XXHighEnd - The Ultra HighEnd Audio Player
October 22, 2018, 11:28:02 am *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: August 6, 2017 : Phasure Webshop open ! Go to the Shop
Search current board structure only !!  
   Home   Help Search Login Register  
Pages: 1 2 3 4 5 [All]
  Print  
Author Topic: Clean Power  (Read 35259 times)
0 Members and 0 Guests are viewing this topic.
christoffe
Audio Addict
***
Offline Offline

Posts: 756


View Profile Email
« on: November 18, 2011, 04:44:49 am »

http://www.whatsbestforum.com/showthread.php?2023-Standard-for-Clean-Power
Logged
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #1 on: November 18, 2011, 09:11:01 am »

Yes, Arthur Kelm is the expert on all of this. Indeed, I based my dedicated mains circuit (isolation transformer, ionic grouding rod, etc) on his ideas.

I'd love to know what he thinks about the use of SMPSs to power hifi components. My feeling is that he would say get the grounding right first, and then don't corrupt it with dirty SMPSs.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Bigear
Audio Loudspeaker
*
Offline Offline

Posts: 57


View Profile
« Reply #2 on: November 18, 2011, 03:20:34 pm »

From personal experience, I know having a good powersupply gives you a better sound, however understanding why is not so easy. All equipment I use has a trafo and rectifier circuit with caps, filtering, regulation etc. So the 'hot' side is floating relative to ground and independent of the AC-net. That leaves us the grounding when you us RCA mode, however when you play balanced, one should think the grounding is not that important anymore either...
Logged

i7 920, Asus MB, Windows 7 Home Premium 64 bit / 6 Gb RAM /no updates / music on SATAII  => XX V9z-6 / adaptive mode / XX buffer 4096 / Q1,2,3,4,5 = 5,0,0,0,0 / unattended / SFS 60 Mb memory straight contiguous / system clock 10ms / CPU Engine 3 / Threadprio RealTime / Playerprio Low / 16x Arc Prediction / Minimise OS / Peak Extend on / Phase invert off => NOS1 USB DAC => Aitos Pre and End amplifiers => PHY-HP KM30 open baffle.  NOT a single capacitor or coil from DAC chip to speaker
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #3 on: November 18, 2011, 03:44:29 pm »

Maybe it is ... But it is hard to judge for the real merits. I mean :

I sure see mains noise when things are not "connected" as should, and I see this through the analyser. Big problem : the analyser has to be connected as well. To mains and through interlinks (which I almost always have balanced). So, you never know what you really see, and one connection too many can create the problem ... which can be in-analyser only.

The other way around "works" too : the analyser "eliminates" the noise (because of its connection through PE etc.), and know you don't see it.

It really, really is tough.
Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #4 on: November 28, 2011, 11:18:13 am »

Well I've decided.

Computer switch mode power supplies are evil.

Right now, I'm sitting in my office with my office system (including a USB-NOS1) sitting right behind me. If I listen to FM radio (BBC Radio 3) with no electrical devices in the office switched on (other than my tuner, preamp and poweramp, all of which use linear PSs), the presenters' voices (live mic feeds) sound so pure and real. As soon as I switch on a PC (either my office work PC or my office music PC), a 'glare' is added to the presenters' voices. It's subtle, but unmissable once you know what to listen out for. Oh and I use Seasonic X-460 power supplies which are supposed to be about the lowest-noise computer PSs you can get.

I hear a very similar effect in my main listening room when I'm using my Atom-based PC with linear PS to play music and I switch on a secondary PC with SMPS - the 'glare' appears there too. Indeed, if I replace the linear PS with a SMPS, the 'glare' is there.

So I've decided that I will elimate all switch mode power supplies from both my office and main hifi systems. If this means that I have to live with a slower XX GUI, and even non-gapless playback, so be it. Now that I know it's there, I can't live with this SMPS 'glare'.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Gerard
Audio Addict
***
Offline Offline

Posts: 1064


WWG1WGA


View Profile
« Reply #5 on: November 30, 2011, 09:27:45 am »

Hey Peter,

Didn't you talk about some power suplie you made for the pc?

 Happy

Logged

Januari 2017 XX 2.05
OriginAE x11v Home build HTCP ~ Asus x79 mobo 3930K 12 core underclock 500 MHz,
16GB, *Windows 10 Pro 64 bit build 14393.0* *from RAM*, music on music server / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *0.10*  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core *3-5* / Not Switch Processors during Playback = Off/ Playback Drive *none* (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter Mid 705600 / -> USB3 (Silverstone both sides isolated = Sw#3 of NOS1a = Up) -> *Clairixa USB 15cm* -> Intona Isolator -> Clairixa USB 0.70m -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink ->> Metrum Amps ->> Metrum Acoustics ESL Open Baffle.
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #6 on: December 01, 2011, 10:27:54 am »

Yes I sure did. This project is postponed because of some necessary chip not doing what it pomises, and now the product is "dangerous" when used other than for what we intend it for (which is out of my control of course). So, not allowed to go to market (yet).

Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #7 on: July 13, 2012, 11:18:53 am »

I just wanted to resurrect this thread because it's so important and of so much relevance to other topics.

I'm no electrical engineer, and was pretty bad at electronics at school and uni. But here are my thoughts on what's going on with your AC mains. Please feel free to comment on any of the points below:

1. The Neutral of your AC mains is bonded to your house's PE.

2. In an ideal situation, the Neutral and the PE should sit at the same potential.

3. A dedicated PE can never truly be isolated from the house's PE. There will always be some finite resistance between them (MegaOhms) and therefore a potential difference.

4. It's a really, really bad idea to have some components (that are themselves connected together via interconnects, USB cables, etc) using the house's PE and others using a dedicated PE.

5. A dedicated PE on its own is pointless. You just ensure that you invalidate point 2.

6. The only way that a dedicated PE makes sense is to use an isolation transformer that allows you to bond the dedicated PE to the Neutral of the secondary.

Thoughts?

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #8 on: July 13, 2012, 12:54:57 pm »

I understand your points Mani .

But imo , the more attention we have to pay , when we connect gear together . For example a cd player to a preamp to an amp .

For decades , in the market hi-fi ( ...) we have seen amplifiers without an earth connection , and lots of differences in potential when we see each component ground .

So I like to connect well my Earth where there is a low impedance point in the chain - the amp . From there I try to arrange well the chain . There is not much to worry about , since you make sure the current would not go trough the imput .
By the way I did try transformer to "center" the ground , but didn't find that worth it , if yes audible .

Just my thoughts


s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
Scroobius
Audio Addict
***
Offline Offline

Posts: 1139


View Profile
« Reply #9 on: July 13, 2012, 02:16:25 pm »

Hi Mani - this is my understanding for what its worth but if anyone knows more please correct:

1. House (modern houses) PE is connected to the armoured sheath of the incoming mains supply cable. Neutral is bonded to PE and earth at the local substation (i.e. not in your house). House PE is ALSO connected to the main incoming water pipe i.e. to earth.

2. I would think that the ideal situation is that Neutral and PE should NOT sit at the same potential see below.

3. By dedicated PE I assume you mean lets say a direct connection into the earth either by a spike or water pipe - I would agree therefore that a dedicated PE cannot be isolated from the house PE because the house PE will be connected to earth at the incoming water pipe and/or at the local sub-station. There will always be the resistance of the earth connecting your dedicated PE to the house PE and it will vary substantially between different houses & locations.

4.
Quote
It's a really, really bad idea to have some components (that are themselves connected together via interconnects, USB cables, etc) using the house's PE and others using a dedicated PE

Yes agreed but the actual impact will vary greatly between different houses / locations. But it seems reasonable to assume that it is a *bad thing*.

5. I do not think that a dedicated PE should be pointless. Assuming a very good bonding of a dedicated PE to earth (i.e. like the one in your very own garden Mani) then it would not be unreasonable to expect that it there should be significantly less hash on it than exists on the house PE. BUT exactly what hash is there in your garden? - mmm how can we know that statement is correct without measuring it?

6. Isolation transformer. Well a standard isolation transformer? yes true it will allow your dedicated PE to be disconnected from your house PE. BUT does it? your dedicated PE is connected to earth in the garden - BUT your house PE is also connected to the earth in the garden via the water pipe and also back at the sub station. What is the effect of that? very dependent on you local situation how near is you dedicated PE to your water pipe? and how close is it to your sub-station?  also how wet is your garden? ha ha very wet at the moment.

Enter the Balanced Isolation Transformer i.e. a 1:1 mains isolation transformer BUT with a centre tap on the secondary that is connected to PE. This means that the output is +115v, PE and -115v (so 230v output with PE sitting in the middle). This has 2 potential benefits:

1. with output PE connected to house PE errors in mains noise should cancel - so if there is an error voltage on the input to the transformer it will appear positive on +115V and negative on - 115v so cancels (a normal transformer will just transfer the error voltage onto the +230v output). Note with a balanced transformer obviously the -115v output is totally disconnected from PE. Note that the bad things like mains droop and distorted waveforms that can occur on the mains supply should be significantly reduced on the output of the balanced transformer.

2. output PE is disconnected from "neutral".

Balanced transformers tend to be used by recording studios etc from what I understand. But they are not necessarily expensive if bought from a proper transformer supplier eg
http://www.airlinktransformers.com/cat10/61/standard_balanced_power_supplies.html

But how to connect a balanced power supply? if the output PE is connected to the house PE then hash from house PE (bad thing) but reduced mains noise is good thing.  Alternatively connect output PE to dedicated PE (garden spike) then mains noise cancelling may not be so effective but a cleaner PE. What is the best way to connect? my guess is that no one could predict it depends on your local situation. One thing that is good I suspect is that the -115v output is totally separate from PE with a balanced transformer - it cannot be with a normal isolation transformer.

My guess is that almost all balanced transformer installations are with the output PE connected to house PE. But and particularly with recording studios some considerable efforts will have been made to ensure that the "house" PE is of the highest quality. So maybe? that means connecting house PE to the dedicated PE - after it is just a more effective water pipe (but do not remove that water pipe connection as that is required by electrical regs). 

At least I think that is how it works  wacko

Paul
Logged

W10-14393.0 - RAM Disc XXHighEnd Mach II Stealth LPS PC -> Xeon E5 2040v4 with Hyperthreading On (20 cores)* @~720MHz, 32GB, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1/ Q1Factor = *1* / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *3.0*  (max 120) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = 43 / Nervous Rate = 100 / Cool when Idle = n.a / Provide Stable Power = 0/ Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter low(12220) 705600 / -> USB3 from MoBo -> Lush USB 70cm -> Phisolator -> 24/768 Phasure NOS1a G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink -> Orelino MkII Active Open Baffle Horn Speakers.
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #10 on: July 13, 2012, 03:26:59 pm »

Hey Paul, thanks for sharing. I'll come back to you once I've had a chance to digest all your points.

Balanced transformers tend to be used by recording studios etc from what I understand. But they are not necessarily expensive if bought from a proper transformer supplier eg
http://www.airlinktransformers.com/cat10/61/standard_balanced_power_supplies.html

Yeah I know. For the last 6 months or so I've been using a pair of 5KVA balanced trannies (one for my main system, one for my study/office system) after my 15KVA isolation trannie. I haven't played around with various configurations yet (just no time), but may well bypass the isolation transformer altogether if it's not helping. Furthermore, I now have a PurePower 2000 regenerator in the mix after the balanced trannie for my study/office system. I've just today ordered a new PurePower+3000 regenerator for my main system. These things do something wonderful to the sound, provided everything is plugged into them (including the PC). I'm not sure if it possible to achieve what they do in any other (cheaper) way, but if there is, I've not found it.

Back with more thoughts later.

Mani


* 3 transformers.jpg (298.61 KB, 775x1037 - viewed 567 times.)
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #11 on: July 13, 2012, 03:28:44 pm »

Quote from: Scroobius [url
http://www.airlinktransformers.com/cat10/61/standard_balanced_power_supplies.html[/url]

But how to connect a balanced power supply? if the output PE is connected to the house PE then hash from house PE (bad thing) but reduced mains noise is good thing.  Alternatively connect output PE to dedicated PE (garden spike) then mains noise cancelling may not be so effective but a cleaner PE. What is the best way to connect? my guess is that no one could predict it depends on your local situation. One thing that is good I suspect is that the -115v output is totally separate from PE with a balanced transformer - it cannot be with a normal isolation transformer.

Boy this is on topic for me right now, clean mains is the most important issue here at the moment. Paul has helped me immensely by highlighting problems with my system that have been plaguing me for months during our recent visits to listen to one another’s systems.

Following this I have been doing some basic testing with the house mains supply by switching off the mains phases in the house that are not supplying to the music systems and then also turning off the other household circuits on the supply phase that feeds the hifi. Early results seem to confirm this is good direction to be looking at.

Following Pauls mentioing balanced and putting me onto Airlink I have been looking at the balanced supply approach carefully and speaking to AirLink. Seems like the first thing to get right is a configuration that sensibly addresses electrical safety. The issue being that when the mains supply is configured to be balanced both the neutral and live connections carry 120v ac whereas the safety measures in a house are in place to trap faults on a single live feed system.

I plan to fuse both the live and neutral connections as they enter into all equipment connected to the balanced supply. Also I really want to have a ganged RCD that disconnects in the output of the balanced supply that isolates both 120v phases of the supply is the case of a fault condition on either phase. This will prevent the danger of having say a 120v “neutral” balanced feed into a hi fi component even when the “live” fuse in the component has blown.

Regards the earthing scheme the plan at first is to use house PE on the balanced side to start with. I know this may not be optimal but again I want to be sure of all the implications of setting up a separate earth on the balanced mains side before going ahead.

There are certainly some very positive comments about sound quality of Balanced supply out there together with descriptions of the sort of problem that I have been having which were addressed by other by moving to Balanced supply. If I can get the electrical safety worked through then I am ready now to order a balanced transformer. Will let you know how it goes.

If anyone has already setup a balanced supply it would be geat to hear how it was done.


Best Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #12 on: July 13, 2012, 03:34:21 pm »

Quote from: manisandher

Yeah I know. For the last 6 months or so I've been using a pair of 5KVA balanced trannies (one for my main system, one for my study/office system) after my 15KVA isolation trannie.


Quote from: Nick

If anyone has already setup a balanced supply it would be geat to hear how it was done.



Bingo !  Happy

Our posts crossed over  Happy

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #13 on: July 13, 2012, 03:38:04 pm »

Hey Nick, yep they did.

I think you're absolutely right to be concerned about the safety of these devices in a domestic setting. I may well bypass them unti there is a common consensus on how they can be used safely.

But I will definitely spend a day or so trying different direct/isolation-trannie/balanced-trannie/regenerator configurations and report back what I find. It could be that simply sticking with the house mains/PE is the best way to go. (But I doubt it...)

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #14 on: July 13, 2012, 04:11:01 pm »

I plan to fuse both the live and neutral connections as they enter into all equipment connected to the balanced supply. Also I really want to have a ganged RCD that disconnects in the output of the balanced supply that isolates both 120v phases of the supply is the case of a fault condition on either phase. This will prevent the danger of having say a 120v “neutral” balanced feed into a hi fi component even when the “live” fuse in the component has blown.

This won't be an issue for me as I will now use a PurePower regenerator to feed all the components. This will take the balanced +115V/-115V and reconvert it back to 0/+230V. If there is a fault condition on either phase feeding it, the PP will simply kick into battery mode.

But the isolation- and/or balanced-trannies will not be for nothing even with a PP in place. Other PP users have reported that the better the incoming AC into the PP, the better the sound. Only with the PP running off battery does the sound really become totally independent of the AC (of course).

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #15 on: July 13, 2012, 04:48:14 pm »

... I really want to have a ganged RCD that disconnects in the output of the balanced supply that isolates both 120v phases of the supply is the case of a fault condition on either phase. This will prevent the danger of having say a 120v “neutral” balanced feed into a hi fi component even when the “live” fuse in the component has blown.

Hey Nick, I've just checked and my Airlink 5KVAs have these installed as standard - one on the input, one on the output. HTH.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #16 on: July 13, 2012, 07:46:55 pm »

By the way I did try transformer to "center" the ground , but didn't find that worth it , if yes audible .

Hi Stefano. Audible in a good way or bad? The main criticism of transformers is that they can sometimes seem to compress dynamics and make the sound a bit dull/lifeless. Did you find this?

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #17 on: July 14, 2012, 03:42:15 pm »

... I really want to have a ganged RCD that disconnects in the output of the balanced supply that isolates both 120v phases of the supply is the case of a fault condition on either phase. This will prevent the danger of having say a 120v “neutral” balanced feed into a hi fi component even when the “live” fuse in the component has blown.

Hey Nick, I've just checked and my Airlink 5KVAs have these installed as standard - one on the input, one on the output. HTH.

Mani.

Mani,

Thanks for mentioning this, I think I will go for a 3 or a 5 kva version like yours. Do you get any audible hum from your units ?

Best

Nick
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #18 on: July 16, 2012, 11:07:17 am »

Hi Nick, I think there is an audible hum from the unit, but sitting as it does in the cellar I don't really pay much attention to it. What I do know is that it's much quieter than the 15KVA 'yacht hull isolation' transformer sitting below it.

There is one thing I'd do if I were you though. AirLink claims, "The Balanced  Power Supply creates a new ground called a derived ground, that decouples equipment connected to it from the circulating ground currents in the rest of the house wiring.  It also provides one common ground reference for all the equipment connected to it."

One of my 5KVAs has the input ground connected to the output ground. The other has the two grounds isolated, although even in this case, there is still a ~6MΩ resisitance between them. Now I have no idea how you 'derive' a new ground from the house ground, but even with my dire knowledge of electronics I know that this simply cannot be the case if the input and output grounds are connected together. It seems to me that someone at AirLink isn't quite sure how to do it either, so I'd check with them before ordering. HTH.

I won't be able to test various power configurations until some new junction boxes that I've ordered come through. But I'll let you know what I find as soon as I get them.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
CoenP
Audio Addict
***
Offline Offline

Posts: 810


View Profile Email
« Reply #19 on: July 16, 2012, 11:43:41 am »

Hi,

I'm breaking in here since this thread got me thinking. No, not about balanced isolation transformers, I'm far from that point yet, but about having the PE (not peak extension) of the NOS1 and PC on the same connection.

Since my computer PSU blew up, I've been having difficulty with getting the sound back I was used to. I tried several PSUs that should even improve on things but all I got was worse sq.

Then in a flash of Zen I swapped the powercable from the PC that sounded fine even on its mobo codec (the ATOM) to the XX music PC and voilá very nice sound again, I dare say even better than before. And this was just a standard cable.

So what was the matter? I inspected the power cable and it looked a little worn on the euroconnector part. Measuring the wires showed that the cable was not making any contact the PE pin!

Now that the PE is tied to the PE of the powerstrip and hence to the NOS1, the SQ issue is solved. Interesing though is that this loose PE connection did not bother me with the other -blown- supply (or did it blow because of this defective cable?).

I am shure there is more to be gained on this topic, but so is in optimising XX!

regards, Coen
Logged

Audio PC: XXHE PC v1 with RAMdisk w.o. videocard and 1 of 2 cpu fans + BRIX/USB3 storage musicserver. Power cable PE not connected, together with nos1 and poweramp in separate "audio" powerstrip.

Lush 1m, Phasure NOS1a-75B G3 USB (buf 16 ms)-> Blaxius ->SE EL95 (0,8W triode) + cheap link to Abaqus plateamps> biwired QED cable-> Bastanis Mandala Duo (upgraded).

[other sources: TD124/3009SII-i/Grace F9/lounge LCR phono; Rega Planet 1997 vintage]
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #20 on: July 16, 2012, 11:58:47 am »

Hey hey ...

To make it somewhat more complicated and in case you did not know : The NOS1 is only connected to PE by means of the case (or : PE is connected to the case and nothing else in there).

But it isn't that simple, because disconnecting (cutting) the PE will make all different, if you only know what to look and measure for. This is all about antenna working, capacitive loads and more sh*t, with always the USB connection in the middle of all.

Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #21 on: July 16, 2012, 12:48:14 pm »

Hi Mani,

Quote
Now I have no idea how you 'derive' a new ground from the house ground, but even with my dire knowledge of electronics I know that this simply cannot be the case if the input and output grounds are connected together.

What will happen in this case (of half truths ?) is that the transformer will create a now ground (reference) to the house ring. But again it is more complicated ...

In order to let you understand I will try to give an example which is not about this at all, but from there you should be able to reason out the remainder, or even measure it;
Btw, I do this by heart now, and I hope I'm not mixing up things, bete here to, at applying the practice (of measuring it) you will see what to do exactly automatically ...

Since you should be able to find two separate ground references (like the earth pin from your house ring and the one from the isolated transformer), use those both to find out some interesting things. I thing DV Voltage measurement now, because it is about that (but allpying an AC measurement here and there may also be interesting, so don't forget to do that too). Now :

Hook up your DC Voltmeter to the two ground references. You should see something. Now, start switching on house devices like phone adapters or motors or possibly amplifiers and the like. Watch the DC Offset change with it.
Now envision that this DC Offset mangles with at least your house ring ground reference. So, mid point e.g. goes up, and think about the part of the (hot point) wave above zero (which should be lower now in amplitude ) and the one under it (should be higher). Here you are turning that sine into something else.
In the mean time your AC output changes too. I never looked at it on the analyser, but while net the result should be the same, it won't because of the wave shape and let's say (for easily envisioning it) clipping at one side.

This all is about your consuming devices fireing back to the mains because of they changing the ground reference.
It may not be an official thing, but this is how I once could look at it. So, that consuming device is not a "balanced device" like in "it draws from both the phases separately" (see earlier posts about this - the 115v/115v thing) but as how I look at it, it will draw the ground towards "itself". So, where ground is to be 0V (which is a virtual thing always, but works for the physics in order) and a device is consuming the 230V, the ground goes up by, say 5V. I know, for AC this can't work, but I think something like switching power supplies ay use the positive side of the wave only or more than the negative, which depends on how the AC/DC conversion is done.

I never worked this out really, but saw it happening and now try to see how it can happen. But it does anyway.
It goes as far as that when devices are switched on subsequently, the Offset always becomes more and more. No new device makes it less again.
In that other larger topic about actually the same subject, I think I mentioned that I once could measure 109V AC between the two grounds. However, this was/is groundloop related too, because (back then) I was able to end up in the exact same situation (devices switched on) with 0V AC.
And when that was the case, a few days later it was there again. It is here when I stopped using the PC on its dirty ring (which in my case wasn't even the house ring but yet again another separate one - for the PC to make dirty), because this can't be a good thing.

No such thing happens when using one ring only, but hey, that is because that other reference is unused (now no current or whatever can flow in between them, and btw 6MOhm *really* is sufficient) but now you also won't be able to see the working of -for example- the PC (etc.) hammering on the same ground reference now directly.

(Ehm, after writing this it looks like a made a few mistakes already, and I now think that you must approach it the other way around : have that clean reference (no devices connected to it) to proove how another device connected to the other ground influences that other ground by looking at the diffrerence between the two grounds. So, switch off/on the PC for example)

It is a quite confusing subject, but when you have two ground references you can find out what is going on by means of some careful thinking.
With your PE-PE connected transformers it will be similar to what I described : they will form a ground all right, but it will level with whatever that other ground is. So, assumed your transformer created ground is really zero (and funnily enough this is not a virtual ground but a real one because really zero), while the ring it is connected to is 5V, think like all now being 2.5V for easy thinking. So, transformer got worse, and house got better. Your transfomer now has a DC Offset of -2.5V.


Maybe I suddenly know how to better explain it :
Switch on the DC Metering of the NOS1. It will show ~5mV. Now play a nice bass piece and see the offset change on the rythm of the playing. What happens is that the plus and minus side are not equelly influenced; thinking transients, the plus amplitute of AC which is what music is, will be larger than the minus. How does that show on the metering ? as DC Offset. Now compare this with my switching PSU example, and how that will influence the DC Offset hence wave shape in general. It only needs thinking about here the "amplitudes" are consumed from both plus and minus - and not equally (pushing DOWN the DC level as how it PHYSICALLY will be ), while with music playing they are stuffed through (LIFTING the DC level as how it is MEASURED (which is just averaging)).

And Mani, before you get confused : I am pretty sure that the DC Metering in the NOS1 shows opposed to reality. So, a more minus offset in reality is a more plus (from a transient and assumed absolute phase is all right). So, assumed you have a slight minus offset showing (which is the case for most NOS1's) you will see it more minus at low frequency transients. But in relatity is it plus; it's just where we connected the wires to, which is by accident the other way around (not really by accident, but just unimportant as we see it).

Was this smart or totally stupid ?
Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #22 on: July 16, 2012, 06:32:03 pm »

I am shure there is more to be gained on this topic...

I'm sure you're right. I remember a while ago, with the PC, NOS1 and amp all connected to the same PE, I tried connecting the chassis of the PC to PE too. And guess what? The sound totally changed. At first, I thought it was an improvement, but over a period of time I found it fatiguing. I now think I was getting false detail and brightness through a ground loop somewhere. Or maybe the chassis was now acting as an RF antenna? Who knows...

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #23 on: July 16, 2012, 06:35:19 pm »

Was this smart or totally stupid ?

Hey Peter, I think even Einstein would need some time to figure this post out... I'll do my best later this evening.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #24 on: July 16, 2012, 07:49:54 pm »

Hi,


Then in a flash of Zen I swapped the powercable from the PC that sounded fine even on its mobo codec (the ATOM) to the XX music PC and voilá very nice sound again, I dare say even better than before. And this was just a standard cable.



regards, Coen
Just want to add , that also the lenght of the cable make differences .

S
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #25 on: July 16, 2012, 08:00:46 pm »

Ehm, after writing this it looks like a made a few mistakes already, and I now think that you must approach it the other way around : have that clean reference (no devices connected to it) to proove how another device connected to the other ground influences that other ground by looking at the diffrerence between the two grounds. So, switch off/on the PC for example.

OK, I've made some measurements between my dedicated PE (measured by electrician as <5Ω resistance to earth) and my house PE (no idea of what this resistance to earth is):

1) with small load (only NOS1 switched on):
- resistance = 2.5MΩ
- DC = 0.25V
- AC = ~0.6V (varies a little, not totally constant)

2) with ~1000W load (Sauermann amps switched on)
- resistance = 2.5MΩ
- DC = 0.25V
- AC = ~0.6V

No, not a 'copy and paste' mistake, but absolutely no difference whatsoever.

In case it's importanct, the whole AC chain here is: house electricity meter -> dedicated consumer unit -> 15KVA isolation transformer -> 5KVA balanced transformer -> PurePower2000 regenerator -> 8-way distribution unit

The dedicated PE consists of: ionic grounding rod and a star earthing arrangement to all components (including PC)

The PurePower2000 uses a switching power supply, which I assume kicks a lot of noise back into the AC feeding it. But the AC feeding it is balanced, which I suspect helps.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #26 on: July 16, 2012, 08:44:51 pm »

Hi Nick, I think there is an audible hum from the unit, but sitting as it does in the cellar I don't really pay much attention to it. What I do know is that it's much quieter than the 15KVA 'yacht hull isolation' transformer sitting below it.

There is one thing I'd do if I were you though. AirLink claims, "The Balanced  Power Supply creates a new ground called a derived ground, that decouples equipment connected to it from the circulating ground currents in the rest of the house wiring.  It also provides one common ground reference for all the equipment connected to it."

One of my 5KVAs has the input ground connected to the output ground. The other has the two grounds isolated, although even in this case, there is still a ~6MΩ resisitance between them. Now I have no idea how you 'derive' a new ground from the house ground, but even with my dire knowledge of electronics I know that this simply cannot be the case if the input and output grounds are connected together. It seems to me that someone at AirLink isn't quite sure how to do it either, so I'd check with them before ordering. HTH.

I won't be able to test various power configurations until some new junction boxes that I've ordered come through. But I'll let you know what I find as soon as I get them.

Mani.

Thanks Mani,

I went ahead and ordered a 3kva transformer from Airlink today. I went for this size purely due to size and the thought that there is no way my system is going near 13 amps of mains voltage draw so I hope this should suffice. If hum is not a problem it will live in my listening room so the 5kva model would rather big.

Air link reckoned that the default PE arrangement  for the unit is that the house PE is connected through to the output ground. They did say this link can be broken to setup your own bespoke PE ground. My plan is to try with the house PE setup first and see what happens.

I have a three phase mains supply into the house and I'am concerned that the load on other phases may also be influencing the PE / neutral potential of the phase used for hifi, particularly after Peters earlier post. Tests switching off phases seem to confirm some interactions between them. I am not sure if the balanced transformer will help here, I can think of arguments for and against but best just to try it and listen I guess.

The Airlink "derived ground" worrys me a bit, I hope they do not mean that the out put is not referenced an external low impedance ground (house or dedicated bespoke earth). Humm I feel the need for another call tommorow with Airlink to understand this.

Did the 5kva transformers come with any Airlink documentation ? There is not much on their website.

Best Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #27 on: July 16, 2012, 09:15:57 pm »

The Airlink "derived ground" worrys me a bit, I hope they do not mean that the out put is not referenced an external low impedance ground (house or dedicated bespoke earth). Humm I feel the need for another call tommorow with Airlink to understand this.

I think this would be prudent. As I said, in one of the transformers I received, the input/output grounds were both conneced, in the other they were not. Which did they mean? I don't know. But if it's the latter, there had better be a connection made pronto if there's a serious fault in the appliance.

Did the 5kva transformers come with any Airlink documentation ? There is not much on their website.

No it didn't, which was a pain. But I intended to connect the transformer's output ground to my dedicated PE anyway so wasn't too bothered.

I've just checked how loudly the 5KVA hums, and it's actually fairly quiet with a low load. It get's considerably louder with my ~1400W RMS though.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #28 on: July 16, 2012, 09:20:35 pm »

In case it's importanct, the whole AC chain here is: house electricity meter -> dedicated consumer unit -> 15KVA isolation transformer -> 5KVA balanced transformer -> PurePower2000 regenerator -> 8-way distribution unit

Well I've just had my own 'Zen moment'. I'm gonna try the following:

electricity meter -> dedicated consumer unit -> PurePower2000 regenerator -> 5KVA balanced transformer (house PE on input, dedicated PE on output) -> 8-way distribution block

Don't think I really need the 15KVA isolation transformer. Anyone want one? It's 120Kg though!

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #29 on: July 16, 2012, 11:15:00 pm »

I've just checked how loudly the 5KVA hums, and it's actually fairly quiet with a low load. It get's considerably louder with my ~1400W RMS though.

That's good to know thanks. I think I might hum a little louder though if I had to supply 1400w rms  Happy . Its nice setup in your cellar though, really no problems if the transformers decide to sing along a little  Happy

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #30 on: July 16, 2012, 11:31:35 pm »

By the way I did try transformer to "center" the ground , but didn't find that worth it , if yes audible .

Hi Stefano. Audible in a good way or bad? The main criticism of transformers is that they can sometimes seem to compress dynamics and make the sound a bit dull/lifeless. Did you find this?

Mani.

Hi Mani ,

the time I did try that it was with a big non feedback amp in depletion mode Mosfets (that amp itself has a big Power Supply with double full waves rails) . So the "balanced" transformer was at the source , for the CD player so to speak .
The sound was very "big" and dynamic . That is what "typical" (?not so typical)  single ended amps without feedback do lol.
So no , I would not say it was like the dynamics were compressed .... it was all more equilibrate and precise .
It was really a good listening . Considering that the amp was connected the Earth ( and I do not play music without ) , I didn't really investigate on the " center " of the transformer .But if I remember well , it was not at the same potential of the Earth , but imo it was safe to connect like that .
It was a nice setup anyway .These  days I listen to 3 lovely amps ( feedback , non feedback various topologies , but all single ended which I love ), and all have different PS arrangements . So saying a definitive word about the transformer would be a bit  too much this side , but it was "different" somehow .
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #31 on: July 17, 2012, 09:06:15 am »

Stefano, just curious ... With SE you mean that your amps can take RCA input only ?
(by now we can go a couple of directions with SE vs. "Differential")
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #32 on: July 17, 2012, 10:22:11 am »

Stefano, just curious ... With SE you mean that your amps can take RCA input only ?
(by now we can go a couple of directions with SE vs. "Differential")

Hi Peter ,

yes indeed , RCA only . For years I have been a fan of the balanced operation with amps and dacs , but now I am more SE oriented ...but I know , there are some balanced configuration that "sounds" like being in paradise  Happy

s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #33 on: July 17, 2012, 10:42:31 am »

I hope it is not too much off topic (okay, it is), but why exactly do you feel SE is better (soundwise, technical ideas about it) ?
Keep in mind, I am NOT suggesting at all that differential is better !

Curious ...
Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #34 on: July 17, 2012, 11:10:13 am »

I hope it is not too much off topic (okay, it is), but why exactly do you feel SE is better (soundwise, technical ideas about it) ?
Keep in mind, I am NOT suggesting at all that differential is better !

Curious ...
Peter
It's is a matter of effects probably , and since I can play with amplifiers with non clinical stages and feedeback ,usually one stage or max 2 stages, and  low distorsion ( between second and third harmonics mainly ) ...I think this has come into a choice of tastes , but also pratical .Pratical becouse single ended is sometimes simpler  Happy
I also used to be a musician playing the accordeon ... the difference between second and third harmonics is still an interesting phenomena...
Of course , balanced operation permits other fantastic permutations and topologies , for example the Supersymmetry (tm) which indeed deserve to be explored and listened ...
So let's say its not an "exact" thing ; some analogies are tremendously close however .
But remember , multistage and waste of gain for the feedback are out of the way for me , and don't even think to amplify with BJT's .

s

Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
juanpmar
Crazy Audiophile
****
Offline Offline

Posts: 1230


View Profile Email
« Reply #35 on: July 17, 2012, 11:14:25 am »

Hi,

Mick Maloney owner of Supratek amplifiers, the ones I used some time ago, has this opinions about Balance/Unbalanced operation:

Balanced operation


I often get asked about the advantages of balanced operation-my reply is based on the following:
 
-Balanced/unbalanced is a conentious subject. It comes from the pro audio world where they need it to balance out the noise pickup from very long cables.

-True balance involves either using balancing transfomers or turning a Single Ended cicuit (as most preamps are) into a push pull circuit with double the components, circuitry etc- this is the only way to achieve "true" balanced throughout the preamp.

-There's a lot of hype about it, with a lot of so called balanced operation being psuedo balanced, which is basically having an XLR socket wired to accept a balanced signal and turning it back to unbalanced.
Personally I prefer Single Ended operation every time- there's an ease and effortlesss about it, wheras true balanced sounds more electronic to me, especially the push pull types of operation.

-One area where it is useful is in helping with RF noise in the big cities- it is not an isue for me in my country town isolation, but places like NY benefit from it and in my mind the best way to use balanced is to take it off the output transformer of a preamp, which fortunately we can do with the Supratek pres- it is true and perfect balanced with the minimum of circuitry to degrade the sound. (there's virtually no circuitry!)

-One balancing point in a preamp is all that is necessary- the output transformer turns the single ended circuit into a balanced circuit with two phases and the noise common to both channels (hum, RF etc) is greatly reduced.

-Going for balanced throughout is sonic overkill that does more harm than good IMO- it has been a successful marketing campaign by some companies, but it is marketed to sell amps.

I personnaly use XLR in my Spectron amps. I live in downtown Madrid sorrounded by offices and computers. My interconnection cables are also quite long. At the end I like more the sound with XLR than with RCA.


Best regards,
Juan
Logged

Audio Pc: Processor i7 970: 3200MHz (reduced to 1668MHz), 6+6 cores/ RAM Corsair DDR3, 24Gb, 1333MHz/ Mb Asus X58 Sabertooth/ OS and XXHE in Peter's RAM-Disk / The CPU fan is the only one in the Audio Pc: NF-S12A (600rpm/6.7db)/ No graphic card/ Power supply: Seasonic SS-400FL2, fanless.

Configuration and Updates in HOW I'VE BUILT MY NEW PC...http://www.phasure.com/index.php?topic=1673.0

XXHighEnd: 2.10. Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN/ KS:Phasure NOS1 Out 4.0/ #4 Engine/ Adaptive Mode/ Q1=30 Q3=1, Q4=1, Q5=1, xQ1=40/ Dev.Buffer: 4096/ ClockRes: 15ms/ Straight Contiguous/ SFS: 5.19 (max 120)/ Not Invert/ Phase Alignment Off/ Allow format change/ Decode HDCD/ Playerprio: Low/ ThreadPrio: Real Time/ Scheme: Core 3-5/ UnAttended/ Not Switch during Playback Off/ Playback Drive none/ UnAttended/ Include Garbage Collect/ Copy to XX Drive by standard/ Always clear Proxy before Playback/ Stop Desktop Services/ Stop Remaining Services/ Stop All Services: Off/ Keep LAN Services: On - Persist: Off/ Use Remote Desktop/ Minimize OS/ XTweaks : Balanced Load 52/ Nervous Rate 10/ Cool when Idle -/ Provide Stable Power 0/ Utilize Cores always 1/ Time Performance Index: Optimal / Time Stability: Stable / Arc Prediction.

Music Server PC (W10) totally silent with OS (W10) in SSD and music inside in SSDs - RDC > Ethernet Gigabyte cable 3m > Audio Pc > 1m USB Lush cable directly from the USB3.0 in the motherboard > PHASURE NOS1a-75B-G3 (Driver v1.0.4) 16ms > Blaxius BNC interconnects > Genelec 1037B 3-Way Active speakers with BNC inputs
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #36 on: July 17, 2012, 12:03:33 pm »

Hi,

Mick Maloney owner of Supratek amplifiers, the ones I used some time ago, has this opinions about Balance/Unbalanced operation:

Balanced operation


I often get asked about the advantages of balanced operation-my reply is based on the following:
 
-Balanced/unbalanced is a conentious subject. It comes from the pro audio world where they need it to balance out the noise pickup from very long cables.

-True balance involves either using balancing transfomers or turning a Single Ended cicuit (as most preamps are) into a push pull circuit with double the components, circuitry etc- this is the only way to achieve "true" balanced throughout the preamp.

-There's a lot of hype about it, with a lot of so called balanced operation being psuedo balanced, which is basically having an XLR socket wired to accept a balanced signal and turning it back to unbalanced.
Personally I prefer Single Ended operation every time- there's an ease and effortlesss about it, wheras true balanced sounds more electronic to me, especially the push pull types of operation.

-One area where it is useful is in helping with RF noise in the big cities- it is not an isue for me in my country town isolation, but places like NY benefit from it and in my mind the best way to use balanced is to take it off the output transformer of a preamp, which fortunately we can do with the Supratek pres- it is true and perfect balanced with the minimum of circuitry to degrade the sound. (there's virtually no circuitry!)

-One balancing point in a preamp is all that is necessary- the output transformer turns the single ended circuit into a balanced circuit with two phases and the noise common to both channels (hum, RF etc) is greatly reduced.

-Going for balanced throughout is sonic overkill that does more harm than good IMO- it has been a successful marketing campaign by some companies, but it is marketed to sell amps.

I personnaly use XLR in my Spectron amps. I live in downtown Madrid sorrounded by offices and computers. My interconnection cables are also quite long. At the I like more the sound with XLR than with RCA.


Best regards,
Juan

Personally I refer to Balanced Amps ( or preamps or dacs ) or Single Ended amps . Knowing that by topology . There are push pull designs which are not balanced , no matter if there is or not XLR connections at the imput or transformers somewhere .
OTHO , there are Balanced Single Ended amps , they accept both RCA and XLR inputs ( no transformer ), and their output are balanced , and they act in push pull also .
An example : the differential pair . The exploitation of a differential pair with local feedbak is called supersymmetry .
Another example : running 2 monoblocks with balanced signals . In this case , -no connection- is taking place between the two amps , if not the speaker itlself .

 Cool
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #37 on: July 17, 2012, 12:10:35 pm »

OTHO , there are Balanced Single Ended amps , they accept both RCA and XLR inputs ( no transformer ), and their output are balanced ,and they act in push pull also .

The highlight is mine.

And there are single-ended amps (i.e. not push/pull) that are true balanced designs. My old Pass Labs Aleph4 was one. A very nice-sounding amp indeed. But a bit 'too polite' for XX/NOS1 duties.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #38 on: July 17, 2012, 12:16:10 pm »

OTHO , there are Balanced Single Ended amps , they accept both RCA and XLR inputs ( no transformer ), and their output are balanced ,and they act in push pull also .

The highlight is mine.

And there are single-ended amps (i.e. not push/pull) that are true balanced designs. My old Pass Labs Aleph4 was one. A very nice-sounding amp indeed. But a bit 'too polite' for XX/NOS1 duties.

Mani.

I know the Aleph 4 . It has the differential pair at the input ( XLR and RCA ) , but the rest is single-ended , infact the speaker referes to ground by design scheme . The Aleph current source is still a modern invention imo .
Very nice piece indeed .

s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #39 on: July 17, 2012, 12:27:28 pm »

Ah, I didn't know this. I sold it a couple of years ago and the guy who came to pick it up just couldn't understand why I would sell such a sweet sounding amp.

But one question: Is possible to have a truly balanced single-ended poweramp? I mean, my Pass X1 preamp is such a design, and if it's possible in a preamp surely it must be possible in a poweramp also, no?

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #40 on: July 17, 2012, 12:29:12 pm »

Ah, I didn't know this. I sold it a couple of years ago and the guy who came to pick it up just couldn't understand why I would sell such a sweet sounding amp.

But one question: Is possible to have a truly balanced single-ended poweramp? I mean, my Pass X1 preamp is such a design, and if it's possible in a preamp surely it must be possible in a poweramp also, no?

Mani.

Yes , sure .

s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #41 on: July 19, 2012, 12:11:07 pm »

Is possible to have a truly balanced single-ended poweramp? I mean, my Pass X1 preamp is such a design, and if it's possible in a preamp surely it must be possible in a poweramp also, no?

Yes , sure .

This seems to contradict something that Juan quoted earlier:

Mick Maloney owner of Supratek amplifiers, the ones I used some time ago, has this opinions about Balance/Unbalanced operation:

Balanced operation

-True balance involves either using balancing transfomers or turning a Single Ended cicuit (as most preamps are) into a push pull circuit with double the components, circuitry etc- this is the only way to achieve "true" balanced throughout the preamp.

Doesn't it?

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #42 on: July 19, 2012, 12:30:02 pm »

I'm gonna try the following:

electricity meter -> dedicated consumer unit -> PurePower2000 regenerator -> 5KVA balanced transformer (house PE on input, dedicated PE on output) -> 8-way distribution block

Spent until the early hours of the morning doing this. (The input ground of the transformer is connected to the house PE and the output ground to my dedicated PE.) I think it was a worthwhile redistribution. I'm getting the same beautiful balance between micro-detail and smoothness as I did when I initially reinstated the PurePower2000 last week, but strangely I'm getting what sounds like a deeper (and more satisfying) low-end. The only thing I'd like more of is a hint more 'sparkle'. But I haven't yet optimised 'Le Monster' (liquid cooler still needs to go in, all fans need to come out and I need to look at BIOS settings)... and of course there's still 0.9z-7 to come!


There's a very interesting review of the top-of-the-line Equi=tech product here: http://www.6moons.com/audioreviews/equitech/1.html. It makes a very strong case for balanced power and even mentions what Peter was talking about (in his long post earlier in this thread). My main concern now is that my Airlink balanced transformers may not be of the quality really required for a truly low-noise balanced system - I doubt they go to the same lengths as Equi=tech. Also a friend of mine who has a great sytem (centred around the dCS Scarlatti) once mentioned that he heard one of the most profound changes to the sound of a system when an Equi=tech was introduced.

Don't think I really need the 15KVA isolation transformer. Anyone want one? It's 120Kg though!

I may need to sell a few things in order to scr*pe enough pennies for an Equi=tech.

On a final note, my 'noise sniffer' doesn't seem to like my current AC setup. It shows quite a high noise reading even with zero load. Anyone have any idea why this might be?

Mani.

EDIT: What's wrong with the word 'scr_a_pe'? It just means 'to rub away'...
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #43 on: July 19, 2012, 01:24:54 pm »

Quote
Anyone have any idea why this might be?

Hey Mani - most stupid answer : because it is so ?
Maybe not stupid when you think it is a false alarm of the meter, but after the sentence about your description of the sound, my immediate idea was : so, thus noisy.

If you meant to ask how that noise can be there, then I for sure don't have an answer to that. Pickup ?


PS:
Quote
It makes a very strong case for balanced power and even mentions what Peter was talking about
Is it allowed to know what this was ? I mean, maybe I am going to read that article, but reading my own posts ?
nea
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #44 on: July 19, 2012, 01:37:15 pm »

Is possible to have a truly balanced single-ended poweramp? I mean, my Pass X1 preamp is such a design, and if it's possible in a preamp surely it must be possible in a poweramp also, no?

Yes , sure .

This seems to contradict something that Juan quoted earlier:

Mick Maloney owner of Supratek amplifiers, the ones I used some time ago, has this opinions about Balance/Unbalanced operation:

Balanced operation

-True balance involves either using balancing transfomers or turning a Single Ended cicuit (as most preamps are) into a push pull circuit with double the components, circuitry etc- this is the only way to achieve "true" balanced throughout the preamp.

Doesn't it?

Mani.

Single Ended Balanced is what we could call a "scaled up" X1 design . Or take a look at the " son of zen " . A differential pair is something "more" than doubling 2 single ended designs . And remember it is a matter of effects how the balanced amp acts with the speaker.... and it is typically another story with respect to a real single ended amp .
Once I did build 2 pairs of Zen V2 ( with the Aleph current source ) for having a balanced design , in that case there wasn't connection between the 2 halves of the amp , but the sound was like paradise , also for the effect of the 2 Aleph arrangement -around- the speaker .
Again , the big circuital differences are when there is the exploitation of the balanced ciruit .It happens when a connection takes place between the halves ( differential pair ).

s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #45 on: July 19, 2012, 01:41:25 pm »

Maybe not stupid when you think it is a false alarm..

Yes I was asking whether it might just be a false alarm - the 'noise sniffer' might not be designed for a balanced setup.

Is it allowed to know what this was ? I mean, maybe I am going to read that article, but reading my own posts ?

Yes, referring to 'Peter's long post' is a bit like referring to Berlusconi's gaffe. I meant this:

Hook up your DC Voltmeter to the two ground references. You should see something. Now, start switching on house devices like phone adapters or motors or possibly amplifiers and the like. Watch the DC Offset change with it.

Now envision that this DC Offset mangles with at least your house ring ground reference. So, mid point e.g. goes up, and think about the part of the (hot point) wave above zero (which should be lower now in amplitude ) and the one under it (should be higher). Here you are turning that sine into something else.

In the mean time your AC output changes too. I never looked at it on the analyser, but while net the result should be the same, it won't because of the wave shape and let's say (for easily envisioning it) clipping at one side.
This all is about your consuming devices fireing back to the mains because of they changing the ground reference.
It may not be an official thing, but this is how I once could look at it. So, that consuming device is not a "balanced device" like in "it draws from both the phases separately" (see earlier posts about this - the 115v/115v thing) but as how I look at it, it will draw the ground towards "itself". So, where ground is to be 0V (which is a virtual thing always, but works for the physics in order) and a device is consuming the 230V, the ground goes up by, say 5V. I know, for AC this can't work, but I think something like switching power supplies ay use the positive side of the wave only or more than the negative, which depends on how the AC/DC conversion is done.


Attached are are a couple of snippets from the article.

Mani.


* Non-balanced AC.JPG (116.59 KB, 711x628 - viewed 580 times.)

* Balanced AC.JPG (155.94 KB, 714x754 - viewed 665 times.)
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #46 on: July 19, 2012, 01:49:29 pm »

Wow Mani. I only read the first part (will read on later) to see that I made up something which more knowledgable people already knew. But as I told earlier, for me this is just empirical finding, trying to make up the story of why around that.

Now I will read further. Thanks for posting this !
Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #47 on: July 19, 2012, 02:43:46 pm »

So in real life , If I use in the practice a transformer ( like in fig.6) from the mains to split the 220V in to two phases (115-0-115) , I can connect the PC to it for example ( included the new ground )  and all works pretty fine.
The only thing I see is a Voltage between the Transformer ZERO and the house earth . This voltage is about 70Volts , which goes to about 10V with a current flow in the order of uA .

If we want to be safe , and sure to protect our audio inputs againts that , it is better to let flow that little current -OR - connect all the HI-FI gear to the transformer .

Just my thoughts .
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #48 on: July 19, 2012, 03:18:45 pm »

The only thing I see is a Voltage between the Transformer ZERO and the house earth . This voltage is about 70Volts , which goes to about 10V with a current flow in the order of uA .

In my case, the DC voltage between teh transformer zero (bonded to my dedicated earth) and the house earth is only 0.25V. The AC voltage is 0.4V. I can live with 100-200 nA.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #49 on: July 19, 2012, 03:22:43 pm »

The only thing I see is a Voltage between the Transformer ZERO and the house earth . This voltage is about 70Volts , which goes to about 10V with a current flow in the order of uA .

In my case, the DC voltage between teh transformer zero (bonded to my dedicated earth) and the house earth is only 0.25V. The AC voltage is 0.4V. I can live with 100-200 nA.

Mani.

Thanks for pointing out that . I was talking of AC voltage , but as I say it carries very little current and it can be reduced very easily .
Of course my example is a diy experiment without a dedicate earth .... ok

s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #50 on: July 20, 2012, 12:00:30 pm »

I spent a long evening yesterday comparing various AC configurations (house AC, PurePower2000, 5KVA balanced in all permutations). I'll share my views when I have more time. But I just want to quickly share something with Nick.

I went ahead and ordered a 3kva transformer from Airlink today.

The Airlink seems to work best with an AC cable with a shield. I use Belden 83803 - a pig to work with but easily available on eBay and reasonably cheap. Connect the shield to earth at both ends.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #51 on: July 22, 2012, 12:13:34 pm »

I spent a long evening yesterday comparing various AC configurations (house AC, PurePower2000, 5KVA balanced in all permutations). I'll share my views when I have more time. But I just want to quickly share something with Nick.

I went ahead and ordered a 3kva transformer from Airlink today.

The Airlink seems to work best with an AC cable with a shield. I use Belden 83803 - a pig to work with but easily available on eBay and reasonably cheap. Connect the shield to earth at both ends.

Mani.

Mani,

i'm looking forwards to hearing the outcomes of your tests.
Thanks for mentioning of the Belden cable. I was concidering a rewire of the whole system on the Mains side and had spotted this cable. With your posertive experiance I have the confidence now to just go ahead and do it.

Also thanks for helping with my questions about the AirLink transformers, I am really happy with what it has done for my system Happy .

Cheers,

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #52 on: July 22, 2012, 01:21:42 pm »

This topic has been really helpful to stuff that I have had to look into recently and I wanted to post some impressions on the balanced mains power transformer that I have recently installed.

A recent visit by Paul's (Scroobius) to my place has helped to highlight a [big] "problem" with my system which has been holding back sound quality for a very long time. Then a return visit to Paul's place absolutely confirmed the problem and rather excitingly for me, totally reset what I consider to be my reference sound (wow Paul's system is good). During our chats Paul recommended trying balanced mains power and this became the get well step.

As a side note, it is turning out the at the "problem" is a compound one caused by a number of factors (so no "silver bullet" single fix  unhappy ). Everything in the system is coming under scrutiny.  Steps so far that have really helped are installation of balanced mains power, and the discovery that using my laptop to RIP CDs has lead to very substandard RIPed WAVs. Now that I have the Balanced mains it is easily allowing me to hear the difference between Laptop and PC RIPs easily so another part of the "problem" has been solved.

I guess that the impact from installing balanced mains power is going to vary depending on how poor the starting mains supply. I must have had poor mains quality because the change is really large here. In its fact difficult to fathom just how much better the sound has become !

Prior to installing I wanted to get the safety aspect of going balanced addressed. The AirLink Model that I went for does much of this by virtue of having primary circuit breakers and most importantly for me the RCD on the output phases installed as standard. In addition I will be fitting secondary fuses internally on all equipment that is connected to the Balanced power to ensure that what was formally the neutral wire has some fault protection within each component. I am no electrician so these measures my only be a starting point but I feel safer using balanced mains with them then without.

This is the unit which is a smaller 3kva version of the 5kva models that that Mani uses, the 3kva model has three pin main plugs installed making output wiring a little simpler than the bigger unit. Build quality is "industrial" but given the price very good, the unit makes virtually no humming sound you have to have your ear right to the case to hear anything at all and then only occasionally can you catch it making any sound. Delivery took just two days.

http://www.airlinktransformers.com/standard_balanced_power_supplies/61-BPS3110S.html

So the impact of the system on sound quality. Well sound is has improved EVERYWHERE, perhaps most importantly a nasty layer of hash and confusion in the sound has been removed throughout the performance right from lows through to highs. This has creatated a much firmer and more understandable foundation for the music to be built on with a much lower noise floor. Transients are better defined and more accurate in their weight. For the first time, in a very long time, highs could almost be described as sweet, with understandable rhythms easily coming across in percussion lines (this was also improved by the discovery that my laptop RIPS are flawed). Mid range is solid less confused has better prominence and improved air and focus between the speakers. Voices, guitars, strings, horns and drums are better reproduced with more colour weight snap and delicacy where appropriate.

My comments are in the context of the balanced power output earth being connected to the household mains PE. I have not setup a dedicated low resistance earth (using earth rods etc) but this may come later.

I was not really that much of a believer in mains conditioning before this and perhaps still don't really subscribe to the conditioning approach of many of the single phase solutions, but the Balanced approach is such a simple one, providing isolation, noise cancellation, and possibly more stable ground reference. I am hooked, this rather industrial looking box has found a place at the heart of my music system.

There are still further "problems" to address for me but I think the mains component is now done, and one strep closer the the sound I fell in love with a few weeks ago whilst at Paul's.

Nick.

Ps these impressions are with about 20 hours on the transformer they are supposed to improve quite a lot with some burn in.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #53 on: July 22, 2012, 02:03:20 pm »

hi Nick, could you please poste a draw or schematic of your new balanced connection in power supply   so that we can have a reference? If possible.   stefano
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #54 on: July 22, 2012, 03:01:33 pm »

I've got a few minutes before my flight, so thought I'd quickly chime in.

So the impact of the system on sound quality. Well sound is has improved EVERYWHERE, perhaps most importantly a nasty layer of hash and confusion in the sound has been removed throughout the performance right from lows through to highs. This has creatated a much firmer and more understandable foundation for the music to be built on with a much lower noise floor.

Yes, this is pretty much exactly how I would describe going to balanced AC in my system. As a result, in my system, the depth of the sound stage is really improved - certain sounds now come from way back behind the speakers. There is nothing at all in your face.

However...

The PurePower2000 provides an even more drastic improvement. It's hard to describe, but with it I'm no longer getting just a nice clean sound, I'm getting music. An acoustic guitar no longer sounds like it's made out of cardboard. It now resonates and 'sings', just like my own Martin (when someone who can actually play a guitar plays it, i.e. not me). I really, really don't want this to sound like an advert for PurePower - they are bloody expensive and may not work out well for everyone. Indeed, mine has been gathering dust in my cellar for the last 12 months because I decided I preferred the sound without it. But then I tried plugging everything into it, including the PC, and voila.

So I now how the following AC setup:
100A dedicated line -> PurePower2000 -> 5KVA balanced transformer (with dedicated earth on output) -> shielded/screened wiring

The biggest bang-for-the-buck is definitely the 5KVA transformer. But having heard what the PP2000 does, I can no longer with without it. Indeed, I've just ordered a new PurePower+3000. (Of course this is just crazy - quite frankly I could have paid my bloody mortgage off with all the money I've spent on this hobby.)

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #55 on: July 22, 2012, 04:16:57 pm »

Quote
The biggest bang-for-the-buck is definitely the 5KVA transformer. But having heard what the PP2000 does...

Dam, why do I read these things ? now I want a PurePower but I just looked at the price ! Happy

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
Scroobius
Audio Addict
***
Offline Offline

Posts: 1139


View Profile
« Reply #56 on: July 22, 2012, 09:48:40 pm »

Wow I was trying to justify the cost of a balanced transformer I had not even started to think about a Purepower!!!!

P

Hey Nick - glad to hear your balanced Tx is sounding good sounds like a good enough reason to come up for another listen!! (Oh yes and another run out in the Ultima ha ha)

Also do you still have your "balanced earth" connected to house earth? as balanced operation cancels reactive currents appearing on the secondary your "balanced earth" should be nice and clean - it seems a shame to connect it to a dirty old house earth. Well maybe who knows how dirty the house earth is? - but anyway I reckon it would be worth lashing up an earth in the garden even on a temporary basis to see how it performs.

I suppose a concern is always the possibility of a significant difference in voltage appearing between the dedicated earth and the house earth - at least Mani's is very small but who knows. It would be interesting though.

Hey Mani - have you tried house earth versus dedicated earth in respect of SQ?




Logged

W10-14393.0 - RAM Disc XXHighEnd Mach II Stealth LPS PC -> Xeon E5 2040v4 with Hyperthreading On (20 cores)* @~720MHz, 32GB, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1/ Q1Factor = *1* / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *3.0*  (max 120) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = 43 / Nervous Rate = 100 / Cool when Idle = n.a / Provide Stable Power = 0/ Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter low(12220) 705600 / -> USB3 from MoBo -> Lush USB 70cm -> Phisolator -> 24/768 Phasure NOS1a G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink -> Orelino MkII Active Open Baffle Horn Speakers.
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #57 on: July 22, 2012, 10:28:18 pm »

Wow I was trying to justify the cost of a balanced transformer I had not even started to think about a Purepower!!!!

P

Hey Nick - glad to hear your balanced Tx is sounding good sounds like a good enough reason to come up for another listen!! (Oh yes and another run out in the Ultima ha ha)



Paul hi,

I was going to pm you, yes it would be great to try balanced power in your system. My unit or Mani's (or both ha ha). Could the best get better ? Let me know the dates you are thinking of.

Speak soon,

Cheers,

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
Scroobius
Audio Addict
***
Offline Offline

Posts: 1139


View Profile
« Reply #58 on: July 23, 2012, 02:06:56 pm »

Interesting extract from the Equitech site and of particular interest to Nick who has a 3-phase supply to his house:

"The level of interference created when a three-phase wye system is split up and used as three single-phase circuits is truly something to behold. For example, as much as 20% (or more) of the power used by fluorescent ballasts is reflected back onto the power grid in the form of reactive or harmonic currents -- now that’s a lot of distortion. In the late 80’s, a 40-plus-story office building in Los Angeles actually burst into flames because of these reactive currents. Incredibly, the origin of the fire was determined to be from excessive harmonic distortion in fluorescent lighting circuits which created a high-frequency current overload and literally a meltdown of the electrical wiring system. The First Interstate Bank fire in Los Angeles in May of 1988 was the event dubbed by the media as "The Towering Inferno" a la the Hollywood movie. Codes were adapted to remedy the fire danger, but the noise problem itself was never completely resolved."

more here  http://www.equitech.com/articles/enigma.html

P
Logged

W10-14393.0 - RAM Disc XXHighEnd Mach II Stealth LPS PC -> Xeon E5 2040v4 with Hyperthreading On (20 cores)* @~720MHz, 32GB, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1/ Q1Factor = *1* / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *3.0*  (max 120) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = 43 / Nervous Rate = 100 / Cool when Idle = n.a / Provide Stable Power = 0/ Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter low(12220) 705600 / -> USB3 from MoBo -> Lush USB 70cm -> Phisolator -> 24/768 Phasure NOS1a G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink -> Orelino MkII Active Open Baffle Horn Speakers.
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #59 on: July 23, 2012, 11:00:04 pm »

Interesting extract from the Equitech site and of particular interest to Nick who has a 3-phase supply to his house:

"The level of interference created when a three-phase wye system is split up and used as three single-phase circuits is truly something to behold. For example, as much as 20% (or more) of the power used by fluorescent ballasts is reflected back onto the power grid in the form of reactive or harmonic currents -- now that’s a lot of distortion. In the late 80’s, a 40-plus-story office building in Los Angeles actually burst into flames because of these reactive currents. Incredibly, the origin of the fire was determined to be from excessive harmonic distortion in fluorescent lighting circuits which created a high-frequency current overload and literally a meltdown of the electrical wiring system. The First Interstate Bank fire in Los Angeles in May of 1988 was the event dubbed by the media as "The Towering Inferno" a la the Hollywood movie. Codes were adapted to remedy the fire danger, but the noise problem itself was never completely resolved."

more here  http://www.equitech.com/articles/enigma.html

P

OMG !

 I'm going to keep the fire extinguisher near by ! I'm starting to think three phase is for industrial plant not hifi ha ha.

What a great article on the equitech site, fascinating and really makes sense when you think about it. I think setting up a dedicated earth is going to be next. I found a bespoke earth in the flower bed next to the front door. The chap that owned the house was into hifi so I'm wondering if he put it in. I think I will have to get the garden spade out this week end to see what's happening. I never thought doing the garden would have the possibility of improving sound quality ha ha.

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #60 on: July 24, 2012, 09:27:35 am »

Let me try to contribute somewhat to this topic, although for one person it is already known (Nick). Hopefully it will come across as "shocking", but I'm just trying to make clear how important this whole stuff is.

From other topics it is maybe known that I myself experienced a few issues here and there, and since they are mains related they are out of (my, your) control. Unless of course we are going to setup our own mains, which seems to be happening now for some. Great, and also great to see back what this is all about. So :

The DC Offset "control" in the NOS1 is what I call "mechanically" organized. No circuitry for it but ways to guide off to central ground or PE (earth) if you want. Now, ever back there was this first version and it worked okay for everybody. But one of course, and this was Nick. Notice that IIRC 5 means of providing mains in the UK exist, and what Nick receives was sort of described a few posts back (but it can be desrcibed more precisely with some Wiki pages about).
Btw, at having Nick's NOS1 back, here it was just okay again.

Next we had the first NOS1 out to the US and our testing at 110-130V overhere showed that the DC Offset control did not work for that. So, we rearranged the lot (a full month of working and trying) and Nick's NOS1 received that too. Everywhere in the US it worked, but not for Nick of course (with his special mains).

In parallel, Mani got in his transformers. All super duper. But from that moment on he had a most nice DC Offset on both channels exactly the same (and for those to whom it may tell something, this was 120mV (while it should be 5-10mV). Just because of his transformer ...

We again changed the setup, and now for everybody it definitely worked. Nick though, still has an issue nobody has.
Always Nick.

This is in a nutshell which is a sort of summarize from what ? 200+ emails ? Something like that. The relation to the mains was "found out" well over a year ago and I am still not sure how to reason it out. So, the fact that knowledgable people in the field are able to show that things matter and is "delicate" for audio is one thing, but the coincidental occurrence of having a DAC which fails on it visibly (and audibly) is something else. Of course, with the actually necessary circuitry in there none of this would have shown. But you can 100% bet that then it would have been audible again because this is a "servo fighting" thing.

As an addition, I can tell that really nasty things can show from the electrical relation to different potentials; as I saw it happening with the DAC chassis, that acting as a capacitor, continuesly charging and uncharging - through air (in say a 5 second frequency). Only when you know how to measure it, it will show. Audible ? why not; everyting is (btw no worries, this was a test setup).

While all is about the phases and how they relate or form (!) ground, this is really something to think about (well, we do in this topic). And still, once that ground is okay in the base, what happens when you connect that device; it will be off somewhat again.

DC Offset has become my life and I got obsessed by it. Electical Engineers tell me that the ground created by two transformers "obviously" is equal to the DC ground further down the line, the DC derived from those same transformers. It is not so. It is not so, because -in my view- the phases hammering on the transformers are not equal in amplitude. DC Offset again. This, while what's derived in DC is formed by a nice calibrated plus and minus DC voltage. Something like that. Now think what happens when PE (which should be the center of the mains phases) is connected to the chassis and "star ground" (etc.) is on the chassis as well. This, while PE fluctuates all over. Can't work eh ? well, open your devices and see how it is arranged for. And know that interlinks connect. So, when one device is setup properly (could be the NOS1) the other destroys it right away.

We are not there with a clean "earth". We must also take care that our devices don't destroy it right away.

Peter
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #61 on: July 25, 2012, 01:34:41 am »

I did some gardening today  Wink I planted three copper grounding rods to wire up a separate ground for the output of my balanced power transformer.

For a while I thought I hit the water supply to the house when driving in one of the rods but ( relief ) the low pressure fault turned out to be in the supply to the street, thank goodness ! I did learn however that my wife values water to the house a lot more than good sounding music, still its always good to know where your partners preferences sit on important matters like this. dntknw

Anyway I managed a couple of hours listening with the ground wired into the system later in the evening. What a surprise ! There is an improvement as large as setting up the balanced supply a few days ago. Actually it may even be a bigger step than that. There is much more of the benifits I described in the post when the 2 phase mains was connected up but with the house PE. Besides more of the same improvements the stand out things on top of those are the extended, controlled bass, smoothness, detail, and realistic tone and body to the music.

I wasn't ready for how big the difference is wit the new isolated ground.. Two big steps in a week, this balanced supply stuff is fun!

It has not solved my other problems with the system but I am hoping that this good gound arraignment is going to help sort it out....

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
Scroobius
Audio Addict
***
Offline Offline

Posts: 1139


View Profile
« Reply #62 on: July 25, 2012, 08:47:50 am »

Hey Nick - that's brilliant news!!!!  but your other problem remains - oh boy is that seriously weird? so we know that your equipment is OK (it worked here without problems) and your mains and earth is now "isolated" (well OK it never can be) from your house mains and earth.

But it has got me thinking - what would balanced supply and dedicated earth sound like here? I would have to think that it could make an improvement.

Gotta say that this hifi as a DIY hobby is fascinating. Much better than going to the dealer and buying the latest greatest product.

Anyway I have another idea for fixing your problem - try these people :-

http://humahealing.blogspot.co.uk/2008/03/removing-ghosts-poltergeist-demons.html

P

Logged

W10-14393.0 - RAM Disc XXHighEnd Mach II Stealth LPS PC -> Xeon E5 2040v4 with Hyperthreading On (20 cores)* @~720MHz, 32GB, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1/ Q1Factor = *1* / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *3.0*  (max 120) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = 43 / Nervous Rate = 100 / Cool when Idle = n.a / Provide Stable Power = 0/ Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter low(12220) 705600 / -> USB3 from MoBo -> Lush USB 70cm -> Phisolator -> 24/768 Phasure NOS1a G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink -> Orelino MkII Active Open Baffle Horn Speakers.
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #63 on: July 25, 2012, 09:21:35 am »

That's what I've been telling Nick all the time.
Point is also : the issue that Nick has got left just can't exist. Not with the knowledge I have. But it does exist ...
Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #64 on: July 25, 2012, 11:45:34 am »

Anyway I have another idea for fixing your problem - try these people :-

http://humahealing.blogspot.co.uk/2008/03/removing-ghosts-poltergeist-demons.html

Paul,
Very good suggestion ha ha. I was thinking I might be able to lead the little ghost blighters away from the system through a star earth system into ground posts in the garden  Happy

Quote
That's what I've been telling Nick all the time.
Point is also : the issue that Nick has got left just can't exist. Not with the knowledge I have. But it does exist ...

Peter
I guess that to good thing is that now i'm getting some good information to exclude possible causes (the Adnaco isolating the PC and the new power set up together with the tests that Paul and performed). This is helping to narrow down the number of areas that could be causing the problem.

Another positive thing is that without this ghost hunt I would probably not have found solutions to other problems like the ripping and balanced power so not all bad !  Happy

I also just wanted to say on forum that I'm sure that no other manufacture would have given me the help and support that you have to try to solve this (absolutely no way). We are quite sure now that it is not the NOS1, but it could be some combination of factors and interactions particular to my system. Whatever the cause I cannot overstate the amount of help you have given me. Many many thanks for this.

Best,

Nick.

EDIT: correction to language
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
PeterSt
Administrator
High Grade Audiophile
*****
Online Online

Posts: 15259



View Profile Email
« Reply #65 on: July 25, 2012, 11:56:44 am »

Thanks for the kind words Nick. If only "that" problem was solved now.

Supposed you'd bring another visit to Paul, can't you take your whole system to him and try there ? That would definitely sort out causes. Of course Paul has to agree with this (a lot of mess going on).
But I already see your horns in front of me - too large to hop in a smal car. So exclude those ? that could be a dangerous thing (when the issue doesn't exhibit at Paul's you'd remain with thinking that it can be the speakers).

Btw Nick, I certainly won't say that the NOS1 can be excluded of being a cause. Of course, your mutual excercises logically tell so, but since this is about ghosting D/A conversion (for others, this is no joke) the first device prone to it would be that D/A converter. So, it can still be an interaction somehow and as expcted it must be mains related (because that's really all what's left I think). You can only prove that by bringing the whole lot to someone like Paul where the phenomenon doesn't occur. If that is proven you'd know what *not* to look for.
Btw, mains related seems obvious, but I would also take into account some antenna working somewhere. If from your audio chain it should show at (e.g.) Paul's, and if from something else (like your iPad charger ha ha) it won't show there, but in your house it is to be considered as a cause outside the mains.

Regards,
Peter

Logged

For the Stealth III LPS PC :
W10-14393.0 - August 18, 2018 (2.10)
XXHighEnd Mach *III* Stealth LPS PC -> *Xeon Scalable 14/28 core* with Hyperthreading On (*set to 10/20 cores*) @~660MHz, 48GB, Windows 10 Pro 64 bit build 14393.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 30/-/1/1/1/ Q1Factor = 10 / Dev.Buffer = 4096 / ClockRes = 15ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *140.19*  (max 140.19) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / Stop Desktop, Remaining, WASAPI and W10 services / Use Remote Desktop / Keep LAN - Not Persist / WallPaper On / OSD Off (!) / Running Time Off / Minimize OS / XTweaks : Balanced Load = *35* / Nervous Rate = 10 / Cool when Idle = n.a / Provide Stable Power = 0 / Utilize Cores always = 1 / Time Performance Index = Optimal / Time Stability = Stable / *Arc Prediction Filtering (16x)* / Always Clear Proxy before Playback = On -> USB3 from MoBo -> *Lush^2 A: B-W & Y-R, B: B-W* USB 1m00 -> Phisolator 24/768 Phasure NOS1a/G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (16ms) -> B'ASS Current Amplifier -> Blaxius Interlink -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere (also from the PC).

For a general PC :
W10-10586.0 - May 2016 (2.05+)
*XXHighEnd PC -> I7 3930k with Hyperthreading On (12 cores)* @~500MHz, 16GB, Windows 10 Pro 64 bit build 10586.0 from RAM, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1 / Q1Factor = 1 / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = 0.10  (max 60) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = *43* / Nervous Rate = 1 / Cool when Idle = 1 / Provide Stable Power = 1 / Utilize Cores always = 1 / Time Performance Index = *Optimal* / Time Stability = *Stable* / Custom Filter *Low* 705600 / -> USB3 *from MoBo* -> Clairixa USB 15cm -> Intona Isolator -> Clairixa USB 1m80 -> 24/768 Phasure NOS1a 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink *-> B'ASS Current Amplifier /w Level4 -> Blaxius Interlink* -> Orelo MKII Active Open Baffle Horn Speakers.
Removed Switching Supplies from everywhere.

Global Moderator
Scroobius
Audio Addict
***
Offline Offline

Posts: 1139


View Profile
« Reply #66 on: July 25, 2012, 01:12:08 pm »

Interesting ..... antennae type problems did occur to me and I was trying to think of ways to test that mmmm.....

Out of interest Nick is your NOS1 / amp electronic ground connected to PE - I mean what is the resistance or voltage ground to PE - I guess it should be close to zero if there is a connection via USB but it would be interesting to know.

Anyway if you can't "take the mountain to Mohammed" how about I bring my complete system up to your place Nick???? 

Paul

Logged

W10-14393.0 - RAM Disc XXHighEnd Mach II Stealth LPS PC -> Xeon E5 2040v4 with Hyperthreading On (20 cores)* @~720MHz, 32GB, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1/ Q1Factor = *1* / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *3.0*  (max 120) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = 43 / Nervous Rate = 100 / Cool when Idle = n.a / Provide Stable Power = 0/ Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter low(12220) 705600 / -> USB3 from MoBo -> Lush USB 70cm -> Phisolator -> 24/768 Phasure NOS1a G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink -> Orelino MkII Active Open Baffle Horn Speakers.
manisandher
Crazy Audiophile
****
Offline Offline

Posts: 2043

from-first-principles.com


View Profile WWW Email
« Reply #67 on: July 25, 2012, 01:25:53 pm »

Nick, I'm back in the UK in a few days' time and would be happy to help out in any way I can.

Mani.
Logged

Audio PC: Phasure Mach II *with clock upgrade* / Win 14393.0 on RAM-OS / controlled by RDC, no GPU card installed / connected directly to music server
XXHighEnd: 2.10 / Minimize OS / Engine#4 Adaptive / DB=4096 / Q1=30 / xQ1=40 / Q3,4,5=1 / SFS=5.19 / XTweaks = 52, 10, 0, 0, 1 / Straight Contiguous / Clock Resolution = 15ms / Scheme 3-5 (low/realtime) / 8x Arc Prediction (switch #5 'up/off') / Unattended
Main System: mobo USB3 port -> Lush^2 -> Phasure NOS1a B75 G3 -> First Watt F5-cloned mono amps -> Tune Audio Anima horn speakers
Office System: Lush -> Phasure NOS1a G3 -> BD-Design BD30-SPR mono amps -> Tune Audio Marvel horn speakers
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #68 on: July 25, 2012, 05:10:19 pm »

Nick, I'm back in the UK in a few days' time and would be happy to help out in any way I can.

Mani.

Mani hi,

I really appreciate the offer of help. I am just planning the next steps. Reading around this it seems that wisdom is that the more you connect to a good quality clean ground in a balanced power setup the lower the hum / noise drops. I am thinking that, as Paul mentioned, it might be worth trying some carefully chosen grounding of PC / DAC and AMPs. Even the signal ground might be worth trying to see if I can earth the stay signal noise before it reaches whatever is performing the unwanted DAC conversion.

I am also focused on the differences between my gianclone build and Pauls (there are differences which might make mine more suseptable to noise). Perhaps the easiest way to check this would be to try Pauls amps. One of the most useful test so far has been changing out components with Paul. I think if we go down this route it may need to be at my house to make sure that all possible factors (including antenna effects, speakers wiring ghosts  Wink etc)  are taken into account.

I will drop you and Paul a PM very soon.

Kind regards,

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #69 on: July 25, 2012, 06:52:14 pm »

Hello All ,

I did furter experiments with the balanced transformer ;

And my question is : do we really need the Earth connection ( safety reasons aside ) when using a balanced transformer?

I ask that because the canncellation by itself its so good in balanced mode , that I don see any of the problems coming throug the skin or speakers , when I connect all the gear throug the transformer and *without* the Earth connected . Problems that are easy to test in unbalanced mode and without an Earth ground connection .

stefano .
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #70 on: July 25, 2012, 08:12:06 pm »

this is my experiment today (diy).
I believe toroidals have better bandwith anyway, but I may be wrong on this .

I did connect PC , Monitor and Amp to it .


* isolated mains.jpg (48.47 KB, 769x400 - viewed 485 times.)
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
Nick
Audio Addict
***
Offline Offline

Posts: 725


View Profile
« Reply #71 on: July 25, 2012, 11:27:35 pm »

Stefano,

I just wanted to say that I did see your earlier question about the circuit I have set up. I have drawn it out and will post it tommorow.

Nick.
Logged

Audio PCs
PC No. 1: 621 motherboard Xeon 32 thread CPU.
PC No 2: X99 motherboard Xeon 28 thread CPU - modified motherboard.

 w10 10393 RAM OS => XX V2.09 / adaptive mode / XX buffer 4096 / NOS USB driver v 1.02 buffer 16ms / Q1,2,3,4,5 = 30,-,1,1,0 / xQ1 =40 / unattended / SFS 5.19Mb / memory straight continuous / system clock 15.0ms / Threadprio RealTime / Playerprio Low / CPU scheme 3-5 / 16x Arc Prediction / Peak Extend off / Phase alignment off / Phase off  / XTweaks : Balanced Load 43 / Nervous Rate 50 / Cool when Idle n/a / Provide Stable Power 0 / Utilize Cores always 1 / Time Performance Index = Optimal / Time Stability  On => Lush USB cable => modified NOS1 USB DAC => no pre amp => Orelo active horn loudspeakers with modified bass channel DSPs.

Music server: X99, Xeon 16 thread PC.

System powered from two 3kva balanced mains isolation transformers.
stefanobilliani
Audio Enthusiast
**
Offline Offline

Posts: 207


View Profile
« Reply #72 on: July 25, 2012, 11:41:32 pm »

Ok , great Nick.No worries. Happy     s
Logged

May 27 -2013 ,  0.9z-9

W8 pro 64bit, UEFI
sata ahci mode .
i7 (8 cores ) 2600k
RAM : 16Gig
XXHE on OS disk
Music on OS disk
Playback drive on USB3 to 4Gig Pendrive
Mobo USB 3 to Oversampling dac at 2X
KS adaptive , ArcPredict  , dev buffer 1024
Q1=8 , Q5=3 , res 1msec
Minimized OS , stop all services
X-tweaks : 45,100,1,1,1, stable , optimal
Disk Utilization : Split File Size: 120 ,max the same as SFS
Memory Organization : Straight Contiguous
Scroobius
Audio Addict
***
Offline Offline

Posts: 1139


View Profile
« Reply #73 on: August 12, 2012, 05:33:17 pm »

I could never quite bring myself to spend the money on a balanced power supply transformer as my system has been sounding good for some time. Anyway I finally took the plunge and bought a 3kv balanced transformer from Airlink. Yesterday I connected it and drove a spike into the garden for an independent earth for the hifi system.

And what do I think? - wow! this is a really fundamental improvement. How would I describe it? I think the best way is just how clear the sound is. The tone of instruments is better there is more detail. Sound staging is better. Just after I had installed it I was wondering around the house clearing up and was playing some piano and it was just so liquid sounding very natural very real.

I find myself playing at a higher volume now maybe because it is a cleaner sound.

Anyway a very fundamental improvement that keeps me in my chair not wanting the music to stop.

Thinking about it? go for it.

Paul

Logged

W10-14393.0 - RAM Disc XXHighEnd Mach II Stealth LPS PC -> Xeon E5 2040v4 with Hyperthreading On (20 cores)* @~720MHz, 32GB, music on LAN / Engine#4 Adaptive Mode / Q1/-/3/4/5 = 14/-/1/1/1/ Q1Factor = *1* / Dev.Buffer = 4096 / ClockRes = 1ms / Memory = Straight Contiguous / Include Garbage Collect / SFS = *3.0*  (max 120) / not Invert / Phase Alignment Off / Playerprio = Low / ThreadPrio = Realtime / Scheme = Core 3-5 / Not Switch Processors during Playback = Off/ Playback Drive none (see OS from RAM) / UnAttended (Just Start) / Always Copy to XX Drive (see OS from RAM) / All Services Off / Keep LAN - Not Persist / WallPaper On / OSD On / Running Time Off / Minimize OS / XTweaks : Balanced Load = 43 / Nervous Rate = 100 / Cool when Idle = n.a / Provide Stable Power = 0/ Utilize Cores always = 1 / Time Performance Index = Not the best (OS from RAM issue) / Time Stability = Off (OS from RAM issue) / Custom Filter low(12220) 705600 / -> USB3 from MoBo -> Lush USB 70cm -> Phisolator -> 24/768 Phasure NOS1a G3 75B (BNC Out) async USB DAC, Driver v1.0.4b (4ms) -> Blaxius BNC interlink -> Orelino MkII Active Open Baffle Horn Speakers.
Pages: 1 2 3 4 5 [All]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1 RC2 | SMF © 2001-2005, Lewis Media Valid XHTML 1.0! Valid CSS!
Page created in 0.199 seconds with 19 queries.