|
Post by Audiophile Neuroscience on Aug 28, 2019 11:49:05 GMT 10
Yea, the NUC or low power consumption computer is the right way to go, even though these still invariably all have DC-DC switchers on board for the varied voltage rails needed by the CPUs & GPUs. But according to Swenson even LPS have some current leakage although far lower than SMPSes (not sure how sonically detrimental this is?) .The best, most relaxed analogue-like sound I've heard came from an SD card player - simple electronics, fed by battery, no switchers or connections to other noise-producing devices. Unfortunately the target use for computers as general purpose computing workhorses often works against the single purpose use for audio. I've heard it argued both ways ie go small or go big in terms of computing power. I went with a server grade MOBO and CPU. It was claimed that the larger CPU would not be as taxed and not produce as much noise. Linear power supplies also apparently block retrograde noise into the sytem from all the switching going on in the board and of course good galvanic isolation between server and DAC.
My DAC does not require the +5V from the usb connection so that has been disconnected leaving only the data feed.
David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong
|
|
sandyk (RIP Alex, 1939 - 2021)
Global Moderator
Posts: 226
About Me: Retired ex Principal Telecommunications Technical Officer with 43 years at Telstra (Australia)
I am also a Moderator in Hi Fi Critic Forum
Electronics hobbyist for >65 years with DIY projects including Loudspeakers, Stereo FM tuner, S/W Regen Receiver, Superhet AM ,
Synchrodyne PLL AM tuner (Phase Lock Loop),Stereo Tape Deck, Amplifiers including I.C. types, Class A, Class AB 100W/Ch. (ETI5000) 240W/Ch. Mosfet (AEM6000) ,several DACs , numerous PSUs including VERY low noise (<4uV) types etc.for myself and friends
Audio Industry Affiliation: NIL
|
Post by sandyk (RIP Alex, 1939 - 2021) on Aug 28, 2019 13:46:31 GMT 10
If a secondary ground is a problem, then we should address the primary ground - perhaps not between the outlet and the device, rather at the entry point to the house. I understand that the electrical code does not want an audio device to be the shunt if there is a short circuit.
Telstra in Au. required the Earth at a PABX (Private Automatic Exchange) and Payphones etc. to be less than 30 ohms before acceptance of the installation . This used to be tested from the local Telephone Exchange. I can assure you that this is often very hard to achieve in many locations where there is a rocky terrain such as places in Sydney such as Frenchs Forest. It's not as easy as just driving in a few parallel earth rods .
|
|
assisi
New Member
Listening to Music Gender: Male Location: Australia – Central Victoria Posts: 1 Date Registered:
Posts: 3
|
Post by assisi on Aug 28, 2019 16:12:11 GMT 10
I do not understand the intricacies of power and primary and secondary grounds. Hence my innocence in the way I started my project. It was fun drilling the hole for the rod and working out what to put in the hole. There was no way I could drive 3 metres of copper rod into the ground.
As I have said when I found out that there was a potential danger, I disconnected the rod. My rod was possibly more serious than the shortish rod for the house and the hole was filled with a mix of bentonite and gypsum that I then wet to reduce the resistance to the surrounding earth. This meant that any fault even from elsewhere may have “found” my ground rod before another.
The Puritan Audio people said that the Ground Master would avoid any danger. Because of comments made on SNA when I posted about the Ground Master I sought clarification. This is the answer I received from Mike Lester.
“The unit is an extremely robust multi band high pass filter and uses inherently safe, (and fail safe), circuit elements that are designed and certified for use in mains and mains fault conditions.
It is designed and constructed to permit free passage of high frequencies but to very effectively and strongly block currents at 50/60Hz and hence is safe.
Grounding systems in the UK and in Australia are similar hence you should not directly connect an auxiliary earth rod to your HiFi in either nation, GroundMaster makes it safe to do so.
I believe that in Australia as in the UK there has long been a culture of connecting additional ground rods or other buried metal to the HiFi earth, simply because sonically it works so staggeringly well and bang for buck is an incredibly great value upgrade. These connections however are not safe and if you are very unlucky could prove expensive and even life threatening. Use of GroundMaster will make these modified systems entirely safe. without compromising the sonic benefit of the direct connection.
As always it is as well to remind everyone that under no circumstances whatever should the actual safety earth be disconnected.”
John
|
|
|
Post by Audiophile Neuroscience on Aug 28, 2019 16:50:18 GMT 10
calling rmsThis is an article on a functinal earth aka technical earth www.eece.uwa.edu.au/students/safety/elec3307/earthing-cpdMy EE suggested a technical earth (clean functional earth) from music room subpanel (I was rewiring so what the heck) back to main switchboard. A new and separate earth/neutral bus bar is used at the mains. This technical earth at the mains is connected via a wire to the mains earth/neutral bus bar. This conforms to Australian regulations and satisfies MEN requirements whereby you are not permitted to have two separate and isolated earthings. Any fault current from the technical circuit will still pass back to the switchboard and no fault current will pass in the opposite direction toward equipment on the technical circuit. The wire connecting the two earths will only conduct very small amounts of noise from the main earth to the technical earth as both will tend to flow in the opposite direction towards the ground rod. Additional option 1 is to create a separate earthing rod for the technical earth i.e. taken from the technical earth at the mainboard and a wire run-off to a separate earth in the ground near the location of the mainboard (where the main earthing rod is).the wire between the technical Earth and MainEarth at the switchboard is maintained as per regulations. with this type of installation you generally read completely silent noise readings on the technical circuit as is required in some medical and other installations Additional option 2 is to place a one-way filter, diode, which acts as a one-way valve between the main earth and the technical earth at the switchboard. This blocks any noise arising from the main circuits and travelling through into the technical circuit. Additional option 3 is to add some power conditioning from the mains. Harmonic filter (?) Additional option 4 is to add an isolation transformer from the mains. These are required by some commercial installations and in the railways for example. it may be too expensive for residential use. It will tend to eliminate ripple and basically create a new 240 V sine wave. Capacitors can be added to smooth out the voltage to create a smooth steady supply.[shielded transformers are used to reduce any induced noise across the windings and some forms of incoming noise cannot pass through the transformer] He does recommend that surge protection be added that the mains level (rather than peripherally). While products like the super safety switch can offer a combination of surge protection, RCD and overcurrent/voltage breaker protection it is probably better to implement them separately. n.b. technical circuits use specially marked outlets and cannot be used for general purposes. regular GPO's cannot be added to the circuits. David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong
|
|
jkenny
Full Member
Posts: 83
About Me: Audio equipment designer forever in pursuit of more realistic & engaging music reproduction purely because of the extra enjoyment of music created by such reproduction.
http://Ciunas.biz
|
Post by jkenny on Aug 29, 2019 2:54:41 GMT 10
Yes, indeed the whole power delivery system in audio equipment is a crucial but often complicated element. Grounds are simply current return paths & need to be considered in this way. So we have power supply return path, signal return path & when we go into more detail we need to consider the frequency of the signal/power as this defines the return path taken - low frequencies following path of least resistance, higher frequencies following a path of least impedance & in between frequencies following a mix of these two paths
So ground isn't just a sink - it's a complex return path for current & power currents need to be kept separate to signal currents. We often see common mode noise (often the result of current leakage on SMPSes) on interconnect cables between devices which is basically power noise being intermixed with supply currents
|
|
jkenny
Full Member
Posts: 83
About Me: Audio equipment designer forever in pursuit of more realistic & engaging music reproduction purely because of the extra enjoyment of music created by such reproduction.
http://Ciunas.biz
|
Post by jkenny on Aug 29, 2019 3:07:07 GMT 10
I've heard it argued both ways ie go small or go big in terms of computing power. I went with a server grade MOBO and CPU. It was claimed that the larger CPU would not be as taxed and not produce as much noise. Linear power supplies also apparently block retrograde noise into the sytem from all the switching going on in the board and of course good galvanic isolation between server and DAC.
My DAC does not require the +5V from the usb connection so that has been disconnected leaving only the data feed.
David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong Sure, it's not cut & dried small Vs large CPU - I would suggest that the selection criteria is to find a system with as steady a current draw as possible, not large fluctuations in current. I don't know what it means that LPS "block retrograde noise into the system" - LPSes can often suffer from current spikes & from both rectifier & supply capacitors plus the suffer from some level of current leakage. I have found that maybe the sonically most important area for USB isolation is actually the two data lines.
|
|
sandyk (RIP Alex, 1939 - 2021)
Global Moderator
Posts: 226
About Me: Retired ex Principal Telecommunications Technical Officer with 43 years at Telstra (Australia)
I am also a Moderator in Hi Fi Critic Forum
Electronics hobbyist for >65 years with DIY projects including Loudspeakers, Stereo FM tuner, S/W Regen Receiver, Superhet AM ,
Synchrodyne PLL AM tuner (Phase Lock Loop),Stereo Tape Deck, Amplifiers including I.C. types, Class A, Class AB 100W/Ch. (ETI5000) 240W/Ch. Mosfet (AEM6000) ,several DACs , numerous PSUs including VERY low noise (<4uV) types etc.for myself and friends
Audio Industry Affiliation: NIL
|
Post by sandyk (RIP Alex, 1939 - 2021) on Aug 29, 2019 9:01:37 GMT 10
I've heard it argued both ways ie go small or go big in terms of computing power. I went with a server grade MOBO and CPU. It was claimed that the larger CPU would not be as taxed and not produce as much noise. Linear power supplies also apparently block retrograde noise into the sytem from all the switching going on in the board and of course good galvanic isolation between server and DAC.
My DAC does not require the +5V from the usb connection so that has been disconnected leaving only the data feed.
David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong Sure, it's not cut & dried small Vs large CPU - I would suggest that the selection criteria is to find a system with as steady a current draw as possible, not large fluctuations in current. I don't know what it means that LPS "block retrograde noise into the system" - LPSes can often suffer from current spikes & from both rectifier & supply capacitors plus the suffer from some level of current leakage. I have found that maybe the sonically most important area for USB isolation is actually the two data lines. Hi John I agree. Introducing Linear PSUs may also lead to earth loop issues. That is why in my general purpose PC I use additional Voltage regulators followed by a John Linsley Hood designed PSU add-on powered from the +12V internal SMPS to power both my internal SSDs with a much cleaner and more isolated power supply . I also use a dual JLH to provide cleaner and highly stable +12V and +5V power for my internal BR writer. In doing this I am using the PC's own Earth Reference. I also have an additional front panel USB 2.0 port, where I have disconnected the internal cable's red +5V wire right next to the end of the plug for the Motherboard socket to prevent noisy +5V SMPS from inducing into the D+ and D- wires. Regards Alex
|
|
|
Post by Audiophile Neuroscience on Aug 29, 2019 9:08:18 GMT 10
I don't know what it means that LPS "block retrograde noise into the system" - LPSes can often suffer from current spikes & from both rectifier & supply capacitors plus the suffer from some level of current leakage. Hi John
I seem to recall (and I may be very wrong) that regarding noise from the pc, linear power supplies have greater noise rejection or filtering of that noise passing back upstream into the line compared to SMPS. This is quite apart from their own noise generation issues being claimed less than SMPS. Alex sandyk (RIP Alex, 1939 - 2021) wasn't this something JS discussed?? David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong
|
|
sandyk (RIP Alex, 1939 - 2021)
Global Moderator
Posts: 226
About Me: Retired ex Principal Telecommunications Technical Officer with 43 years at Telstra (Australia)
I am also a Moderator in Hi Fi Critic Forum
Electronics hobbyist for >65 years with DIY projects including Loudspeakers, Stereo FM tuner, S/W Regen Receiver, Superhet AM ,
Synchrodyne PLL AM tuner (Phase Lock Loop),Stereo Tape Deck, Amplifiers including I.C. types, Class A, Class AB 100W/Ch. (ETI5000) 240W/Ch. Mosfet (AEM6000) ,several DACs , numerous PSUs including VERY low noise (<4uV) types etc.for myself and friends
Audio Industry Affiliation: NIL
|
Post by sandyk (RIP Alex, 1939 - 2021) on Aug 29, 2019 9:15:10 GMT 10
I don't know what it means that LPS "block retrograde noise into the system" - LPSes can often suffer from current spikes & from both rectifier & supply capacitors plus the suffer from some level of current leakage. Hi John
I seem to recall (and I may be very wrong) that regarding noise from the pc, linear power supplies have greater noise rejection or filtering of that noise passing back upstream into the line compared to SMPS. This is quite apart from their own noise generation issues being claimed less than SMPS. Alex sandyk (RIP Alex, 1939 - 2021) wasn't this something JS discussed?? David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong Hi David Yes, E.E. John Swenson who designs the Uptone PSUs etc. did discuss this aspect. Even John Swenson's own Ultracap PSU does this too to a small extent, due to the combined SMALL capacitances of the switching transistors that provide isolation from the A.C. Mains supply. Alex
|
|
jkenny
Full Member
Posts: 83
About Me: Audio equipment designer forever in pursuit of more realistic & engaging music reproduction purely because of the extra enjoyment of music created by such reproduction.
http://Ciunas.biz
|
Post by jkenny on Aug 29, 2019 9:49:20 GMT 10
I don't believe leakage of noise back into the mains is really much of a problem now in audio devices? Swenson seems to suggest that current leakage from AC mains to DC output of power supply is much more sonically important
|
|
|
Post by Audiophile Neuroscience on Aug 29, 2019 10:01:08 GMT 10
I don't believe leakage of noise back into the mains is really much of a problem now in audio devices? Swenson seems to suggest that current leakage from AC mains to DC output of power supply is much more sonically important Yes but just saying if there is a noisy pc component on one psu rail you don't want that noise passing upstream and passing onto other rails (whether it passes back into the mains AC or not). LPSU supposedly do a better job at rejecting this noise but I guess it all comes down to implementation. Cheers David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong
|
|
jkenny
Full Member
Posts: 83
About Me: Audio equipment designer forever in pursuit of more realistic & engaging music reproduction purely because of the extra enjoyment of music created by such reproduction.
http://Ciunas.biz
|
Post by jkenny on Aug 29, 2019 19:05:37 GMT 10
I don't believe leakage of noise back into the mains is really much of a problem now in audio devices? Swenson seems to suggest that current leakage from AC mains to DC output of power supply is much more sonically important Yes but just saying if there is a noisy pc component on one psu rail you don't want that noise passing upstream and passing onto other rails (whether it passes back into the mains AC or not). LPSU supposedly do a better job at rejecting this noise but I guess it all comes down to implementation. Cheers David ----------------------------------------------------------------------------------------------------------------------------- "All music is folk music. I ain't never heard no horse sing a song." - - Louis Armstrong But this "noise passing upstream" would be via interconnected signal cables, not via power cables so I'm not sure how this relates to power supply? Admittedly the implementation of the power distribution system withing each device is crucial to keeping return paths for currents in as small a loop as possible & keeping signal returns away from power returns but again I'm not sure I can relate this to LPS Vs SMPS. But it is a complicated area especially with multiple interconnected audio devices/boxes
|
|
jkenny
Full Member
Posts: 83
About Me: Audio equipment designer forever in pursuit of more realistic & engaging music reproduction purely because of the extra enjoyment of music created by such reproduction.
http://Ciunas.biz
|
Post by jkenny on Aug 29, 2019 19:44:18 GMT 10
Hey, Alex, I see you now use a battery pack to power an iso-Regen Went over to the dark-side using batteries, I see?
|
|
sandyk (RIP Alex, 1939 - 2021)
Global Moderator
Posts: 226
About Me: Retired ex Principal Telecommunications Technical Officer with 43 years at Telstra (Australia)
I am also a Moderator in Hi Fi Critic Forum
Electronics hobbyist for >65 years with DIY projects including Loudspeakers, Stereo FM tuner, S/W Regen Receiver, Superhet AM ,
Synchrodyne PLL AM tuner (Phase Lock Loop),Stereo Tape Deck, Amplifiers including I.C. types, Class A, Class AB 100W/Ch. (ETI5000) 240W/Ch. Mosfet (AEM6000) ,several DACs , numerous PSUs including VERY low noise (<4uV) types etc.for myself and friends
Audio Industry Affiliation: NIL
|
Post by sandyk (RIP Alex, 1939 - 2021) on Aug 29, 2019 20:36:02 GMT 10
Hey, Alex, I see you now use a battery pack to power an iso-Regen Went over to the dark-side using batteries, I see? Hi John I have had for quite a few years. It's the best way to get away from the degradation with USB , even when there is a small amount of capacitive coupling to mains earth. When using a Linear PSU with an R-Core transformer I can hear the difference when connecting it's screen to mains earth. I actually have a toggle switch on a supply like this with a JLH PSU Add-on so that I can select either screen earthed or not. I only feel the need for a battery sourced supply with USB though, when using a normal USB Regen with a USB Memory stick plugged into it, and the Regen plugged into a front USB 2.0 port via a DIY USB-A to USB-B adaptor with both screen and +5V not connected through. With this setup, I can readily manipulate the sound of the files from dull to over detailed, or even a valve like warmth, simply by the choice of electro types in the capacitance multiplier section of the JLH No 1s or 0s are harmed in the process either. In fact, the same applies to Digital video , where I can create comparison music videos from SNL etc. where you can both see and hear differences at the same time between Videos with identical checksums. David has seen this through his own 4K Samsung TV too from comparison files on a USB memory stick. acg (Anthony) has also seen this from uploaded video comparison files when he played both versions side by side on his 4K monitor.
P.S I don't use USB for Audio other than this. I prefer well implemented Coax SPDIF .
|
|
|
Post by Audiophile Neuroscience on Aug 30, 2019 11:35:54 GMT 10
But this "noise passing upstream" would be via interconnected signal cables, not via power cables so I'm not sure how this relates to power supply? Admittedly the implementation of the power distribution system withing each device is crucial to keeping return paths for currents in as small a loop as possible & keeping signal returns away from power returns but again I'm not sure I can relate this to LPS Vs SMPS. But it is a complicated area especially with multiple interconnected audio devices/boxes
Oh, I assumed it could pass up one rail and go down another rail but by no means certain about this
David
|
|