means nothing in itself. A snap of what the reg does when faced with sharp step of reducing the load current by 50% is more interesting, as is the reaction when AC is injected into the rail.
Series regs usually are not particulary pretty in either case and these transient problems as well as their inability to sink current make the output capacitors so critical (sonically). With a shunt-reg all you need to do is to ensure stability, which needs usually a fairly small value capacitor, low ESR types tend to be applicable with no stability penalty.
Sayonara
Koinichiwa,
Fred Dieckmann said:
Of course. I did not say that shunt reg with remote sensing etc would be less complex (actually, adding the cascaded current source for feeding the shunt makes it more complex), I merely remarked that it is "better", IMHO.
That said, limited power shunt regulators can be made VERY SMALL and located directly at the supplied pin. Their AC behaviour is IMHO better than series regs.
A low measured output impedance means nothing in itself. A snap of what the reg does when faced with sharp step of reducing the load current by 50% is more interesting, as is the reaction when AC is injected into the rail.
Series regs usually are not particulary pretty in either case and these transient problems as well as their inability to sink current make the output capacitors so critical (sonically). With a shunt-reg all you need to do is to ensure stability, which needs usually a fairly small value capacitor, low ESR types tend to be applicable with no stability penalty.
Sayonara
TO MY EARS TOO...
Hi,
Sure is, you lose the series element.
IME, they have very good bass performance...always a good indicator of low Zo at low frequencies.
What I don't quite understand is that the industry hardly ever uses any...
Or, do they?
Cheers,😉
Hi,
Sure is, you lose the series element.
IME, they have very good bass performance...always a good indicator of low Zo at low frequencies.
What I don't quite understand is that the industry hardly ever uses any...
Or, do they?
Cheers,😉
Re: lateral career moves
Koinichiwa,
ARE YOU SURE? I mean REALLY SURE?
Drive a nice pulse into a nice line cable (or a speaker) and have a look what is coming back....
I can think of a few. To keep it simple, wherever a device will drive a reactive load.
Sayonara
Koinichiwa,
Fred Dieckmann said:
ARE YOU SURE? I mean REALLY SURE?
Drive a nice pulse into a nice line cable (or a speaker) and have a look what is coming back....
Fred Dieckmann said:
I can think of a few. To keep it simple, wherever a device will drive a reactive load.
Sayonara
Re: Re: lateral career moves
The only real advantage of a shunt regulator might be better PSRR from the unregulated supply. The ability to source and sink current is no better than a preloaded series regulator. Transient response can be made the same for either and doesn't depend on series or shunt topology, but things like the open loop gain and phase margin. A lot of people love to talk about the mystical properties of the shunt regulator with no engineering data to back them up. a regulator is an amplifier with a big DC offset and the same design principles apply.
"Drive a nice pulse into a nice line cable (or a speaker) and have a look what is coming back...."
The output capacitor on the regulator will be sinking the current for cable reflections these are very high frequecies. You are going to have a regulator fast enough to deal with these. If you going to drive pulses down cables you terminate them with a resistor to match the cable impedance and it all looks resistive then. This is really an issue for digital signals and not audio frequency analog. That's people put capacitors on the outputs of regulators for high frequency transients. The caps are doing most of the work on a digital circuit. There just are not that many reactive loads for line level circuits and even headphones are a pretty tame load. I still don't see the need for a push pull type regulator on a line level circuit. Even on an amplifier supply with a good sized capacitor on the output of the regulator.
Kuei Yang Wang said:
The only real advantage of a shunt regulator might be better PSRR from the unregulated supply. The ability to source and sink current is no better than a preloaded series regulator. Transient response can be made the same for either and doesn't depend on series or shunt topology, but things like the open loop gain and phase margin. A lot of people love to talk about the mystical properties of the shunt regulator with no engineering data to back them up. a regulator is an amplifier with a big DC offset and the same design principles apply.
"Drive a nice pulse into a nice line cable (or a speaker) and have a look what is coming back...."
The output capacitor on the regulator will be sinking the current for cable reflections these are very high frequecies. You are going to have a regulator fast enough to deal with these. If you going to drive pulses down cables you terminate them with a resistor to match the cable impedance and it all looks resistive then. This is really an issue for digital signals and not audio frequency analog. That's people put capacitors on the outputs of regulators for high frequency transients. The caps are doing most of the work on a digital circuit. There just are not that many reactive loads for line level circuits and even headphones are a pretty tame load. I still don't see the need for a push pull type regulator on a line level circuit. Even on an amplifier supply with a good sized capacitor on the output of the regulator.
Re: Re: Re: lateral career moves
Koinichiwa,
Yup. Exactly. And a series regulator operates in SE (as does a shunt), meaning it is very limited in it's ability to cope with AC. That is why PP Amplifiers where made in the first place.
This is one of the reasons why I recommend bidirectional output Regs or Shunt.
With a shunt reg the key advantages are in the arrangement of the (AC) current loops and the constant current draw from the supply.
Sayonara
Koinichiwa,
Fred Dieckmann said:
Yup. Exactly. And a series regulator operates in SE (as does a shunt), meaning it is very limited in it's ability to cope with AC. That is why PP Amplifiers where made in the first place.
This is one of the reasons why I recommend bidirectional output Regs or Shunt.
With a shunt reg the key advantages are in the arrangement of the (AC) current loops and the constant current draw from the supply.
Sayonara
Single ended or single minded.
"Yup. Exactly. And a series regulator operates in SE (as does a shunt), meaning it is very limited in it's ability to cope with AC. That is why PP Amplifiers where made in the first place."
Nonsense. Tell that to people who build single ended amplifiers. I guess Mr. Pass doesn't understand either? Push pull amps were designed for the increased efficiency. I use singled ended mosfet follower power amps with no problem and great sonics as do many others.
Your just digging yourself a deeper hole the further you go with this.......... Did you ever stop to think how few push pull audio supplies are actually out there. If the are such a good idea and so needed; it is suprising that they are not used more. Maybe you have the push pull regulator market wide open for yourself. Who am I to argue with such a visionary? I eagerly wait for your design if you plan to share it with us on the forum........
"Yup. Exactly. And a series regulator operates in SE (as does a shunt), meaning it is very limited in it's ability to cope with AC. That is why PP Amplifiers where made in the first place."
Nonsense. Tell that to people who build single ended amplifiers. I guess Mr. Pass doesn't understand either? Push pull amps were designed for the increased efficiency. I use singled ended mosfet follower power amps with no problem and great sonics as do many others.
Your just digging yourself a deeper hole the further you go with this.......... Did you ever stop to think how few push pull audio supplies are actually out there. If the are such a good idea and so needed; it is suprising that they are not used more. Maybe you have the push pull regulator market wide open for yourself. Who am I to argue with such a visionary? I eagerly wait for your design if you plan to share it with us on the forum........
Re: Single ended or single minded.
Koinichiwa,
I myself use SE Amplifiers. However, virtually ALL such amplifiers operate in "shunt mode", not in "follower mode".
Yes. Did you ever stop to think how few Speakers exist that deserve the moinker "High Fidelity"? Just because something is not usually "done" does not mean it has no merit.
Unlike you I have worked much with the various regulator topologies. I find that there are both objective (measured) and subjective (sound) differences. Maybe you would wish to actually empirically evaluate the subject instead of just arguning from a theoretical viewpoint that is by far to abstract and simplified to be usable in reality?
It has already been shared. Using a normal Op_Amp without added series pass device IS a "push-pull" regulator. Note that I still feel local shunts are a better choice in most cases.
Sayonara
Koinichiwa,
Fred Dieckmann said:
I myself use SE Amplifiers. However, virtually ALL such amplifiers operate in "shunt mode", not in "follower mode".
Fred Dieckmann said:
Yes. Did you ever stop to think how few Speakers exist that deserve the moinker "High Fidelity"? Just because something is not usually "done" does not mean it has no merit.
Unlike you I have worked much with the various regulator topologies. I find that there are both objective (measured) and subjective (sound) differences. Maybe you would wish to actually empirically evaluate the subject instead of just arguning from a theoretical viewpoint that is by far to abstract and simplified to be usable in reality?
Fred Dieckmann said:
It has already been shared. Using a normal Op_Amp without added series pass device IS a "push-pull" regulator. Note that I still feel local shunts are a better choice in most cases.
Sayonara
local reg
Hmm, Ahh, right, ehh, Sayonara, I do find Freds arguments a bit more convincing, to be honest. I was thinking that an opamp with a standing current of say 10mA delivering a signal current of say 20mA peak would need a bidirectional supply. But if you check it out, each signal half is delivered from each supply polarity, so looking at each supply the current drawn is unidirectional. Inductive loads would change that, but I don't think low-level circuits have that.
To make sure this is not a totally useless post, what means:
Koinichiwa?
Sayonara?
Jan Didden
Hmm, Ahh, right, ehh, Sayonara, I do find Freds arguments a bit more convincing, to be honest. I was thinking that an opamp with a standing current of say 10mA delivering a signal current of say 20mA peak would need a bidirectional supply. But if you check it out, each signal half is delivered from each supply polarity, so looking at each supply the current drawn is unidirectional. Inductive loads would change that, but I don't think low-level circuits have that.
To make sure this is not a totally useless post, what means:
Koinichiwa?
Sayonara?
Jan Didden
Re: local reg
Koinichiwa,
Fine.
ANY reactive load will "change that", as ultimatly it means that energy is stored somewhere and released delayed. This is stuff that steady state testing does not and cannot account for. For example my sources are mostly analogue, with MC catridges you get vert steep slope impulses at appreciable levels.
How a system reacts to these pulses has much to do with their subjective audibility or perhaps better subjective intrusiveness. Regualtir topologies with no relaibly measurable steadt state behaviour sound still different.
Play around a little, is sometimes helps to break through the blocks set up in most EE's heads at school & university. Whenever you come across behaviour that "should not happen" you find the world is a little more complex than it would seem. I agree the divergence only shows up at the "margins", but that is exactly where one works in High Performance Audio.
From "Basic Japanese phrases"
Day-time greeting = koinichiwa (koh-ee-nee-chee-wah)
Bye! = sayonara (sah-yoh-nah-rah)
Some more for those interrested:
Evening-time greeting = konbanwa (koh-nn-bah-nn-wah)
I'm sorry = gomen nasai (goh-meh-nn-nah-sah-ee)
Thank you = arrigato (ah-ree-gah-toh)
Thanks = domo (doh-moh)
Big thank you = domo arrigato
Thank you very much (present tense)= domo arrigato gozaimas (goh-zah-ee-mahss)
Thank you very much (past tense) = domo arrigato gozaimashita
Sayonara
Koinichiwa,
janneman said:
Fine.
janneman said:
ANY reactive load will "change that", as ultimatly it means that energy is stored somewhere and released delayed. This is stuff that steady state testing does not and cannot account for. For example my sources are mostly analogue, with MC catridges you get vert steep slope impulses at appreciable levels.
How a system reacts to these pulses has much to do with their subjective audibility or perhaps better subjective intrusiveness. Regualtir topologies with no relaibly measurable steadt state behaviour sound still different.
Play around a little, is sometimes helps to break through the blocks set up in most EE's heads at school & university. Whenever you come across behaviour that "should not happen" you find the world is a little more complex than it would seem. I agree the divergence only shows up at the "margins", but that is exactly where one works in High Performance Audio.
janneman said:
From "Basic Japanese phrases"
Day-time greeting = koinichiwa (koh-ee-nee-chee-wah)
Bye! = sayonara (sah-yoh-nah-rah)
Some more for those interrested:
Evening-time greeting = konbanwa (koh-nn-bah-nn-wah)
I'm sorry = gomen nasai (goh-meh-nn-nah-sah-ee)
Thank you = arrigato (ah-ree-gah-toh)
Thanks = domo (doh-moh)
Big thank you = domo arrigato
Thank you very much (present tense)= domo arrigato gozaimas (goh-zah-ee-mahss)
Thank you very much (past tense) = domo arrigato gozaimashita
Sayonara
The money shot
They're at home in lots of places, a Naim environement is just one of them. I have plenty of other bits of kit around the house though. Much of it with technically better measured performance - none of it sounds remotely as good though 🙄
I note you've railed frequently at Naim designs, particularly power amps, all you have done in such circumstances is demonstrated your ignorance to me. You stated them as suffering all manner of distortions, all of which are notable by their abscence in the genuine article.
Note how I refrained during previous responses from veiled comments about any of the equipment you, or anyone else here uses. I respect your right to dislike mine, but find your inability, in almost everything you've written elsewhere, to resist having a little snipe at every opportunity does nothing for my respect of your opinions.
I was really trying to be open-minded too - the first rule of a salesman NEVER slag off the opposition.
And before you classify me under the 'Naim faithful' tag, you would do well to read my much of what I've posted elsewhere before doing so. I've been strongly critical too, the interesting thing is that the more I've learnt, the less critical I've become. It's easy to look at a circuit diagram and be critical. Unfortunately it's what you cannot see that's often more important.
But it doesn't by your criteria above, the PSU's are remote, on long cables, with often RC feeds to the circuits in question. Supply impedance for starters would not be a primary issue, yet the gains from PSU improvements are huge.
Do you know why - I do!
Andy.
They're at home in lots of places, a Naim environement is just one of them. I have plenty of other bits of kit around the house though. Much of it with technically better measured performance - none of it sounds remotely as good though 🙄
I note you've railed frequently at Naim designs, particularly power amps, all you have done in such circumstances is demonstrated your ignorance to me. You stated them as suffering all manner of distortions, all of which are notable by their abscence in the genuine article.
Note how I refrained during previous responses from veiled comments about any of the equipment you, or anyone else here uses. I respect your right to dislike mine, but find your inability, in almost everything you've written elsewhere, to resist having a little snipe at every opportunity does nothing for my respect of your opinions.
I was really trying to be open-minded too - the first rule of a salesman NEVER slag off the opposition.
And before you classify me under the 'Naim faithful' tag, you would do well to read my much of what I've posted elsewhere before doing so. I've been strongly critical too, the interesting thing is that the more I've learnt, the less critical I've become. It's easy to look at a circuit diagram and be critical. Unfortunately it's what you cannot see that's often more important.
But it doesn't by your criteria above, the PSU's are remote, on long cables, with often RC feeds to the circuits in question. Supply impedance for starters would not be a primary issue, yet the gains from PSU improvements are huge.
Do you know why - I do!
Andy.
"I myself use SE Amplifiers. However, virtually ALL such amplifiers operate in "shunt mode", not in "follower mode"."
I don't know what the hell "shunt mode" is. If it is single ended the amp is in parallel with the load from an AC standpoint with a high impedance to the connection to one of the supply rails. With an AC coupled output it can be a single rail. I don't know what ALL amplifiers means because there are plenty with follower outputs. It's a good way to get low output impedance without loop negative feedback. Maybe you need to look at a few before making statements about all amplifiers.
Single Ended Pure Class A Power Follower Amplifier
Designed by: Andrea Ciuffoli.
http://www.audiodesignguide.com/my/follower_99c_sch.gif
http://www.audiodesignguide.com/my/index.html
http://www.passlabs.com/pdf/seclassa.pdf
http://home.c2i.net/semfielec/dcmosamp.htm
http://invalid.ed.unit.no/~nilsst/electronics/single/construction.php3
http://www.pha.inecnet.cz/macura/audiopage.html
"Unlike you I have worked much with the various regulator topologies. I find that there are both objective (measured) and subjective (sound) differences. Maybe you would wish to actually empirically evaluate the subject instead of just arguning from a theoretical viewpoint that is by far to abstract and simplified to be usable in reality? "
You presume to know a hell of a lot about my experience..... You might actually want to read a few of my post before jumping in the deep end with such statements. There are several hundred to choose from, the earlier being under the harry haller member name. I have posted several simple circuit designs on the forom that were far from abstract and theoretical. I have built audio circuits for over 15 years (including commercial experience) with listening to them as the most important part of the design process. I have spent nearly as much time discussing subject listening results as technical matters. These include the sound of transformers, diodes, resistors, capacitors, solder,fet vs. bipolar circuits, the influence of RF, and mechanical damping of passive components and ICs. I have posted many simple circuits (a passive preamp simple enough or you) and passive component recommendations as well. I have spent a lot of time lately discussing the necessity of examining BOTH the subjective and the measurable in serious audio design.
By the way, how is something both abstract and simplified. I never found basic amplifier theory to be abstract. I have seldom been accused over simplifying things but often of providing too much information. I can't seem to please anybody it seems........
I don't know what the hell "shunt mode" is. If it is single ended the amp is in parallel with the load from an AC standpoint with a high impedance to the connection to one of the supply rails. With an AC coupled output it can be a single rail. I don't know what ALL amplifiers means because there are plenty with follower outputs. It's a good way to get low output impedance without loop negative feedback. Maybe you need to look at a few before making statements about all amplifiers.
Single Ended Pure Class A Power Follower Amplifier
Designed by: Andrea Ciuffoli.
http://www.audiodesignguide.com/my/follower_99c_sch.gif
http://www.audiodesignguide.com/my/index.html
http://www.passlabs.com/pdf/seclassa.pdf
http://home.c2i.net/semfielec/dcmosamp.htm
http://invalid.ed.unit.no/~nilsst/electronics/single/construction.php3
http://www.pha.inecnet.cz/macura/audiopage.html
"Unlike you I have worked much with the various regulator topologies. I find that there are both objective (measured) and subjective (sound) differences. Maybe you would wish to actually empirically evaluate the subject instead of just arguning from a theoretical viewpoint that is by far to abstract and simplified to be usable in reality? "
You presume to know a hell of a lot about my experience..... You might actually want to read a few of my post before jumping in the deep end with such statements. There are several hundred to choose from, the earlier being under the harry haller member name. I have posted several simple circuit designs on the forom that were far from abstract and theoretical. I have built audio circuits for over 15 years (including commercial experience) with listening to them as the most important part of the design process. I have spent nearly as much time discussing subject listening results as technical matters. These include the sound of transformers, diodes, resistors, capacitors, solder,fet vs. bipolar circuits, the influence of RF, and mechanical damping of passive components and ICs. I have posted many simple circuits (a passive preamp simple enough or you) and passive component recommendations as well. I have spent a lot of time lately discussing the necessity of examining BOTH the subjective and the measurable in serious audio design.
By the way, how is something both abstract and simplified. I never found basic amplifier theory to be abstract. I have seldom been accused over simplifying things but often of providing too much information. I can't seem to please anybody it seems........
Re: The money shot
Koinichiwa,
Fine. I used to work as Engineer in a London 2nd Hand HiFi Shop. There was enough coming through my hands (with all sorts of others). And Naims gear does not only not measure well (which has not much of relevance I will redily agree), it's sound also leaves much to be desired.
I know enough people who like "Naim Sound", I find it tiring, unpleasant and lacking in exactly what Naim always claimed as a forte', namely musicality, as in sounding like music.
Let's see, Naim Poweramps and TIM - plenty of it, truckloads in fact. The input filter does not do enough to prevent it, only with a Naim preamp with severely ristricted HF response becomes the whole system moderatly acceptable WRT TIM.
Naim poweramps and crossover distortion - present and accounted for (quiescent current starved output stage).
Naim poweramps and high order harmonic distortion plus IMD - present and accounted for.
I do not know if you are the designer for Naim or not (nor do I care). Naims circuits have a number of specific "features", which represent themselves in certain behaviours (like the almost absent PSRR in preamp circuits and poor transient performance in poweramps).
My main point with this "sniping" was actually to suggest that in the context of Naim gear the benefit of improving the PSU is disproportionally larger as it is with many other circuits, that have fewer ideosyncracies.
Based on my experiences in the context of SE valve circuitry (which by neccesity is as sensitive to powersupplies) - yes. And I normally avoid active regulation as much as possible - I find it much easier to get the neccesary behaviour of the Powersupplies by using passive methodes. I will agree that this is only a matter of "taste" and very extensive active regualtion circuits MAY be as good (I have yet to find one though).
The only reason for me to be interested in PSU regulators is the fact that much of the equipment I need to play music & movies is based on solid state. But even there I tend to prefer an approach that is as "passive" as possible.
Sayonara
Koinichiwa,
ALW said:
Fine. I used to work as Engineer in a London 2nd Hand HiFi Shop. There was enough coming through my hands (with all sorts of others). And Naims gear does not only not measure well (which has not much of relevance I will redily agree), it's sound also leaves much to be desired.
I know enough people who like "Naim Sound", I find it tiring, unpleasant and lacking in exactly what Naim always claimed as a forte', namely musicality, as in sounding like music.
ALW said:
Let's see, Naim Poweramps and TIM - plenty of it, truckloads in fact. The input filter does not do enough to prevent it, only with a Naim preamp with severely ristricted HF response becomes the whole system moderatly acceptable WRT TIM.
Naim poweramps and crossover distortion - present and accounted for (quiescent current starved output stage).
Naim poweramps and high order harmonic distortion plus IMD - present and accounted for.
ALW said:
I do not know if you are the designer for Naim or not (nor do I care). Naims circuits have a number of specific "features", which represent themselves in certain behaviours (like the almost absent PSRR in preamp circuits and poor transient performance in poweramps).
My main point with this "sniping" was actually to suggest that in the context of Naim gear the benefit of improving the PSU is disproportionally larger as it is with many other circuits, that have fewer ideosyncracies.
ALW said:
Based on my experiences in the context of SE valve circuitry (which by neccesity is as sensitive to powersupplies) - yes. And I normally avoid active regulation as much as possible - I find it much easier to get the neccesary behaviour of the Powersupplies by using passive methodes. I will agree that this is only a matter of "taste" and very extensive active regualtion circuits MAY be as good (I have yet to find one though).
The only reason for me to be interested in PSU regulators is the fact that much of the equipment I need to play music & movies is based on solid state. But even there I tend to prefer an approach that is as "passive" as possible.
Sayonara
Koinichiwa,
Interestingly, non of these circuits are commercial, the SE Amplifiers that are actually being manufacturerd with minimal exceptions all operate with outputs that amounts to common cathode/emitter/source (shunt) mode. Moreover, so do most DIY Amp's.
And once one observes the distortion spectrum and the load interaction of a common anode/collector/drain (aka follower) mode circuit and actually listens to the resulting sound it becomes clear why.
Let me ask you then:
"Have you ever compared regulator circuits with broadly the same static behaviour (bandwidth, impedance), where one circuit has SE Follower outputs, the other SE Shunt outputs and the third SEPP outputs?"
Good for you, I have done so since the early 1980's. Are we now at the point "I got more experience than you!"?
And? I have been arguing the same point also since the 1980's. What's new pussycat?
Abstraction is the process of simplifying a given phenomena to those variables considered relevant. Thus an abstratc is a simplified expression of reality.
I agree, using abstratc and simplified in the same sentence is an unneccesary repetition.
Sayonara
Fred Dieckmann said:
Interestingly, non of these circuits are commercial, the SE Amplifiers that are actually being manufacturerd with minimal exceptions all operate with outputs that amounts to common cathode/emitter/source (shunt) mode. Moreover, so do most DIY Amp's.
And once one observes the distortion spectrum and the load interaction of a common anode/collector/drain (aka follower) mode circuit and actually listens to the resulting sound it becomes clear why.
Fred Dieckmann said:
Let me ask you then:
"Have you ever compared regulator circuits with broadly the same static behaviour (bandwidth, impedance), where one circuit has SE Follower outputs, the other SE Shunt outputs and the third SEPP outputs?"
Fred Dieckmann said:
Good for you, I have done so since the early 1980's. Are we now at the point "I got more experience than you!"?
Fred Dieckmann said:
And? I have been arguing the same point also since the 1980's. What's new pussycat?
Fred Dieckmann said:
Abstraction is the process of simplifying a given phenomena to those variables considered relevant. Thus an abstratc is a simplified expression of reality.
I agree, using abstratc and simplified in the same sentence is an unneccesary repetition.
Sayonara
koinichiwa Choo Baka-chan
" And I normally avoid active regulation as much as possible - I find it much easier to get the neccesary behaviour of the Powersupplies by using passive methodes. "
And yet you feel qualified to argue about active power supply regulator topologies........
"Let's see, Naim Poweramps and TIM - plenty of it, truckloads in fact. The input filter does not do enough to prevent it, only with a Naim preamp with severely ristricted HF response becomes the whole system moderatly acceptable WRT TIM.
Naim poweramps and crossover distortion - present and accounted for (quiescent current starved output stage).
Naim poweramps and high order harmonic distortion plus IMD - present and accounted for."
So much for subjective listening test. I would think someone this knowledgeable about amp design and measurement could follow my technical discussions.
"I do not know if you are the designer for Naim or not (nor do I care). Naims circuits have a number of specific "features", which represent themselves in certain behaviours (like the almost absent PSRR in preamp circuits and poor transient performance in poweramps).
"My main point with this "sniping" was actually to suggest that in the context of Naim gear the benefit of improving the PSU is disproportionally larger as it is with many other circuits, that have fewer ideosyncracies."
All the more reason that Andy might have a desire to design good power supplies and a good testbed for their evaluation. Op amps which have very large PSRRs at audio frequencies still are very sensitive to supply quality. Why don't you take a pot shot at the Passlabs Pearl phono preamp which also has a single ended topology with low PSRR? I guess that is a hack design also. This forum would be a good place to tell Mr. Pass how to redesign it to a topology you find expectable.
I am curious if you even know who the original designer of the Naim products was, if you think Andy is a designer for them now.
Why would someone build an after market power supply upgrade for a product from a company he worked for? I think Naim would frown on that. Do you know anything about the people you feel so qualified to second guess. I have talked to most of the designers of the high end electronics I own and know a little more about them than just their names. P.S. Andy has a website if you have the slightest curiosity about him.
http://www.alw.audio.dsl.pipex.com/
I am glad it is just not me that you think doesn't know anything. I am starting to feel privileged to belong to that group, though.
Ookiosewada,
Fred-san
" And I normally avoid active regulation as much as possible - I find it much easier to get the neccesary behaviour of the Powersupplies by using passive methodes. "
And yet you feel qualified to argue about active power supply regulator topologies........
"Let's see, Naim Poweramps and TIM - plenty of it, truckloads in fact. The input filter does not do enough to prevent it, only with a Naim preamp with severely ristricted HF response becomes the whole system moderatly acceptable WRT TIM.
Naim poweramps and crossover distortion - present and accounted for (quiescent current starved output stage).
Naim poweramps and high order harmonic distortion plus IMD - present and accounted for."
So much for subjective listening test. I would think someone this knowledgeable about amp design and measurement could follow my technical discussions.
"I do not know if you are the designer for Naim or not (nor do I care). Naims circuits have a number of specific "features", which represent themselves in certain behaviours (like the almost absent PSRR in preamp circuits and poor transient performance in poweramps).
"My main point with this "sniping" was actually to suggest that in the context of Naim gear the benefit of improving the PSU is disproportionally larger as it is with many other circuits, that have fewer ideosyncracies."
All the more reason that Andy might have a desire to design good power supplies and a good testbed for their evaluation. Op amps which have very large PSRRs at audio frequencies still are very sensitive to supply quality. Why don't you take a pot shot at the Passlabs Pearl phono preamp which also has a single ended topology with low PSRR? I guess that is a hack design also. This forum would be a good place to tell Mr. Pass how to redesign it to a topology you find expectable.
I am curious if you even know who the original designer of the Naim products was, if you think Andy is a designer for them now.
Why would someone build an after market power supply upgrade for a product from a company he worked for? I think Naim would frown on that. Do you know anything about the people you feel so qualified to second guess. I have talked to most of the designers of the high end electronics I own and know a little more about them than just their names. P.S. Andy has a website if you have the slightest curiosity about him.
http://www.alw.audio.dsl.pipex.com/
I am glad it is just not me that you think doesn't know anything. I am starting to feel privileged to belong to that group, though.
Ookiosewada,
Fred-san
I don't know why I'm doing this, but...
A Naim power amp must not be used without a Naim pre - it's part of the manufacturers written criteria. The input signal is then limited by a very high quality low-pass filter within the preamp, that limits bandwidth to that which is audible. It does this without introducing any musically important anomalies (e.g. phase etc).
You also appear though to posess the hearing abilities of a bat - you can't hear above 20kHz, if you can, make some money by offering yourself to medical science.
Wrong again - it's an optimally biased output stage for the implementation. The amp does not suffer crossover distrotion.
Wrong again - hey 3 out of 3, not bad.
I see you use single-ended circuits, the easiest topology to get working correctly, since push-pull class AB brings about details that need addressing, that single ended doesn't even care about.
Andy.
A Naim power amp must not be used without a Naim pre - it's part of the manufacturers written criteria. The input signal is then limited by a very high quality low-pass filter within the preamp, that limits bandwidth to that which is audible. It does this without introducing any musically important anomalies (e.g. phase etc).
You also appear though to posess the hearing abilities of a bat - you can't hear above 20kHz, if you can, make some money by offering yourself to medical science.
Wrong again - it's an optimally biased output stage for the implementation. The amp does not suffer crossover distrotion.
Wrong again - hey 3 out of 3, not bad.
I see you use single-ended circuits, the easiest topology to get working correctly, since push-pull class AB brings about details that need addressing, that single ended doesn't even care about.
Andy.
Aaah I get it now. You worked in a shop that had no official Naim support.
You then got amplifiers, of unknown condition, about which you know nothing, then, like so many others, assume that they're simple designs that anyone can sort.
On the basis of that you then choose to make repeatedly innacurate descriptions of them, because of course you are a well known designer with a long pedigree of commercial successes behind you. You also on the basis of those cursory glances, managed to work out all the relevant details that make the circuit perform so well.
or not...
Andy.
Sorry about this everybody, there's just so much material here!
Interesting - would that be the PSRR that's actually better than most circuits at the higher reaches of the audio band that you're talking about? The same PSRR that's better than op-amps at high frequencies, by 10's of dB?
Poor transient performance by what measure. The amps do not suffer any transient-related distortions of any kind, when used within the design criteria. Great efforts are taken to ensure that the amplifiers are driven within their performance bounds.
You may (probably will) with 'illegal' inputs get performance anomalies, these do not happen in real life though.
And by the way, I do NOT design for Naim.
Andy.
Interesting - would that be the PSRR that's actually better than most circuits at the higher reaches of the audio band that you're talking about? The same PSRR that's better than op-amps at high frequencies, by 10's of dB?
Poor transient performance by what measure. The amps do not suffer any transient-related distortions of any kind, when used within the design criteria. Great efforts are taken to ensure that the amplifiers are driven within their performance bounds.
You may (probably will) with 'illegal' inputs get performance anomalies, these do not happen in real life though.
And by the way, I do NOT design for Naim.
Andy.
- Status
- Not open for further replies.