In general yes, but most likely you'll encounter problems with switching noise, but if you can overcome that via careful design and layout it is OK.
Thanks, what about ripple, output impedance, bandwidth etc…are these factors as good as or better in switcmode supply’s. And why don’t we see more of thes PSU if they are superior….
Baard said:
Maybe only few people know how to make them work well in audio?
You will need a SMPSU with an RMS power rating equal to the Peak power of the connected amp (+loses) in order not to run out of current.
\Jens
A good regulated SMPS outperforms any conventional unregulated 50hz or 60Hz approach in all measurable fields. However, in order to develop good switching circuits (SMPS and class D) a solid understanding of the following topics is required:
- Switching behaviour of solid state devices
- PCB parasitistics
- Component parasitistics
- Magnetic components
- Wiring parasitistics
- Conducted and radiated EMI
- EMI filtering
Since almost everyone designing linear audio amplifiers lacks that understanding, we are not seeing much SMPS in commercial products. Also, safety approval agencies make things more complex and costly when it comes to get permission to sell SMPS-based products, so most people sticks a pre-approved toroid transformer into their design and forgets about the subject.
I feel that it's better that way because those linear designers are just not skilled to do switching stuff. Indeed, their PCB layouts and wirings (seen on most popular amplifiers and kits) look as if they were unadvertedly intended to be radio transmitters (if ever operated above the audio band).
Unfortunately, some of those designers are migrating to class D and the results seem quite discouraging. I'm currently developing a 2400W class D amplifier (not intended for audio, though) and since I'm fully aware of the previously mentioned issues I'm getting plenty of headaches from time to time trying to make the circuit cleaner in EMI terms. Everything seems much easier for those people unaware of EMI issues and parasitistics because they have much less stuff to worry about.
Also, those popular-wisdom claims about "switching noise" audibility are proving to be absolutely false, because plenty of the so called "audiophiles" and "tweakers" seem to be falling in love with the currently emerging generation of poorly designed class D amplifiers, where said switching "noise" is happily radiated and fed direcly to the loudspeaker in great amounts, instead of being carefully filtered and handled like the most harmful thing in the world as it was in SMPS-based linear amplifiers
- Switching behaviour of solid state devices
- PCB parasitistics
- Component parasitistics
- Magnetic components
- Wiring parasitistics
- Conducted and radiated EMI
- EMI filtering
Since almost everyone designing linear audio amplifiers lacks that understanding, we are not seeing much SMPS in commercial products. Also, safety approval agencies make things more complex and costly when it comes to get permission to sell SMPS-based products, so most people sticks a pre-approved toroid transformer into their design and forgets about the subject.
I feel that it's better that way because those linear designers are just not skilled to do switching stuff. Indeed, their PCB layouts and wirings (seen on most popular amplifiers and kits) look as if they were unadvertedly intended to be radio transmitters (if ever operated above the audio band).
Unfortunately, some of those designers are migrating to class D and the results seem quite discouraging. I'm currently developing a 2400W class D amplifier (not intended for audio, though) and since I'm fully aware of the previously mentioned issues I'm getting plenty of headaches from time to time trying to make the circuit cleaner in EMI terms. Everything seems much easier for those people unaware of EMI issues and parasitistics because they have much less stuff to worry about.
Also, those popular-wisdom claims about "switching noise" audibility are proving to be absolutely false, because plenty of the so called "audiophiles" and "tweakers" seem to be falling in love with the currently emerging generation of poorly designed class D amplifiers, where said switching "noise" is happily radiated and fed direcly to the loudspeaker in great amounts, instead of being carefully filtered and handled like the most harmful thing in the world as it was in SMPS-based linear amplifiers
Well there you go. The average audio designer is too naive to understand a SMPS. Hmm.
For me switch mode is primarily a miniaturization technology: mostly because HF transformers (100kHz) are much smaller per watt than 50/60Hz transformers and because switching regulators dissipate less heat than linear regulators. But you don't get something for nothing.
Perhaps addressing the wider question of potential benefits of SMPS is more useful here. Size and power dissipation would seem to be the key benefits. The downsides are complexity, noise, impedance issues. I'm not convinced a SMPS for an audio amp of quality will save any money over a linear, regulated PS.
So, if you need small size and low heat go for SMPS. If you want high fidelity and minimum complexity go for a linear PS. That's my recommendation. 😎
For me switch mode is primarily a miniaturization technology: mostly because HF transformers (100kHz) are much smaller per watt than 50/60Hz transformers and because switching regulators dissipate less heat than linear regulators. But you don't get something for nothing.
Perhaps addressing the wider question of potential benefits of SMPS is more useful here. Size and power dissipation would seem to be the key benefits. The downsides are complexity, noise, impedance issues. I'm not convinced a SMPS for an audio amp of quality will save any money over a linear, regulated PS.
So, if you need small size and low heat go for SMPS. If you want high fidelity and minimum complexity go for a linear PS. That's my recommendation. 😎
Dear Eva,
could you recommend a manufacturer that, in your opinion, has solved these issues in a manner allowing the use of their switching supplies.
I am interested in 12V/2A and 24V/2A.
Thank you,
M
could you recommend a manufacturer that, in your opinion, has solved these issues in a manner allowing the use of their switching supplies.
I am interested in 12V/2A and 24V/2A.
Thank you,
M
From my experience SMPS better than conventional, fast transfer source, no hum.
Noise? No big problem for sound frequency area. I have experience problem with TV receiving, but not for sound.
The other advantage is less weight, less space, less caps size, etc.
The requirements SMPS for Amp is that the SMPS able to handle the variety load. Normally SMPS is designed for continuous load.
We are know that all big car amps powered with SMPS anyway.
Noise? No big problem for sound frequency area. I have experience problem with TV receiving, but not for sound.
The other advantage is less weight, less space, less caps size, etc.
The requirements SMPS for Amp is that the SMPS able to handle the variety load. Normally SMPS is designed for continuous load.
We are know that all big car amps powered with SMPS anyway.
I recall that Ben Duncan spoke very highly of SMPS for audio amps in an article in "Electronics World" some years ago. I think it was about the bass response of amps.
Yes, IMHO SMPS seems that the amp connected to infinity source, good fo music with full bass, while conventional supply with big caps seems a big tank that need time to charging.
Now all of my DIY amps (hi fi) powered with ex PC PSU with little modification. I have four amps 2 chips, 1 solid state, 1 class-D DIY UCD. All of them have better bass response. But I did not test with testing tool, perharps only my impression too.
regards,
kartino
Now all of my DIY amps (hi fi) powered with ex PC PSU with little modification. I have four amps 2 chips, 1 solid state, 1 class-D DIY UCD. All of them have better bass response. But I did not test with testing tool, perharps only my impression too.
regards,
kartino
SMPS
We are building amplifiers with SMPS +PFC with good results. it is easy to regulate vs dumb iron+copper.
If you follow the rules safety is not an issue
We are building amplifiers with SMPS +PFC with good results. it is easy to regulate vs dumb iron+copper.
If you follow the rules safety is not an issue
You might find the 200W/8 Yamaha B-6 interesting, also called the Pyramid amplifier because of its shape.
Yamaha only made a couple of them around 1980 because the powersupply in the B-6 was the switching 'Magnetic Field Powersupply' of Bob Carver. Single time i've heard about a Japanese corporation having to back down because of an IP claim.
The amplifier circuitry in the B-6 had fast devices, such as the Toshiba 2SA1095/2SC2565 ring-emitter output transistors.
An integrated Yamaha model also employed the switching PS, and was taken out of production as well.
Sound quality of the B-6 was quite good compared to regular PS models in the same price range.(if you can stand the sound of Yamaha amps)
The user manual of the B-6 can still be downloaded from the www2.Yamaha.co.jp site. The B-6 service manual can be found on download sites such as eservice, has a detailed description of the powersupply.
Yamaha only made a couple of them around 1980 because the powersupply in the B-6 was the switching 'Magnetic Field Powersupply' of Bob Carver. Single time i've heard about a Japanese corporation having to back down because of an IP claim.
The amplifier circuitry in the B-6 had fast devices, such as the Toshiba 2SA1095/2SC2565 ring-emitter output transistors.
An integrated Yamaha model also employed the switching PS, and was taken out of production as well.
Sound quality of the B-6 was quite good compared to regular PS models in the same price range.(if you can stand the sound of Yamaha amps)
The user manual of the B-6 can still be downloaded from the www2.Yamaha.co.jp site. The B-6 service manual can be found on download sites such as eservice, has a detailed description of the powersupply.
MOER, have you got a plot of one of your SMPS output Z versus f for f from 1kHz to 10MHz, and with dc output current ranging from 10mA to 10A?
It is a huge technical challenge to make a raw SMPS perform as well, from an audio point of view, as a linear supply.
In terms of replacing a big, "dumb" iron/Cu transformer then I agree. I think treating a SMPS as a transformer replacement...but only that...is a good starting point. And only if you really dislike transformers
It is a huge technical challenge to make a raw SMPS perform as well, from an audio point of view, as a linear supply.
In terms of replacing a big, "dumb" iron/Cu transformer then I agree. I think treating a SMPS as a transformer replacement...but only that...is a good starting point. And only if you really dislike transformers
I think we need someone who cabable to figuring out both characteristic of power supply on music load.
Hi Kanwar,
Have you the complete Crest amp schematic? This is possible send me:
fabrasantos@@@yahoo.es
Thanks,
Fabrício
Have you the complete Crest amp schematic? This is possible send me:
fabrasantos@@@yahoo.es
Thanks,
Fabrício
kartino said:
Could you post a little more information on your Modification(s) of the PC Power supplies? It sounds like something I'd like to try.
Thanks!
Tall Shadow
I am sorry, I am forget the link, anybody have link for destroyer-x at SS amp related this topic please?
I second that. I just used two PC PSUs in series to power my son's Class-T MC8X100 amp. I get 24V at 11A, and the amp works great at my place. However, there's a lot of noise when the amp is connected at my son's place. The noise goes away if the amp inputs are disconnected. The PSUs also disrupt my son's wireless keyboard and mouse 😱 Unfortunately, my knowledge of electronics does not go beyond soldering skills. No modifications were done to the PSUs, except for the ground disconnected from the 12V rail on one of the units.Tall Shadow said:
Any help would be appreciated. TIA.
- Status
- Not open for further replies.