hi,
using a class a amp +/-45v@2amp will be more then 180watts per channel, we have 0.94% trafo efficiency and 1.4volts drop on the rectifiers
90/0.94 + 1.4=97.2volts; 90/97.2= 92.6% total efficiency for the power supply.
THEN WHAY TO USE A SHITCHING POWER SUPPLY?
Did I calculate wrong? What are the benefits of switching power supply when using class a amps?
Best regards
Williams
using a class a amp +/-45v@2amp will be more then 180watts per channel, we have 0.94% trafo efficiency and 1.4volts drop on the rectifiers
90/0.94 + 1.4=97.2volts; 90/97.2= 92.6% total efficiency for the power supply.
THEN WHAY TO USE A SHITCHING POWER SUPPLY?
Did I calculate wrong? What are the benefits of switching power supply when using class a amps?
Best regards
Williams
The following advantages just came to my mind:
- less weight, because less 'iron' is used (at the power transformer);
- if regulated, the output voltage should be stable like a rock;
- less space (volume) necessary, caused by smaller transformer and much smaller bulk capacitor(s);
- slightly higher efficiency possible, because of less copper losses (transformer), rectifier losses (Schottky diodes or even active synchronous rectification with MOSFETs), etc. under full load condition, which would be always the case for class-A.
The ladder case is just an estimation and could be the opposite case if not carefully designed.
- less weight, because less 'iron' is used (at the power transformer);
- if regulated, the output voltage should be stable like a rock;
- less space (volume) necessary, caused by smaller transformer and much smaller bulk capacitor(s);
- slightly higher efficiency possible, because of less copper losses (transformer), rectifier losses (Schottky diodes or even active synchronous rectification with MOSFETs), etc. under full load condition, which would be always the case for class-A.
The ladder case is just an estimation and could be the opposite case if not carefully designed.
Corax already gave already good points; I think less space with regulated rails for free sounds interesting. It's just a pity that these SMPS are not for DIY as even pcb-layout is critical and not as forgiving as audio. Next you need to wind chokes yourself...so interest is limited.
Have fun, Hannes
By the way, did I say SMPS can be a real noise-bomb if not carefully designed?
And use lethal voltages throughout?
Have fun, Hannes
By the way, did I say SMPS can be a real noise-bomb if not carefully designed?
And use lethal voltages throughout?
the efficiency of the transformer is (output current * output voltage)/(input current * input voltage). the difference between input wattage and output wattage is lost as heat. so you should measure the line (mains) voltage and current, and then measure the output voltage and current. these measurements will give you the total efficiency, not just the rectifier efficiency. the rectifiers are probably the MOST efficient part of your power supply. i think the measurements including the power transformer will give you VERY different results.
actually, back in 1982 or so there was a design in the works at APT corp, for an APT-2 amplifier, which would have been an improved APT-1 with a switching supply. this was when switching supplies were in ther infancy, and SMPS's had some exciting failure modes back then. one of them was diode failure causing the filter caps to either frag or launch......... it was just about a year after they decided not to build it right away that TI came out with SMPS driver chips with shutdown and protection modes that worked.
Yes, it still applies, but you have got ridden of the resistance due to the large transformer windings. This resistance(well, not only it, but it's the most importanat) limits the power you can draw form a transformer and also it's regulation so they are not negligible. In SMPS you get rid of it. Also you can correct the power factor easily. Furthermore, in order to have a correct comparison, one should put head to head a linear regulated supply against a switcher.
It's the line regulation where the SMPS wins.
SMPS varies the duty cycle to adj. for variation of input line voltage.
Linear unregulated will pass excess voltage on the class A stage where it is dissapated as extra heat.
But in actual use when the line voltage is known and fairly constant then an unreg linear is on par with a SMPS ~ 200 W.
As the power goes down from 200W or so the linear will take the lead.
SMPS varies the duty cycle to adj. for variation of input line voltage.
Linear unregulated will pass excess voltage on the class A stage where it is dissapated as extra heat.
But in actual use when the line voltage is known and fairly constant then an unreg linear is on par with a SMPS ~ 200 W.
As the power goes down from 200W or so the linear will take the lead.
I know Jeff Rowland Design Group uses SMPS in his newer
amps and they can be low noise, smaller than linear,
plus the benefit of being regulated already as already
mentioned.
I picked up several Astec MP units on the flee bay for
cheap, but not quite the voltages I was looking for, but
looking into possibilities of modifying or sending back to
Emerson and having modules swapped if price is resonable.
http://www.powerconversion.com/assets/mp_1227738018.pdf
amps and they can be low noise, smaller than linear,
plus the benefit of being regulated already as already
mentioned.
I picked up several Astec MP units on the flee bay for
cheap, but not quite the voltages I was looking for, but
looking into possibilities of modifying or sending back to
Emerson and having modules swapped if price is resonable.
http://www.powerconversion.com/assets/mp_1227738018.pdf
Hi all,
I think this is a very interesting issue the differences between the two kinds.
But we have to compare PS to PS stuff.
A none regulated Pwr Supply with a large amount of capas and adequate rectifiers block can deliver huge amounts of instant current.
The regulation factor of a good trafo is 95% and efficiency is much higher then 85%-90% so is not so bad, the supply is always a sine voltage.
Using a switching PS is a regulated servo topology and must be switched at higher freq. then 40KHz, components at this freq. and this pwr are not cheap
The RFI-EMI should be taken in consideration, very much and carefully.
I think the best is the middle way, using a good linear regulated PS for the VAS and drivers and in the final stage un regulated linear high current PS.
As prove of it, most of very high end does not use switching PS.
Best regards
Williams
I think this is a very interesting issue the differences between the two kinds.
But we have to compare PS to PS stuff.
A none regulated Pwr Supply with a large amount of capas and adequate rectifiers block can deliver huge amounts of instant current.
The regulation factor of a good trafo is 95% and efficiency is much higher then 85%-90% so is not so bad, the supply is always a sine voltage.
Using a switching PS is a regulated servo topology and must be switched at higher freq. then 40KHz, components at this freq. and this pwr are not cheap
The RFI-EMI should be taken in consideration, very much and carefully.
I think the best is the middle way, using a good linear regulated PS for the VAS and drivers and in the final stage un regulated linear high current PS.
As prove of it, most of very high end does not use switching PS.
Best regards
Williams
Williams Audio said:
Taking in the economies of scale SMPS can be very very inexpensive, looking at my desktop PC. It doesn't take too much copper and steel to make SMPS look attractive even on a much lower scale. Also to mention the cost of electricity savings over the running lifetime.
RE> EMI/RFI A number of people take very dynamic audio measurements with PC sound cards with cheap SMPS powered computers.
Hi all,
OK, if is so good show me what you have done!
Did you make a SPS for +/-45Vdc at 2A, with 30 A peaks capability for 50% duty cycle at a complex load?
If is that easy why not using it, if is that cheap let do it, if is that good every body should have it in their amps,
Not to forget PC SPS is made by millions of one design and the high currents are at low voltages, the sound blasters have a separate PS unit with on boards; High quality regulators that sense the noise and spikes and a servo loop cancel the noises.
So, if some body has a simple good SPS for a high pwr amp, I will be the first one to get in touch and build it
Any one can help?
Best regards
Williams
OK, if is so good show me what you have done!
Did you make a SPS for +/-45Vdc at 2A, with 30 A peaks capability for 50% duty cycle at a complex load?
If is that easy why not using it, if is that cheap let do it, if is that good every body should have it in their amps,
Not to forget PC SPS is made by millions of one design and the high currents are at low voltages, the sound blasters have a separate PS unit with on boards; High quality regulators that sense the noise and spikes and a servo loop cancel the noises.
So, if some body has a simple good SPS for a high pwr amp, I will be the first one to get in touch and build it
Any one can help?
Best regards
Williams
Williams Audio said:
RE SMPS and audio amps. It can and is being done. More in Pro audio but not too much in high end yet because of psycological hangovers.
Have a look around some more. SMPS technology is tried and true. No it's not that easy, but best understood by SMPS people who also design audio systems incl. speakers, amps, and DACs. Your specs are not real world 30A peaks but for how long ...not too many speakers need that. 2 ohm electrostatics is the only one that comes to mind.
Besides there is a proper PS forum at DIY with many folks doing it.
Hi all,
I see this combat is not balanced, but still, speakers of ribbon and ESL nature are the heavy loads, and for a long bass replay say at 30Hz, 50% duty cycle of loud volume will be need very high currents for long time, as you can see what is my occupation, be sure I know a thing or two about high pwr SPS at high voltage.
In any case take a look at VICORE there are exotic in this field.
In any case if you find some thing good, be kind and let me know.
Best regards and happy weekend
Williams
I see this combat is not balanced, but still, speakers of ribbon and ESL nature are the heavy loads, and for a long bass replay say at 30Hz, 50% duty cycle of loud volume will be need very high currents for long time, as you can see what is my occupation, be sure I know a thing or two about high pwr SPS at high voltage.
In any case take a look at VICORE there are exotic in this field.
In any case if you find some thing good, be kind and let me know.
Best regards and happy weekend
Williams