Hi,
I have recently studied the principles of some well known commercial power supplies, such as the Paul Hynes SR-7, Uptone JS2, etc... , and it appears that the quality of the transformer has an impact on the final result (sound quality).
As I am not an expert in the transformers domain, I try to identify what are the criterias that make a transformer better (sounding) than another. I know a firm that can manufacture custom transformers, Toroidals and C-core.
Of course the Transformer is not the only part to make a difference, but for now I just focus on the transformer.
The cost is not really a concern for the time being: I just try to identify the criteria to care about.
The P.S. should be able to deliver 10A on 12V, with 2 rails.
So:
-Toroids or C-Core ?
-And of course what are the other things that could be asked to the manufacturer ? screen, what quality of materials, etc...
Thank you for your replies
I have recently studied the principles of some well known commercial power supplies, such as the Paul Hynes SR-7, Uptone JS2, etc... , and it appears that the quality of the transformer has an impact on the final result (sound quality).
As I am not an expert in the transformers domain, I try to identify what are the criterias that make a transformer better (sounding) than another. I know a firm that can manufacture custom transformers, Toroidals and C-core.
Of course the Transformer is not the only part to make a difference, but for now I just focus on the transformer.
The cost is not really a concern for the time being: I just try to identify the criteria to care about.
The P.S. should be able to deliver 10A on 12V, with 2 rails.
So:
-Toroids or C-Core ?
-And of course what are the other things that could be asked to the manufacturer ? screen, what quality of materials, etc...
Thank you for your replies
In most instances, the transformer is not going to have an effect on the sound quality, unless you adopt the philosophy of cheap computers speakers.
For professional sound systems having to deliver sustained power levels for long periods, it does matter (but not that much).
For a domestic use, there are many other things to prioritize.
As a test, use your favorite amp with a generic transformer supply, then with the best toroid or R-core, and try to (honestly) evaluate the difference.
If you listen at near max power, a cheap transformer will run out of steam and clip a bit earleir, but that should be the only difference (if your test is fair).
In fact, with some types of music, and somewhat unfair conditions, you could find a crap X former more pleasant
For professional sound systems having to deliver sustained power levels for long periods, it does matter (but not that much).
For a domestic use, there are many other things to prioritize.
As a test, use your favorite amp with a generic transformer supply, then with the best toroid or R-core, and try to (honestly) evaluate the difference.
If you listen at near max power, a cheap transformer will run out of steam and clip a bit earleir, but that should be the only difference (if your test is fair).
In fact, with some types of music, and somewhat unfair conditions, you could find a crap X former more pleasant
I agree that "in most cases", the transformer does not have a lot of impact, but when it comes to Audio and when the sound quality is the only priority, the transformer does make a difference.
I have verified this impact on various devices such as DAC, preamplifiers and amps: I have lesoned to the sound quality with different transformers having all the same VA rating.
Toroidal, EI types (the rectangular ones) and R-core (but they were cheap R-core): they had all delivered a different sound, and even 2 transformers of the same type (let say 2 toroidals with the same VA rating) did not sound exactly the same.
On a DAC for example, I did prefer the EI type (the rectangular ones) over the toroidal ones (same VA): the sound was more natural, more opened and relaxed...
Unfortunately, I have not enough knowledge regarding the transformers to know what criteria(s) makes a difference.
Actually I work on a DC power supply: 12V, 6A to 10A max.
So maybe some of you have the required expertise ?
I’m also learning how to build a small power supply, and like you, I’ve noticed that every transformer has its own sonic character.
Unfortunately, I don’t know what factors effect sound quality, but I’m happy to share my limited understanding of transformer design with you.
I’m pretty sure that transformer design has to start with the Transformer EMF Equation. Ie:
V(rms) x (sqrt)2 = 2 x pi x f x A x B x N
Where:
V(rms) = mains voltage
f = mains frequency (Hz)
A = area of transformer core = ((OD - ID) / 2) x HT
B = max flux density (T)
N = number of primary turns
For you, this can be reduced down to:
220 x (sqrt)2 = 2 x 3.142 x 50 x A x B x N
Or,
A.B.N = 0.99
For audio usage, the next step is to choose a value for Bmax. I’ve read that one should specify a Bmax of 1.0T. This probably comes from looking at the B-H curve for the core material and choosing a value below the curves “knee”, which is usually at about 1.2T. But, why choose 1.0T? Why not 0.5T, or 0.9T, or 1.1T. This is something I would like to know for myself.
If we go with B = 1.0T,
A.N = 0.99
For sound quality, do we want a greater core area with lower turns (and lower dc resistance), or smaller core area with more turns (and higher dc resistance)? I don’t know.
There are three variables which need to be kept in balance, with possibly hundreds of “correct” combinations, but what combination produces the best sound?
Unfortunately, I don’t know what factors effect sound quality, but I’m happy to share my limited understanding of transformer design with you.
I’m pretty sure that transformer design has to start with the Transformer EMF Equation. Ie:
V(rms) x (sqrt)2 = 2 x pi x f x A x B x N
Where:
V(rms) = mains voltage
f = mains frequency (Hz)
A = area of transformer core = ((OD - ID) / 2) x HT
B = max flux density (T)
N = number of primary turns
For you, this can be reduced down to:
220 x (sqrt)2 = 2 x 3.142 x 50 x A x B x N
Or,
A.B.N = 0.99
For audio usage, the next step is to choose a value for Bmax. I’ve read that one should specify a Bmax of 1.0T. This probably comes from looking at the B-H curve for the core material and choosing a value below the curves “knee”, which is usually at about 1.2T. But, why choose 1.0T? Why not 0.5T, or 0.9T, or 1.1T. This is something I would like to know for myself.
If we go with B = 1.0T,
A.N = 0.99
For sound quality, do we want a greater core area with lower turns (and lower dc resistance), or smaller core area with more turns (and higher dc resistance)? I don’t know.
There are three variables which need to be kept in balance, with possibly hundreds of “correct” combinations, but what combination produces the best sound?
Transformer design
The power capability of the transformer if it is underpowered can affect the sound as the amp's circuity runs out of juice.
Torroidal transformers produce smaller magnetic fields -adventitious in designing psu's in equipment, but they also saturate easier than C or E core ones. Not to normally much of a problem in lower power circuits preamps dacs streamers etc but a bigger problem in power amps if it is a bit undersized (normally for cost/size reasons), and you use it at high volumes, where C&E cores work better due to having a less abrupt saturation curve. But then the design engineer has to deal with the much larger magnetic field. That field, may also effect other equipment.
Like all engineering there are choices of where you want to have the compromises. Hope this helps.
A&R over spec'd their torroid in the A60 amp, so it drove the power very well and produced a nice dynamic amplifier. The overload fuse would blow well before the transformer saturated.
The power capability of the transformer if it is underpowered can affect the sound as the amp's circuity runs out of juice.
Torroidal transformers produce smaller magnetic fields -adventitious in designing psu's in equipment, but they also saturate easier than C or E core ones. Not to normally much of a problem in lower power circuits preamps dacs streamers etc but a bigger problem in power amps if it is a bit undersized (normally for cost/size reasons), and you use it at high volumes, where C&E cores work better due to having a less abrupt saturation curve. But then the design engineer has to deal with the much larger magnetic field. That field, may also effect other equipment.
Like all engineering there are choices of where you want to have the compromises. Hope this helps.
A&R over spec'd their torroid in the A60 amp, so it drove the power very well and produced a nice dynamic amplifier. The overload fuse would blow well before the transformer saturated.
Numbers of misconceptions are expressed in the two previous posts.
Just to pick a very common one: saturation doesn't limit/occur with heavy loads, quite the opposite in fact, because of parasitic parameters like winding resistance and leakage inductance.
Saturation is a no-load/high primary voltage issue, and if anything, it tends to regulate/stabilize the output voltage in some (undesirable) way.
People worrying about their transformers should also worry about their local pole transformers, and other transformers upstream in their distribution system (there are lots of them, and the aerial power lines connecting them have a large characteristic impedance, translating into a large equivalent leakage inductance).
But of course, hearing is believing
Just to pick a very common one: saturation doesn't limit/occur with heavy loads, quite the opposite in fact, because of parasitic parameters like winding resistance and leakage inductance.
Saturation is a no-load/high primary voltage issue, and if anything, it tends to regulate/stabilize the output voltage in some (undesirable) way.
People worrying about their transformers should also worry about their local pole transformers, and other transformers upstream in their distribution system (there are lots of them, and the aerial power lines connecting them have a large characteristic impedance, translating into a large equivalent leakage inductance).
But of course, hearing is believing
Ditto: ?
Core saturation only occurs when the core cannot absorb and more of the magnetic flux. For a magnetic field to be produce current has to flow it is normally the secondary windings that cause the problem as the current is drawn by the load applied to it.
Although quite what is has to do with the power companies distributions system confuses me. -In the UK was have a very good 3 phase system with high power transformers at the end of blocks, with proper 3 phase underground cabling,rated at 150+amps. Not 2 phase on poles, with wires handing from them.
> Numbers of misconceptions are expressed in the two previous posts.
Agreed.
> But, why choose 1.0T? Why not 0.5T, or 0.9T, or 1.1T. This is something I would like to know for myself.
Work it out. A 0.5T design will be much larger (more expensive!!) than a 1.0T design.
"Saturation" is about applied voltage (and winding and core). Load current does not make saturation.
I too wonder how my utility company came into this. My 500 foot(!) 240V 100A drop has reactance less than resistance. Both so large that my pole-pig is moot. My 20KV street feed is iron and has large resistance, but on my side of the 20KV:240V transformer that impedance looks 7,000 times less, small fraction of total impedance.
MY opinion is that if a power transformer affects the audio, either the winder or the amplifier designer screwed-up bad.
Agreed.
> But, why choose 1.0T? Why not 0.5T, or 0.9T, or 1.1T. This is something I would like to know for myself.
Work it out. A 0.5T design will be much larger (more expensive!!) than a 1.0T design.
"Saturation" is about applied voltage (and winding and core). Load current does not make saturation.
I too wonder how my utility company came into this. My 500 foot(!) 240V 100A drop has reactance less than resistance. Both so large that my pole-pig is moot. My 20KV street feed is iron and has large resistance, but on my side of the 20KV:240V transformer that impedance looks 7,000 times less, small fraction of total impedance.
MY opinion is that if a power transformer affects the audio, either the winder or the amplifier designer screwed-up bad.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.