That is surprising, when I was reading up on power supplies, all of the sources I found indicate that linear power supplies are lower efficiency than switching:
Linear vs Switching Power Supplies - Advanced Conversion Technology
Linear Power Supply vs. Switching Power Supply: Advantages and Disadvantages
http://www.audiodevelopers.com/misc/Apogee_App_Notes/AN-01.pdf
Ok ... I think we all know that basically a switching power supply has a higher efficiency than an "analog" one, but in the HiFi world it has a relative importance
If I have to drive an audio power amplifier and the toroidal transformer sounds better than the switching the fact that the efficiency is lower I think it is not very interesting ... don't you think?
If I have to drive an audio power amplifier and the toroidal transformer sounds better than the switching the fact that the efficiency is lower I think it is not very interesting ... don't you think?
Linear regulated power supplies are less efficient than SMPS power supplies. The only losses in an unregulated power supply are diodes, which SMPS also have and transformer losses due to DC resistance in the windings. Plug a 1KW transformer into the wall. A decent one won't even get warm.
Transformer Efficiency, Losses and Heat
Transformer Efficiency, Losses and Heat
The caveat being that the spending more for a better SMPS could match the sound quality of the linear power supply, then it's a matter of money vs efficiency once again. Would you pay more money for much improved power efficiency? Some of us would.
For me these are useless talk ... the only sector where SMPS power supplies win over linear is in the professional world, where the amps exceed 500-1000W and where Audiophile quality is not in first place, but efficiency and low weight. they make a difference
In a 50-100-150W Audiophile power amp these advantages fall to the last places ...
Specifying that a linear power supply for a power amplifier includes a transformer + diodes + capacitors, without a stabilizer circuit as it is not necessary, so the only "problem" could be the size of the toroidal
In a 50-100-150W Audiophile power amp these advantages fall to the last places ...
Specifying that a linear power supply for a power amplifier includes a transformer + diodes + capacitors, without a stabilizer circuit as it is not necessary, so the only "problem" could be the size of the toroidal
No regulation needed for class D amps? I think the efficiency is lost in the regulation process. I have no clue what kind of PSU would better suit class D amps.
Here are few pros for SMPS though, from DIY perspective:
SMPS is easy, ready made module or even a plug in power brick which both kind of ease the scary fiddling with high voltage. Also lots less work to hook complete module into a chassis than piecing together multiple odd bits. Module vs. multiple odd bits easily saves at least an hour worth of work and many more if not building stuff constantly. Getting rid of mains hum with linear supplies is a bit of work, see the other amplifier forums here on Diyaudio, I suspect most common problem for first builds is the hum.
There is always circumstances where either is better option so no need to fight over it
If linear supply sounds better but is more work, someone is willing to do it and some not.
Here are few pros for SMPS though, from DIY perspective:
SMPS is easy, ready made module or even a plug in power brick which both kind of ease the scary fiddling with high voltage. Also lots less work to hook complete module into a chassis than piecing together multiple odd bits. Module vs. multiple odd bits easily saves at least an hour worth of work and many more if not building stuff constantly. Getting rid of mains hum with linear supplies is a bit of work, see the other amplifier forums here on Diyaudio, I suspect most common problem for first builds is the hum.
There is always circumstances where either is better option so no need to fight over it
If linear supply sounds better but is more work, someone is willing to do it and some not.
Last edited:
Here is some data TPA3255 - all about DIY, Discussion, Design etc.
On a quick look at the data a regulated linear supply beats default SMPS brick in that example case, but the unregulated linear supply comes about the same performance (THD+N) as the SMPS brick. I have no idea if the difference is audible or not since the figure is very low but there seems to be some difference on the harmonic distribution and what not. In my book either would be valid approach depending on the other stuff, ease or comfort or any other preference.
We have to be cautious generalizing too heavily. Checked out the ASR forum amplifier measurement ranking chart and there the SMPS psu amplifiers occupy the top and there is probably both on the tail end, but of course there is more to it than what PSU they use starting from the metrics they use to make the graph. Anyway, use what ever makes one happy
On a quick look at the data a regulated linear supply beats default SMPS brick in that example case, but the unregulated linear supply comes about the same performance (THD+N) as the SMPS brick. I have no idea if the difference is audible or not since the figure is very low but there seems to be some difference on the harmonic distribution and what not. In my book either would be valid approach depending on the other stuff, ease or comfort or any other preference.
We have to be cautious generalizing too heavily. Checked out the ASR forum amplifier measurement ranking chart and there the SMPS psu amplifiers occupy the top and there is probably both on the tail end, but of course there is more to it than what PSU they use starting from the metrics they use to make the graph. Anyway, use what ever makes one happy
Last edited:
Some clarification needed here...
A SMPS will be more efficient than a linear supply where the supply output is REGULATED. Regulated meaning the output is held stable under varying loads.
A transformer, rectifier and capacitors make an UN-regulated supply. Strictly speaking this is not a LINEAR since there is no output regulation, and therefore the losses associated with linear supplies are not present.
Since we are in the Class-D / TPA3255 forum, the term 'linear' is perhaps used inaccurately to describe non-switching supplies.
So does this TPA3255 need a regulated power supply? I suspect that a suitable unregulated supply would be more than acceptable. Another member here as already commented that this is appears to be the case. Cannot provide a concrete reason why this should be true but probably has something to do with the output being modulated in class-D and that the TPA3255 has a bridged output. Class-D output level after filtering being the average level of the PWM level rather than the level of the PWM voltage. Bridged meaning that the output of each channel is connected between the output of two amplifiers (one output an inversion of the other). So long as the reservoir capacity of the unregulated supply is sufficient for the load demand I suspect there will be no issues.
A SMPS will be more efficient than a linear supply where the supply output is REGULATED. Regulated meaning the output is held stable under varying loads.
A transformer, rectifier and capacitors make an UN-regulated supply. Strictly speaking this is not a LINEAR since there is no output regulation, and therefore the losses associated with linear supplies are not present.
Since we are in the Class-D / TPA3255 forum, the term 'linear' is perhaps used inaccurately to describe non-switching supplies.
So does this TPA3255 need a regulated power supply? I suspect that a suitable unregulated supply would be more than acceptable. Another member here as already commented that this is appears to be the case. Cannot provide a concrete reason why this should be true but probably has something to do with the output being modulated in class-D and that the TPA3255 has a bridged output. Class-D output level after filtering being the average level of the PWM level rather than the level of the PWM voltage. Bridged meaning that the output of each channel is connected between the output of two amplifiers (one output an inversion of the other). So long as the reservoir capacity of the unregulated supply is sufficient for the load demand I suspect there will be no issues.
Hi Everyone, new poster, long time reader of this forum.
I am building a 3e Audio TPA3255 class d amp, using the 3e audio PS. I finally hooked it up for the first time. However the sound is really lacking in bass, from around 100hz.
Has anyone experienced this with this 3e Audio amp??
I’ve stripped out everything else in the line and am getting my source from my Topping D50s dac which sounds great on my older AB hitachi.
Any help or ideas are very much appreciated. Please also let me know if it’s better I start a new thread. And I know sound, especially bass is really subjective but I think it’s really rolling off.
Cheers,
Matt
I am building a 3e Audio TPA3255 class d amp, using the 3e audio PS. I finally hooked it up for the first time. However the sound is really lacking in bass, from around 100hz.
Has anyone experienced this with this 3e Audio amp??
I’ve stripped out everything else in the line and am getting my source from my Topping D50s dac which sounds great on my older AB hitachi.
Any help or ideas are very much appreciated. Please also let me know if it’s better I start a new thread. And I know sound, especially bass is really subjective but I think it’s really rolling off.
Cheers,
Matt
Folk, sorry to change subject, but a quick power supply question regarding the TPA 3255. For the same output Watts, is it best to go low voltage and high current or high voltage and low current ? According to data sheet, I can get about 105 Watts per side using 32v/6.4A. But it appears I can get the same 105 Watts per side using a 48v/4.2A. Whats best?
Has anyone looked into using IcePower 200AS2? I believe it's a TPA3255 board ? Pros and Cons vs other TPA3255 boards? Also, was thinking of getting a PecanPi setup (with the Volume control) from them and possibly mounting in the same case as amp. Has anyone seen this done? Good or bad idea?
200AS2 - ICEpower
200AS2 - ICEpower
Last edited:
Audioarchitech
This ICE board is not based on the TPA3255 chip and is an old design (>15 years) and is unlikely to compete on performance to any well implemented TPA3255 solutions.
There is a good review on ICEpower 200AC & 200ASC Review (Class D Amp) | Audio Science Review (ASR) Forum.
This ICE board is not based on the TPA3255 chip and is an old design (>15 years) and is unlikely to compete on performance to any well implemented TPA3255 solutions.
There is a good review on ICEpower 200AC & 200ASC Review (Class D Amp) | Audio Science Review (ASR) Forum.
Chip is on the bottom (between the inductors) to transfer heat to the aluminum base plate. Here is a photo of the 100as2 chip:TPA3251/55 High Power Class D Amplifiers
I've seen several people claim that the 200AS2 uses TI's TPA3255. However, this information isn't listed on the datasheet or anywhere on ICEpower's web site (that I can find), which makes me a little suspicious. Either this claim is incorrect, or ICEpower is hesitant to advertise the fact that they're relying on another company's technology.
ICEpower really needs a better product numbering system, because it's too hard to distinguish their newer models from older ones. Presumably because their main focus is on making bulk sales to other vendors, they're not as interested in clarifying such details to casual DIY buyers.
ICEpower really needs a better product numbering system, because it's too hard to distinguish their newer models from older ones. Presumably because their main focus is on making bulk sales to other vendors, they're not as interested in clarifying such details to casual DIY buyers.