Low cost high quality excellent performing LPSU for audio.

[jabbr]

 

4 hours ago, sandyk said:

 

Buffers are as only good for cleaning up a signal as the PSU area and Earthing in particular in the PC /server permits.

See this thread started by one and a half (Gary)

https://audiophilestyle.com/forums/topic/60381-hdd-to-case-bonding-uptick-in-sq/

 

 

We are talking about networked audio here, and yes absolutely the noise in the network attached device will arise from the device itself including PCB design, PSU etc.

 

I think its very reasonable to use a good power supply on the endpoint

15 hours ago, One and a half said:

So does the buffer also playback the noise?

 

The buffer reads out, or plays back, the noise in the buffer. The buffer, assuming a modern functional network, does not contain memory of differential nor phase noise elsewhere in the networked system e.g. server.

 

what is the caveat here? these are the types of noise: each type of noise is a graph of value vs frequency (different types of noise e.g. shot, flicker, johnson have different effects at different frequencies):

1) voltage

2) current

3) differential mode

4) common mode

5) phase

 

Differential digital circuitry in general is excellent at rejecting differential mode noise. Buffers are excellent at rejecting phase noise.

Common mode noise is pesky.

 

"How I learned to stop worrying and love the glass"

 

My own use of fiberoptic ethernet arose from my desire to completely eliminate the *consideration* of common mode noise transmission over a network. I can't say that any network device *does* or doesn't transmit common mode noise: I have said that 10Gbe and faster network protocols test for the transmission of both phase and differential mode noise, and don't, because this is tested for, but these tests don't test for common mode noise. Fiberoptic eliminates this consideration.

 

[plissken]

28 minutes ago, jabbr said:

 

The buffer reads out, or plays back, the noise in the buffer.

 

Is this a word slip? Data is read in/out.

[jabbr]

20 minutes ago, plissken said:

 

Is this a word slip? Data is read in/out.

 

From an electronics point of view, the buffer is an electronic device. "Data" is an abstraction. You are dealing with a "bit" which is defined as two ranges of voltages. Use of a buffer does not eliminate all noise, rather replaces noise in the transmitter with noise in the receiver.

[plissken]

9 minutes ago, jabbr said:

 

From an electronics point of view, the buffer is an electronic device. "Data" is an abstraction. You are dealing with a "bit" which is defined as two ranges of voltages. Use of a buffer does not eliminate all noise, rather replaces noise in the transmitter with noise in the receiver.

 

Ok, jitter is going to exist in some form on either read or write. But if there is no read/write operation, that it is data at rest. Just wanted to make sure we are on the same page and we are. Some could read that as the upstream noise is read back out and I just wanted to make that a clarifying point.

[jabbr]

Just now, plissken said:

 

Ok, jitter is going to exist in some form on either read or write. But if there is no read/write operation, that it is data at rest. Just wanted to make sure we are on the same page and we are. Some could read that as the upstream noise is read back out and I just wanted to make that a clarifying point.

To be clear there will be both some phase noise (jitter) on the readout, as well as voltage and current noise due to the electronics. "the noise in the buffer" is just that, the intrinsic noise in the buffer, when at "rest" there are still electrons bumbling around in the gates.

[plissken]

3 minutes ago, jabbr said:

To be clear there will be both some phase noise (jitter) on the readout, as well as voltage and current noise due to the electronics. "the noise in the buffer" is just that, the intrinsic noise in the buffer, when at "rest" there are still electrons bumbling around in the gates.

 

I totally agree. Just wanted to make the point that there is no 'accumulative' noise going on. If it's nonvolatile storage.... ;-)

[R1200CL]

1 hour ago, jabbr said:

Buffers are excellent at rejecting phase noise.

I wonder then why it’s not used in clock design. Where low phase noise is the key to good SQ. 

[jabbr]

2 hours ago, R1200CL said:

I wonder then why it’s not used in clock design. Where low phase noise is the key to good SQ. 

 

Don't get stuck on the vastly too simplistic idea that the only important things in digital audio electronics are clocks and power supplies.

 

There are clocks everywhere. Do you have any idea of the number of clocks that exist between my typing and you reading this message? I have no idea because each digital receiver buffers and reclocks each packet of information and the message gets through despite the number of links. Even the best clocks have phase noise and if noise were additive across each hop in a network then the network wouldn't work, the required jitter tolerances are that tight. What network receivers do is *reject jitter*. Broadly speaking there are two ways:

1) clock recovery aka PLL

2) FIFO buffers

[R1200CL]

1 hour ago, jabbr said:

Don't get stuck on the vastly too simplistic idea that the only important things in digital audio electronics are clocks and power supplies.

Don’t disagree.
When the power part has been taken care of in a audio system, and wires, shielding, and fiber applied, then 10 MHz clocks in some systems seems to be next logical step. 

There are other clock frequencies. I’m not there yet 😀

 

I think there is much yet to discover and learn about 10 MHz  clocks impact on audio. As you probably noticed, there’s is a post soon coming from John and Alex that hopefully will be very helpful in better understanding OCXO clocks.

Still these magic buffers obviously doesn’t solve phase noise on clocks.

 

I’ve asked several probably not easy to answer questions lately over at the EtherRegen clock tread. 
I hope some answers by Easter. An indication of price point of top notch clocks, makes me understand that what did look like an easy solution, isn’t straightforward.

 

[The Computer Audiophile]

New topic

 

 

[jabbr]

2 hours ago, R1200CL said:

Don’t disagree.
When the power part has been taken care of in a audio system, and wires, shielding, and fiber applied, then 10 MHz clocks in some systems seems to be next logical step. 

There are other clock frequencies. I’m not there yet 😀

 

I think there is much yet to discover and learn about 10 MHz  clocks impact on audio. As you probably noticed, there’s is a post soon coming from John and Alex that hopefully will be very helpful in better understanding OCXO clocks.

Still these magic buffers obviously doesn’t solve phase noise on clocks.

 

I’ve asked several probably not easy to answer questions lately over at the EtherRegen clock tread. 
I hope some answers by Easter. An indication of price point of top notch clocks, makes me understand that what did look like an easy solution, isn’t straightforward.

 

Nope. Never. Just nope.

[Jonder]

Hi @airguitar

 

I am using this lps https://www.aliexpress.com/item/4001288349011.html

with great success along with Sommer square 4-core mk2 highflex cable for the dc supply.

[FIndingit]

11 hours ago, Jonder said:

Hi @airguitar

 

I am using this lps https://www.aliexpress.com/item/4001288349011.html

with great success along with Sommer square 4-core mk2 highflex cable for the dc supply.

Is this Zerozone psu now a “Denafrips rip-off” like Musician Pegasus dac was supposed to be? 😄
 

It has all the signs of a Denafrips! O-core transformer and loads of caps. 

[TomJ]

On 2/17/2021 at 9:22 AM, Jonder said:

Hi @airguitar

 

I am using this lps https://www.aliexpress.com/item/4001288349011.html

with great success along with Sommer square 4-core mk2 highflex cable for the dc supply.

Hi Jonder,

 

Interesting LSP!

Do you have done any comparisons with other LPS?

Or are there any reviews on the net available?

 

Regards,

 

Tom

[Jonder]

Hi Tom

 

I’m not aware of any reviews about this particular lps. I do have a cheaper lps of a different design also bought from China which is noticeably poorer and degrades the sound quality compared to this one.

 

I’m currently in the process of putting together a three rail supply using Zerozone Leo power boards from China

 

https://www.aliexpress.com/item/1005002587132298.html

 

and fully potted toroidal transfomers from Poland.

 

https://sklep.toroidy.pl/pl/c/Transformatory-AUDIO-Supreme-V2/99

 

[TomJ]

17 hours ago, Jonder said:

Hi Tom

 

I’m not aware of any reviews about this particular lps. I do have a cheaper lps of a different design also bought from China which is noticeably poorer and degrades the sound quality compared to this one.

 

I’m currently in the process of putting together a three rail supply using Zerozone Leo power boards from China

 

https://www.aliexpress.com/item/1005002587132298.html

 

and fully potted toroidal transfomers from Poland.

 

https://sklep.toroidy.pl/pl/c/Transformatory-AUDIO-Supreme-V2/99

 

The Zerozone Leo power boards seems to be out of stock.

 

Any reviews of the Leo board?

 

Regarding affordable Chinese LPS I have following suggestions so far:

 

Zerozone S11 LPS: https://www.ebay.de/itm/High-end-SUPER-PSU-30VA-DC12V-1-3A-Linear-Power-supply-LPS-for-Audio-Sources/224070931302

 

Jays Audio LPS: 

https://de.aliexpress.com/item/1005002408959425.html  

This seems to be the same device like the original from Jays Audio: https://www.jays-audio.com/product-page/lps25va

 

[Jonder]

The Leo boards are showing out of stock because I requested then so the listing was for me. I’m sure if you wanted some they will put a listing up for you. They reply to messages quickly.

[TomJ]

13 minutes ago, Jonder said:

The Leo boards are showing out of stock because I requested then so the listing was for me. I’m sure if you wanted some they will put a listing up for you. They reply to messages quickly.

So what is the price for the board, and why do you have chosen the Leo board?