I admit to not fully understand the affect of the switching freq. in digital circuit aspects but all modern computers use SMPS and they work at all kind of different very high-speed clocks. other than that a computer has at least one hard drive and a CD rom and both devices use electric motors in them.
my gut feeling (for lack of better term) says that if a modern computer can run using the usually cheap power supply that came with it, a DAC or CD player would do just fine. unless the fact they are working in high freq. and not ultra high freq. got something to do with it but that goes back to the specific switching speed that I understand nothing about
the aspect that I really find interesting is the power form factor correction abilities of these designs.
we all use AC power circuits, the thing is that because of capacitance and inductance some energy is stored in these loads and this creates a time difference between the two waveforms of the voltage and current - or in other words it makes them out of phase.
now I was thinking about amplifiers, class A and A/B mostly since I don't know enough about other types, it sounds to me like this phenomena will mess up the quality of signal reproduction on a circuit, no matter how well it is designed. not to say anything about the fact that many capacitors in the signal path will cause further disturbances resulting from the same effect.
the nice things about SMPS is that the power form can be corrected by using various technologies and some of them are not expensive to implement and that would "re-shape" the waveform coming out of the supply to the same shape as the input from the power outlet. this is a big advantage over a standard transformer based power supply which would cause phase issues right at the power supply before the amplifier itself can add to that problem on its own
the only real problem I see is noise resulting from oscillators, if the supply is not switching quickly enough I believe there will be some auditable effect but this noise can be filtered out rather easily and it's not like transformer based power supplies don't need filtering.
also, the more I think of it this means that even the voltage / current waveforms are out of phase at the power outlet in our homes! and it seems like it would be idea to correct this problem using enhanced SMPS.
my gut feeling (for lack of better term) says that if a modern computer can run using the usually cheap power supply that came with it, a DAC or CD player would do just fine. unless the fact they are working in high freq. and not ultra high freq. got something to do with it but that goes back to the specific switching speed that I understand nothing about
the aspect that I really find interesting is the power form factor correction abilities of these designs.
we all use AC power circuits, the thing is that because of capacitance and inductance some energy is stored in these loads and this creates a time difference between the two waveforms of the voltage and current - or in other words it makes them out of phase.
now I was thinking about amplifiers, class A and A/B mostly since I don't know enough about other types, it sounds to me like this phenomena will mess up the quality of signal reproduction on a circuit, no matter how well it is designed. not to say anything about the fact that many capacitors in the signal path will cause further disturbances resulting from the same effect.
the nice things about SMPS is that the power form can be corrected by using various technologies and some of them are not expensive to implement and that would "re-shape" the waveform coming out of the supply to the same shape as the input from the power outlet. this is a big advantage over a standard transformer based power supply which would cause phase issues right at the power supply before the amplifier itself can add to that problem on its own
the only real problem I see is noise resulting from oscillators, if the supply is not switching quickly enough I believe there will be some auditable effect but this noise can be filtered out rather easily and it's not like transformer based power supplies don't need filtering.
also, the more I think of it this means that even the voltage / current waveforms are out of phase at the power outlet in our homes! and it seems like it would be idea to correct this problem using enhanced SMPS.
space-cake said:I admit to not fully understand the affect of the switching freq. in digital circuit aspects but all modern computers use SMPS and they work at all kind of different very high-speed clocks. other than that a computer has at least one hard drive and a CD rom and both devices use electric motors in them.
my gut feeling (for lack of better term) says that if a modern computer can run using the usually cheap power supply that came with it, a DAC or CD player would do just fine. unless the fact they are working in high freq. and not ultra high freq. got something to do with it but that goes back to the specific switching speed that I understand nothing about
OK - but I didn't say that the CD player or DAC wouldn't work.
There is a huge difference between "working" and "sounding good". I haven't heard of anyone having great results by placing their DAC inside their PC. In fact, most DVD players use SMPS and this one of the reasons that most of them sound terrible.
space-cake said:now I was thinking about amplifiers, class A and A/B mostly since I don't know enough about other types, it sounds to me like this phenomena will mess up the quality of signal reproduction on a circuit, no matter how well it is designed. not to say anything about the fact that many capacitors in the signal path will cause further disturbances resulting from the same effect.
the nice things about SMPS is that the power form can be corrected by using various technologies and some of them are not expensive to implement and that would "re-shape" the waveform coming out of the supply to the same shape as the input from the power outlet. this is a big advantage over a standard transformer based power supply which would cause phase issues right at the power supply before the amplifier itself can add to that problem on its own
With amplifiers - you are only dealing with the audible range and it is relatively easy to filter out noise in such a limited range.
space-cake said:
With a CD player or DAC, you have to deal with the sub-sonic, audible and higher frequency ranges. Unnecessary noise in any of these areas will increase jitter and have a detrimental effect on the critical D/A process. I can't see a SMPS being easily tuned to satisfy all of these requirements.
CD Transport quality
Computer sound cards
Clock PSU
Electrical Jitter
Influence of PS on clock modules
Why does a CD player need a clean PS?
Low noise PS for oscillator high end SMPS for audio
In another thread, Fin asked for my 'help' in this one.
I'm not sure what I can add as I am no electronics expert and can only speak from hands on experience. For me, the SMPS that I have used with class-T amplifiers is clearly better than a linear supply in just about every aspect.
As regards what value of cap is best (on the output) I always give the same advice. Try a small one, say 47 UF, a large one, say 10K and something in between. We are talking about a comparatively small outlay for these three caps and you get to find out which sounds best (if any) for yourself. That's far better than any post you will read on these forums or any other!
Perhaps more importantly, you should try blocking any noise coming out of the SMPS (at both ends) by adding ferrite rings to each wire that is connected to the SMPS.
SMPS with DAC? Again, we can put forward all sorts of theories but there is no substiture for actually trying it! I have and I find it works well. The only proviso I will make is that it may sound different to a linear supply, not necessarily better or worse!
Finally, let me say that all my comments regarding SMPS are made with regard to the Skynet 8080 module.
The above is posted by the World Organisation to retain Sanity in DIY audio (WASADO) !
I'm not sure what I can add as I am no electronics expert and can only speak from hands on experience. For me, the SMPS that I have used with class-T amplifiers is clearly better than a linear supply in just about every aspect.
As regards what value of cap is best (on the output) I always give the same advice. Try a small one, say 47 UF, a large one, say 10K and something in between. We are talking about a comparatively small outlay for these three caps and you get to find out which sounds best (if any) for yourself. That's far better than any post you will read on these forums or any other!
Perhaps more importantly, you should try blocking any noise coming out of the SMPS (at both ends) by adding ferrite rings to each wire that is connected to the SMPS.
SMPS with DAC? Again, we can put forward all sorts of theories but there is no substiture for actually trying it! I have and I find it works well. The only proviso I will make is that it may sound different to a linear supply, not necessarily better or worse!
Finally, let me say that all my comments regarding SMPS are made with regard to the Skynet 8080 module.
The above is posted by the World Organisation to retain Sanity in DIY audio (WASADO) !
Well - I have been converted to SMPS to a certain degree - but just for class T-amps at this stage. Before reading what you, Michael Mardis and a few other have to say - I would never have even entertained the thought of bringing a SMPS anywhere near my Hi-Fi gear.
I suppose what I have been trying to get at is that if I, or most other novice DIYers with limited knowledge of how a SMPS can be optimised for a particular purpose, were to build a DAC or modify and add power supplies to a DAC or CD player - we would probably stand a much better chance of getting a good result with a traditional linear supply than with an off-the-shelf SMPS.
I suppose what I have been trying to get at is that if I, or most other novice DIYers with limited knowledge of how a SMPS can be optimised for a particular purpose, were to build a DAC or modify and add power supplies to a DAC or CD player - we would probably stand a much better chance of getting a good result with a traditional linear supply than with an off-the-shelf SMPS.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.