I am not sure I agree with that.
First, if you have a 1000 Watt switchmode power supply, but only use 200 Watts then the work is not over the top.
Second, if you use a CRC capacitor bank after the power supply the noise can be removed.
However I do agree, as a general rule switchmodes are not a good choice.
First, if you have a 1000 Watt switchmode power supply, but only use 200 Watts then the work is not over the top.
Second, if you use a CRC capacitor bank after the power supply the noise can be removed.
However I do agree, as a general rule switchmodes are not a good choice.
Last edited:
Banned
Joined 2002
OK, find me a few NICE amplifiers like pass labs or krell or bryston etc etc, that use SMPS in there amplifiers, ? find any NOOOO! because they are not ment or used for it, these people have put more time and engineering into it, thus why no one still does it.
This is about perceived value. The heavier the amp the more it's worth. Who's gonna pay good money for a featherweight amp? Ever notice the first thing they describe in an advertisement is the weight of the amp and the VA rating of the transformer. Then it's the number of output transistors along with the power rating of the devices.
You could use a SMPS and rail jumping scheme to reduce the number of output devices for a Crest-like amp with 1400 watts/ch in a 15 pound package but it's still only worth like $75-100 per pound.
Banned
Joined 2002
Im told that classA likes a supply with some resistance ( not the caps I suppose)
Maybe because it on all the time
Switch mode supply is the opposite
Maybe switch noise could make it hard on the sensible input jfets
Or even result in instability
But I know nothing about such things
#
Maybe because it on all the time
Switch mode supply is the opposite
Maybe switch noise could make it hard on the sensible input jfets
Or even result in instability
But I know nothing about such things
#
Last edited:
What about all of the high frequency noise that is already on the AC power lines.
If I follow your philosophy, then I better start building my own coal fired or nuclear powered plant to ensure noise free ac.
If you can remove high frequency noise that is already on the ac lines then I don't see why you can't remove noise from a switch mode power supply.
I never said this was a good idea. I only asked if people had tried this.
I am the type of person that likes to learn from first hand experience and not to blindly follow.
If you say the world is flat, I say why.
The only other thing I would like to say about this, is it is not good to make generalisations here.
You could easily spend a $1000 on components and make an amplifier that sounds worse than an amplifier that Nelson could build for $200, if you don't know what you are doing.
It's not so much what you use, but more importantly, how you implement it.
If I follow your philosophy, then I better start building my own coal fired or nuclear powered plant to ensure noise free ac.
If you can remove high frequency noise that is already on the ac lines then I don't see why you can't remove noise from a switch mode power supply.
I never said this was a good idea. I only asked if people had tried this.
I am the type of person that likes to learn from first hand experience and not to blindly follow.
If you say the world is flat, I say why.
The only other thing I would like to say about this, is it is not good to make generalisations here.
You could easily spend a $1000 on components and make an amplifier that sounds worse than an amplifier that Nelson could build for $200, if you don't know what you are doing.
It's not so much what you use, but more importantly, how you implement it.
Last edited:
Linn makes an amp with a switched power supply which seems to be quite favorably reviewed. From what I've heard- yes you can do it with a switched mode supply but you need a good one, and a good one will cost as much or more that a conventional supply. But it would be lighter and smaller.....
Last edited:
IIRC some Perreaux amps were also SMPS-based. Specially some of their best ones.
The well-regarded MiniWatt also uses a SMPS - and it's a tube amp with only 2.5 watts, so one assumes super-sensitive speakers will be used, ones that would tell tales of power supply noise if there were any.
The well-regarded MiniWatt also uses a SMPS - and it's a tube amp with only 2.5 watts, so one assumes super-sensitive speakers will be used, ones that would tell tales of power supply noise if there were any.
I think there is real merit in trying to improve the performance of switch mode power supplies for audio use. The big benefit I see is the high efficiency.
While I agree shunt regulators offer pretty spectacular performance, trying to implement one for a class A amplifier can be a real pain in terms of the amount of heat that is dissipated. The heatsinks required can be as large as those used in the amplifer itself.
If you look at some of the reviews these days on top quality pc power supplies (switch mode of course), you will notice they have improved tremendously.
Seasonic, while manufacturing their own brand of powersupplies, also manufacture power supplies for the best brands in the business. A lot of them are capable of 1000W, the only problem is they don't have the necessary output voltages for amplifiers.
I am sure though, some bright spark around here would know enough to mod them to produce the voltages we are interested in.
While I agree shunt regulators offer pretty spectacular performance, trying to implement one for a class A amplifier can be a real pain in terms of the amount of heat that is dissipated. The heatsinks required can be as large as those used in the amplifer itself.
If you look at some of the reviews these days on top quality pc power supplies (switch mode of course), you will notice they have improved tremendously.
Seasonic, while manufacturing their own brand of powersupplies, also manufacture power supplies for the best brands in the business. A lot of them are capable of 1000W, the only problem is they don't have the necessary output voltages for amplifiers.
I am sure though, some bright spark around here would know enough to mod them to produce the voltages we are interested in.
Last edited:
I have heard many amplifiers better than the Linn Klimax, but it was nice sounding and cool running with tons of power for driving just about anything. And like Mark said after your blast, it uses an SMPS. Opinions are great....as long as the spirit of good will and helping one another is maintained. There are several high end amplifiers that use them including tube, solid state, and 12vdc auto applications. Look around and consider the possibilities...
To be honest I was a little surprised when the question was posed but was equally curious to see what might come out of the community of extremely bright folks that hang around here.
Chris
You can see among Babowana's posts as i know he has built one and opened a thread at this purpose.
Yes you are right about that, but we don't necessarily need to build one from scratch. We just need enough info to mod a pc power supply to give us the voltages we need.
To get a feel for what is achievable have a look at the link Seasonic M12D 750 W Power Supply Review | Hardware Secrets
Have a look at figure 19 and 20.
23mV p-p of noise isn't great but it isn't bad either, espescially when you consider this is the level of noise at the maximum specified power output (ie 750Watts) of the power supply. Things improve quite a bit when you operate them at less than 100%.
Last edited:
Now, consider what could be achieved if we put our own CLC or CRC (or even pure capacitor smoothing) power supply at the output.
It would not be very difficult to reduce that 23 mV p-p noise to below 1 mV p-p noise, and that is at 100% of the rated power output of the switchmode power supply.
Now to put things further into perspective, consider what would happen if you tried to run a 750VA transformer at 750W continuous. I have never tried this but I know it wouldn't be pretty.
Using Nelsons rule. To run 750 Watts continuous you would need a 1500VA (1.5 kVA) transformer, as a bare minimum.
Now, a 1.5 kVA transformer is not cheap.
I think if we want to seriously consider this furher then we should start up a new thread dedicated to it.
It would not be very difficult to reduce that 23 mV p-p noise to below 1 mV p-p noise, and that is at 100% of the rated power output of the switchmode power supply.
Now to put things further into perspective, consider what would happen if you tried to run a 750VA transformer at 750W continuous. I have never tried this but I know it wouldn't be pretty.
Using Nelsons rule. To run 750 Watts continuous you would need a 1500VA (1.5 kVA) transformer, as a bare minimum.
Now, a 1.5 kVA transformer is not cheap.
I think if we want to seriously consider this furher then we should start up a new thread dedicated to it.
Last edited:
In my world its more simple
ClassD needs a very fast supply
ClassAB also needs a fast supply, but less
ClassA is on continiously and doesnt depend as much on such a fast supply
Switcmode seems ideal fore classD, yet some who have tried both switchmode and ordinary supply with big trafo on classD have stated that traditional trafo and caps sounds better
Switchmode is cheap, light and small, if thats what your priority
I am though very careful not to put too much weight on such statements
I might consider THELs regulated supply in his new symmetrical version, takes AC directly
Though, I still remember mr Thel himself writing that traditional oldstyle supply being the best sounding, if you dont need the voltage regulation
When a man who makes his living selling regulated supply designs say he favours traditional supply, then I tend to believe the man
Sure, might be nice to have the option to be able to change supply voltage easily
Issues
does a regulated supply still need a big trafo, or would it work juast as well with smaller trafo ?
Burning 100watt continiously or maybe more, would such a regulated supply get very hot ?
ClassD needs a very fast supply
ClassAB also needs a fast supply, but less
ClassA is on continiously and doesnt depend as much on such a fast supply
Switcmode seems ideal fore classD, yet some who have tried both switchmode and ordinary supply with big trafo on classD have stated that traditional trafo and caps sounds better
Switchmode is cheap, light and small, if thats what your priority
I am though very careful not to put too much weight on such statements
I might consider THELs regulated supply in his new symmetrical version, takes AC directly
Though, I still remember mr Thel himself writing that traditional oldstyle supply being the best sounding, if you dont need the voltage regulation
When a man who makes his living selling regulated supply designs say he favours traditional supply, then I tend to believe the man
Sure, might be nice to have the option to be able to change supply voltage easily
Issues
does a regulated supply still need a big trafo, or would it work juast as well with smaller trafo ?
Burning 100watt continiously or maybe more, would such a regulated supply get very hot ?
Attachments
Last edited: