AC Filtering, Grounding Boxes, Linear PSU and Balanced Power.

[zilch0md]

Oh, you started out with it unbalanced?   That's interesting that it is running more quietly now.  From what I've been reading, any hum that can be heard with unbalanced ITs is due to the presence of DC offset coming in from the mains, but I guess your transformer somehow doesn't "see" the DC offset, after center-tapping the secondary for balanced output.  ???

[zilch0md]

Quoting the paragraph below Figure 4. Reducing Common-mode noise, as seen at this page:

 

http://www.mastec.co.nz/JST/PK8.html

 

Quote

With all transformers, stray capacitance, called inter-winding capacitance, can exist between the windings.  It is via this inter-winding capacitance that [incoming] high frequency [common-mode] noise [from the mains] can couple through to the secondary winding. However, when a grounded [Faraday] shield is used to separate the primary and the secondary windings, the inter-winding capacitance is significantly reduced. This effectively increases the impedance (resistance) of the coupling path and therefore reduces the amount of high frequency [common-mode] noise that can couple through to the secondary winding.

 

Thus, the lower the inter-winding capacitance, the greater the reduction of high frequency common-mode noise.

 

Which is what John Swenson has been telling us.  Not just any isolation transformer will do the job.   :-)

[zilch0md]

For the most passionate of students, here's a very well-written article, which, among other things, provides very succinct definitions of the different types of noise. 

 

The author, Marc Dekenah, all but carves 11th and 12th Commandments in stone. 

 

Paraphrasing... Thou shall not use the term "Normal-Mode Noise," but rather "Differential Noise" and "Transverse Noise" is a superset of "Differential Noise;" they are not one-in-the-same.

 

Let the reader be schooled. LOL

 

http://www.marcspages.co.uk/pq/3270.htm

 

Here's an index to several other topics - all, equally well-written and in-depth:

 

http://www.marcspages.co.uk/pq/0000.htm

 

[zilch0md]

I've been doing a lot of research into how the Topaz (MGE or Daitron) Ultra-Isolator transformers could possibly manage to provide -65 dB of Normal-Mode noise reduction in addition to -146 dB of Common-Mode noise reduction (as go the specs of their "-31" and "-32" models, which have the extremely low 0.0005 pF inter-winding capacitance, that allows attenuation of noise in the treble frequencies.

 

Way back on page 13 of this thread, I had posted a graphic of my own creation which, I've only recently concluded, promotes the erroneous contention that a 1:1 transformer having a grounded-neutral secondary (such as the Topaz or less-effective Tripp-Lite units) will attenuate Common-Mode noise (TRUE), while 1:1 transformers having a floating-neutral secondary will attenuate Normal-Mode noise (FALSE).

 

I've held that contention for over two years, but have finally done enough homework to realize that it's absolutely incorrect - and - I've been misleading readers with that incorrect understanding, all along, just as I had been mislead by a consensus of quacks who had made a lot of sense to me at the time.  It's really not debatable.  

 

Back in November of 2016, in the thread titled "Discussion of AC mains isolation transformers..." John Swenson stepped up, without naming names, to set the story straight.  As much out of self-doubt as for my respect for John's expertise, I didn't openly contest his post, but his words did manage to plant a seed in my thick skull, to investigate this further.  

 

It wasn't until I bought a Topaz 19095-32, recently, that I finally started digging into what impact, if any, floating the secondary's neutral has Normal-Mode vs. Common-Mode noise reduction.  I am now thoroughly convinced that John Swenson is correct. Doh!  Both types of isolation transformer, all else being equal, will attenuate Common-Mode noise, not Normal-Mode noise.

 

There is a slight twist, not yet discussed in these pages:  It turns out that grounding the secondary's neutral might improve the Common-Mode noise reduction, ever so slightly, by providing one last path to ground for any remaining CM that has made it through the Faraday shield and is otherwise headed for the load.  A floating-neutral secondary would not provide this infinitesimally small advantage for CM noise reduction.

 

So, here's are a couple of revised graphics, that incorporate corrections to previously "published" versions (along with a lot of other information, which is hopefully not misleading or incorrect in any way.

 

:-)

 

Mike

 

 

 

 

[YashN]

Some very cool info and links there, Mike, thanks for sharing.

 

Additionally: always cool to read John Swenson's explanations on the less obvious or more intricate phenomena in computer-audiophile land.

[YashN]

Some simple experiments and captures regarding the direct Computer-to-DAC, with various power supplies including a DIY Linear Regulated PSU for the DAC, and a Custom USB Connector done during the Dec 2017 - Jan 2018 holidays (so last year), a ripe time for DIY as usual. Meant to post this before, but was really taken up this year and besides, the evil Photobucket suddenly blocked sharing - unless you pay them a hefty fee of course... This gives an idea of what we potentially lose by direct powering our DACs (not counting DACs which are specifically designed to have as complete an isolation from the computer or such other solutions as using an NAA) form the USB port of the server Computer.

 

- Captured using my old M-Audio FastTrack Pro at the DAC outputs (iFi iDSD Nano)

- Sine 1000Hz generated by Audacity

- Spectrogram in Audacity

 

Caveats: these aren't professional by any means, but give an inkling of what each PSU is affecting and how that correlates to subjective listening at least for me and in my system. Someone with better equipment/scopes can surely do much better. It's a fun thing to do nevertheless.

 

Initially, I only did a filter for the USB power and GND leads from the computer to the DAC. However, this wasn't satisfactory: I lost a lot of dynamics, so I had to relegate that filter to another duty, but I kept the circuit at hand.

 

1. Default iMac to DAC, using a generic USB Cable:

 

 

There's a lot of things going on here... Very bad things. There ideally should be a single line at 1kHz, the original signal played by the DAC in non oversampling mode.

 

So that's what you get when you power your (my current) DAC by computer USB...

 

I set out to try and isolate the DAC from the pesky +5V and GND coming from the noise infested computer, but needed both a USB connector allowing external injection of power as well as a new PSU. So, the best thing I wanted to try is my own Linear Regulated PSU and power the DAC from that.

 

But first I had to hack the USB cable to allow the external power. Easier said than done though: while powing externally is just having two leads from the new PSU to the DAC (easy), there's handshake and any required connection back the computer to take into account. I struggled a bit with this until I found a way to filter out the GND line noise (that filter above came in handy). In some cases, if I disconnected that line, my DAC wouldn't be recognised.

 

So, USB cable hacked to allow external USB-level power, and hence the next thing I tried and which I had at hand was an iPad charger:

 

2. DAC externally powered by iPad charger with a custom short USB cable:

 

 

This already sounded better, clearer if a bit on the 'thin' side! There's still a lot of horrible things going on if you judge by the chart. The iPad charger is a compact SMPS. The better audio probably has to do with not much happening compared to #1 regarding 20Hz - nearly 100Hz because the rest does look as bad if not worse (worse especially in the 10KHz to even 29KHz region which could potentially cause all sorts of non-linearities/bad resonance at the load).

 

However, 'better' here is purely relative to example 1 above, which is a really bad system. Because just think of it: I just replaced an SMPS (computer-side) by another smaller SMPS and just the fact that the DAC is externally powered now makes sound much better. How bad must the computer-side noise be, then?

 

Now, I wanted to try a battery of sorts to power the DAC when it's not powered by its internal battery (needed say, when the internal battery has discharged too low for normal operation). So I got myself a cheap Battery Charger Pack which conveniently has a USB style output port:

 

3. DAC externally powered by cheap Battery Charger Pack:

 

 

Just be aware here that the scale on the left has changed: everything else cleared out around the 1K Sine line, so I wanted to zoom in a bit around it to see if I saw anything. Much, much better sound here, very clean. But is that all there is?

 

DIY Linear PSU to the test. Spent some time refreshing on Electronics and circuits, fished around for John's old posts on his linear PSU Design and some more inspiration around Linear Regulated PSUs circuits and posts and built a first iteration.

 

4. DAC externally powered by DIY Linear Regulated PSU (some John Swenson design in it):

 

 

Wow! Nothing prepared me for this: THE THING I noticed most and immediately was the bass content, the slam, dynamics. This sounds huge. I built the PSU as a floating design.

 

Here, if you reflect a little on what we've seen so far, this is what is lost when I'm using either the computer-side SMPS or an external SMPS: I lose a lot of the energy that should be there to reproduce the bass properly.

 

I did several other experiments, including changing capacitors post Regulator, turning the computer-side filter on and off (not a big difference, but still noticeable), combinations of the data lines shielding which showed even further improvements, etc...

 

One of the things which was especially interesting to me was finding out I could also make the DIY LInear Regulated PSU better by taking particular care of the grounding and layout.

 

In fact, check this new improved layout no components changed, just more proper thinking and design of return paths:

 

 

Again, much cleaner throughout on the graph, more good sound.

 

I am not sure what the really low level bump on the left is or whether I can also get rid of it.

 

So last year during the holidays, I had only set out to filter the direct power and GND lines from the computer, but I inadvertently re-implemented one of the computer-side USB filter devices. I also had to re-think what a USB short connector should be and got better sound this way. I also got the best sound in all these experiments with the DIY Linear Regulated PSU, in this configuration:

 

1. USB Filter Computer-side ON

2. Custom configuration with the shield for the data lines

3. DIY LInear Regulated PSU when powering and charging the DAC

 

I also re-added the AC Filter into which I plugged the iMac itself and a few other devices like the TV and the Blu-Ray player. This added yet another level of improvement. While I didn't expect it to be that much better (since I block a lot of bad influence from the computer to the DAC with both the custom connector allowing external power and the special shield configuration), but it was there and very noticeable too.

 

I've been enjoying a wicked sound here over the year, accompanying all my work through 2017.

 

It will be interesting to test a few things more in that configuration, like the copper tube external 'signal grounding'  experiment, some of the Star Quad wiring that is being discussed, further enhancements to the circuitry for the DIY PSU etc...

[sandyk]

52 minutes ago, YashN said:

Some simple experiments and captures regarding the direct Computer-to-DAC, with various power supplies including a DIY Linear Regulated PSU for the DAC, and a Custom USB Connector done during the Dec 2017 - Jan 2018 holidays (so last year), a ripe time for DIY as usual.

 

Well done !!! 

I can relate very closely to much of what you have done here.

Unfortunately, the closed minded Objective set will ignore your evidence and continue on their merry way claiming there is nothing wrong with normal USB implementation and keep dishing out their flawed advice.

 

Perhaps you can put this post in a blog for easy reference by others ?

Also, it deserves posting in a more friendly area of the Forum such as the Uptone area where members trying to sort out these problems can more readily locate this information.

 

Kind Regards

Alex

[Bob Stern]

1 hour ago, sandyk said:

Unfortunately, the closed minded Objective set will ignore your evidence and continue on their merry way claiming there is nothing wrong with normal USB implementation and keep dishing out their flawed advice.

 

You are unnecessarily degrading the friendliness of the forum by directing nasty criticisms against a hypothetical entity.  Even if you were defending your position in a genuine disagreement, you should do it politely.  Gratuitously throwing insults in response to non-existent enemies is inexcusable.  I hope you find another forum to vent your spleen.

[sandyk]

25 minutes ago, Bob Stern said:

 

You are unnecessarily degrading the friendliness of the forum by directing nasty criticisms against a hypothetical entity.  Even if you were defending your position in a genuine disagreement, you should do it politely.  Gratuitously throwing insults in response to non-existent enemies is inexcusable.  I hope you find another forum to vent your spleen.

 

 You haven't recently been labelled as a Liar in the General Area of the forum , have you ?

 A couple of members here mercilessly attack almost every post that I make in the General Area of the forum.

I agree though that my comments were coloured as a result of these recent personal attacks, and shouldn't have been made.

 e.g.

  https://www.computeraudiophile.com/forums/topic/38000-operating-systems-and-there-sound-signatures/?page=7&tab=comments#comment-759699

 

[Ralf11]

wow - take a nap

[gstew]

6 hours ago, YashN said:

 

@gstew Some pics added! 

And very nice pix and process. Thanks for sharing!

 

Greg in Mississippi

[marce]

It shows nicely how bad lap tops are for noise, just look at the 3M products for EMC inside a laptop. Interesting pics,  curious as to where the really low level noise if from 2-4Hz?

Despite SandyK being a bit emotional, the majority have agreed that generic computer USB outputs can be noisy, Laptops being particularly bad. I would be interested in the new generation of laptops/tablets, even less space so more compromises.

Again though, one would have thought that the DAC input should have filtering on the supply if it is designed to be USB powered.

[sandyk]

32 minutes ago, marce said:

It shows nicely how bad lap tops are for noise, just look at the 3M products for EMC inside a laptop. Interesting pics,  curious as to where the really low level noise if from 2-4Hz?

Despite SandyK being a bit emotional, the majority have agreed that generic computer USB outputs can be noisy, Laptops being particularly bad. I would be interested in the new generation of laptops/tablets, even less space so more compromises.

Again though, one would have thought that the DAC input should have filtering on the supply if it is designed to be USB powered.

 

Marc

 Give it a few days, and for quite a few members it will be back exactly the way it was. : You don't need better USB cables, you don't need USB widgets, you don't need low noise Linear PSUs etc., AND if there IS a problem, it is due to poor design of the USB input of the DAC !

 Are you so naïve to believe that a single post like this will change those members minds ?

All that matters according to some, is that the 1s and 0s leaving the USB port are correct.

 

Alex

[YashN]

6 hours ago, marce said:

It shows nicely how bad lap tops are for noise

 

Again though, one would have thought that the DAC input should have filtering on the supply if it is designed to be USB powered.

 

It wasn't a laptop, it was from a 2009 24" Core 2 Duo iMac running Mavericks.

 

This said, ifthe laptop is battery-powered rather than having the power cord connected, there could be a possibility that there is less 'noise'. Problem: my 2009 MBP runs El Capitan and that old M-Audio card doesn't even run on newer OS, so I haven't been able to capture that one.

 

Now, about DACs 'should have filtering': this is happening, the newer designs do have them. I was supposed to review one of the new DACs but unfortunately couldn't because of being taken up.

 

You expect things - but that means nothing in reality. Reality and your assumptions must be tested in real life. Most people aren't even thinking about this as an issue at all despite some people warning about grounding issues, leakage currents and so on.

 

How many people actually sit down to dissect a generic USB cable and see how it's made inside, vary the geometry and listen, verify how the shielding configuration is, test all combinations at the shielding / connector interface and listen and try to make sense of it.

 

Not many. Most people have a model inside their heads that goes like this: 'It's digital, as long as the bits arrive at the other end, the sound shouldn't change, why don't you capture the DAC outputs and show us.' 

[YashN]

6 hours ago, sandyk said:

 

Marc

 Give it a few days, and for quite a few members it will be back exactly the way it was. : You don't need better USB cables, you don't need USB widgets, you don't need low noise Linear PSUs etc., AND if there IS a problem, it is due to poor design of the USB input of the DAC !

 Are you so naïve to believe that a single post like this will change those members minds ?

All that matters according to some, is that the 1s and 0s leaving the USB port are correct.

 

Marce isn't totally guilt-free in this respect 

[YashN]

A few other things I would have liked to do at the time I made the captures:

 

Since the custom USB connector  - by that time I couldn't even call it a USB 'cable' anymore, as there were components in a circuit computer-side as well as along the GND line - isolated the DAC quite a lot, and this was even enhanced by a custom shielding configuration - I started to think about what I was doing at the capture side.

 

The M-Audio can  be powered two ways: by its own external PSU or by USB (the same very noisy computer USB supply from the same iMac - oh no!).

 

So here, I chose not to use the USB power from the iMac through the M-Audio USB cable, but to power the soundcard by its external power supply.

 

However, this doesn't guarantee me that whatever noise is being transmitted via a generic USB cable from the Computer still doens't affect the card itself and therefore the measurements.

 

I could only assume (and here for lack of time and better equipment) that since the M-Audio is a card used for music composition, so also for capturing Microphones and other instruments ar 24-bit 96kHz max, that the manufacturer did some work for the isolation (its an older USB implementation so doesn't require the more recent and previously inexistent USB 2+ galvanic isolation implementations).

 

Secondly, the M-Audio external power supply I used was also not a Linear one...

 

To make things better for measurements, I would have liked to:

 

1. Have a clone of my Custom USB Connector to use that too for the M-Audio

 

2. Ideally power the M-Audio by an external, low-noise battery or something like that

 

 

Additionally, a few things to notice:

 

- I specifically did not work directly on isolating the data lines. However, filtering computer-side still improves SQ.

 

- The filtering computer-side still works when I A/B it (I installed a switch and a LED) even after I set up a custom shielding configuration for the USB Connector (which by itself already increased SQ too)

 

What this seems to suggest is that the data lines are still affected by whatever is happening computer-side, although I am very deliberately not powering from the computer, and although I also further isolated by working on the shielding configuration.

 

And finally this:

 

At this point in time, I thought I had a really good setup and that any leakage currents were not passing through my custom USB Connector, so I was thinking that if I re-connected my AC Filter in the system, it wouldn't make a big difference.

 

Here again, big difference!

 

This is what allowed me to say last year that I have a long-term plan of avoiding SMPSes in the listening setup: the work needed to really filter out all the nasties with SMPSes is extensive: you need to ensure the interconnects do not propagate unwanted things, you need filters post-SMPS as well as filters AC-side.

[YashN]

13 hours ago, gstew said:

And very nice pix and process. Thanks for sharing!

 

Greg in Mississippi

 

Wish I had taken some more pictures and captures before and after a layout change.

 

Here again, 'just a layout change' right? That means just changing where the capacitor returns pre-regulator went (the overall schematics and circuit did not change).

 

Huge SQ difference.

 

What this showed me is that even at that micro level, loop areas, connections, resulting EMI/RFI, returns/'grounding' can have large effects.

 

You can have 3 different PCBs with slightly different layouts, from the same schematics giving you vastly different results.

[YashN]

Relevant: SMPS and Grounding

 

A portion of leakage currents of SMPSes can be shunted by that technique.

 

If I'm not mistaken, this is similar to what the Tripoint 'grounding' boxes are doing (albeit with some additional passive filtering and other things) because the box is indeed connected to Earth too (as opposed to the early Entreq boxes which specifically only do what they term 'signal grounding' with an external box, some additional material and no connection to Earth).

 

So I think John found the explanation of that the Tripoint does: it is shunting part of Leakage Currents to ground, allowing this diversion not to affect system SQ as much.

[YashN]

52 minutes ago, YashN said:

 

[Cornan]

15 minutes ago, YashN said:

Relevant: SMPS and Grounding

 

A portion of leakage currents of SMPSes can be shunted by that technique.

 

If I'm not mistaken, this is similar to what the Tripoint 'grounding' boxes are doing (albeit with some additional passive filtering and other things) because the box is indeed connected to Earth too (as opposed to the early Entreq boxes which specifically only do what they term 'signal grounding' with an external box, some additional material and not connection to Earth).

 

So I think John found the explanation of that the Tripoint does: it is shunting part of Leakage Currents to ground, allowing this diversion not to affect system SQ as much.

 

Spot on! My conclution exactly (as I've posted on another thread). I connect my JSGT to Entreq grounding box which actually sounds better than connected to the same star-earth wired PSD as my floating SMPS in my setup (floating SMPS is connected to a balanced IT with floating secondary).

Dual floats/GI/isolations is another thing that I have found improves SQ vastly everywhere I've tried it. No isolation is absolute I'll guess! ?

 

 

[marce]

1 hour ago, YashN said:

 

Marce isn't totally guilt-free in this respect 

You and many others know my stand on certain aspects of the hobby.

[gstew]

2 hours ago, YashN said:

A few other things I would have liked to do at the time I made the captures:

 

Since the custom USB connector  - by that time I couldn't even call it a USB 'cable' anymore, as there were components in a circuit computer-side as well as along the GND line - isolated the DAC quite a lot, and this was even enhanced by a custom shielding configuration - I started to think about what I was doing at the capture side.

 

The M-Audio can  be powered two ways: by its own external PSU or by USB (the same very noisy computer USB supply from the same iMac - oh no!).

 

So here, I chose not to use the USB power from the iMac through the M-Audio USB cable, but to power the soundcard by its external power supply.

 

However, this doesn't guarantee me that whatever noise is being transmitted via a generic USB cable from the Computer still doens't affect the card itself and therefore the measurements.

 

I could only assume (and here for lack of time and better equipment) that since the M-Audio is a card used for music composition, so also for capturing Microphones and other instruments ar 24-bit 96kHz max, that the manufacturer did some work for the isolation (its an older USB implementation so doesn't require the more recent and previously inexistent USB 2+ galvanic isolation implementations).

 

Secondly, the M-Audio external power supply I used was also not a Linear one...

 

To make things better for measurements, I would have liked to:

 

1. Have a clone of my Custom USB Connector to use that too for the M-Audio

 

2. Ideally power the M-Audio by an external, low-noise battery or something like that

 

 

Additionally, a few things to notice:

 

- I specifically did not work directly on isolating the data lines. However, filtering computer-side still improves SQ.

 

- The filtering computer-side still works when I A/B it (I installed a switch and a LED) even after I set up a custom shielding configuration for the USB Connector (which by itself already increased SQ too)

 

What this seems to suggest is that the data lines are still affected by whatever is happening computer-side, although I am very deliberately not powering from the computer, and although I also further isolated by working on the shielding configuration.

 

And finally this:

 

At this point in time, I thought I had a really good setup and that any leakage currents were not passing through my custom USB Connector, so I was thinking that if I re-connected my AC Filter in the system, it wouldn't make a big difference.

 

Here again, big difference!

 

This is what allowed me to say last year that I have a long-term plan of avoiding SMPSes in the listening setup: the work needed to really filter out all the nasties with SMPSes is extensive: you need to ensure the interconnects do not propagate unwanted things, you need filters post-SMPS as well as filters AC-side.

 

YashN,

 

All of this sounds good. AND should be  a good set of guidelines and areas of inquiry for those looking to optimize their audio setup.

 

AND to me one of the most important points is starting with really low-noise power supplies, at their output, on how much noise they feed into the AC line, and how they react with other gear (John Swenson's high-impedance SMPS leakage). 

 

One of my guiding principals has been to eliminate sources of noise rather then try to filter or shield from them. My thought is that once you have a particular noise source, filtering or shielding will only reduce its impacts, not eliminate it. The best way to have low noise is to not create or introduce it in the first place.

 

So your long-term plan of avoiding SMPSs in my setup is something I've worked hard to achieve... and with just recently moving to amplifiers that don't use SMPSs, I've gotten there. It was good. AND well worth the effort and expense to get there.

 

Greg in Mississippi

 

[sandyk]

3 hours ago, YashN said:

 

Marce isn't totally guilt-free in this respect 

 

 No, but he appears to be far more open minded than most though.

We need people like Marc to use their vast experience to help further improve this area.

Perhaps your detailed results may help to at least cause a few people to investigate further themselves with their own gear, and then perhaps pass on any additional information gained ? 

[Cornan]

6 minutes ago, sandyk said:

 

 No, but he appears to be far more open minded than most though.

We need people like Marc to use their vast experience to help further improve this area.

Perhaps your detailed results may help to at least cause a few people to investigate further themselves with their own gear, and then perhaps pass on any additional information gained ? 

+1 @marce is a great guy with a lot of knowledge and an open mind to unexplained things. He is a great resource here on CA IMHO. So are you @sandyk! ?

 

[marce]

Thank you both and happy new year to all at CA. For my self I am glad to be rid of 2017.