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

[YashN]

On 4/1/2018 at 2:27 PM, jabbr said:

 

Very good start.

What I would like to see are some good measurements.

 

I did recording captures at the DAC output, using a pure Sine generated by Audacity using the M-Audio Fast Track Pro as capture interface, then did FFT on the results. Maybe you missed those? That's in and for my system. Not ideal for sure (I did mention caveats in my post).

 

For good measurements, go ahead and do the same in your own system. I am fairly sure you have better measuring gear than I do.

 

 

Quote

 

Normally I'd say listening is fine and listen to what you like HOWEVER in the case of grounding boxes which have no rational explanation based in what physics of electromagnetism we understand, if the claim so goes against traditional electronics I ask for real measurements to support a difference. 

 

I disagree with your perception of 'no rational explanation': from what I gather it is the same phenomenon by which grounding works within a system, apart from the intriguing thing that some people mention with Entreq, i.e. it takes some time for the effect to come on. I haven't experienced this in my system: the external 'grounding' effect occurs immediately.

 

So for my own experiments: I see nothing other than the simple contraption allowing dirty currents away from areas where they were previously wreaking sonic havoc.

 

It's nothing more than improving returns IMO, so no exotic or new physics is needed here (apart from that thing above).

 

What's not needed though is the bookish statism and inherent misconclusions ("this shouldn't make a difference, therefore this doesn't work").

 

 

Quote

 

Our ability to measure electrical phenomena goes way above our ability to listen, not that we always know what we are measuring but for example, if there is less noise on the ground plane then that is measurable. 
 

 

@PeterST has done a lot of measurements exactly like that, especially at the computer-side planes IIRC.

 

As for me, I have nowhere near the gear I'd like to do extensive and more precise measurements but you probably do, don't you? So the captures and FFT are what they are.

 

 

Quote

If you are trained in science and engineering once your listening test detects a difference, you validate this with measurements. You can't publish a physics paper without either theory and/or measurements.

 

I agree.

 

However, I am not publishing any paper here.

 

Quote

No one does, but extraordinary claims require extraordinary validation. Listening tests do not alone validate nor invalidate physics claims -- perhaps psychology.

 

Try it for yourself like in a build, listen, and make your own conclusions and measurements.

 

You may not have any better sound, but that's fine too by me: it might be that your system already is properly grounded throughout, so no additional benefit can be obtained.

[sandyk]

25 minutes ago, YashN said:

The system still sounds good, but I know it's several levels below its potential.

 

 As are the systems of those who believe that correct check sums guarantee that your cheap SMPS Laptop will sound as good as a very low electrical noise computer. All that the checksums guarantee is that you MAY be able to REGENERATE files that sound CLOSE to that of the original Master.

[YashN]

6 minutes ago, sandyk said:

 

 As are the systems of those who believe that correct check sums guarantee that your cheap SMPS Laptop will sound as good as a very low electrical noise computer. All that the checksums guarantee is that you MAY be able to REGENERATE files that sound CLOSE to that of the original Master.

 

What's up?

 

Tell me about it. Been looking for some new cool gear to do multi-track recordings in my studio. Can you believe one manufacturer went from a Linear PSU to an SMPS in a new model of a multi-track interface and thinks they resolved an overheating issue like that?

 

So now, this is quite a common issue I have seen on the production side of things lately: various SMPS used in different pieces of equipment, even some new(ish) synths.

 

Mind-boggling, really.

[YashN]

On 4/5/2018 at 11:38 PM, jabbr said:

AFAIK Swenson’s Supply was an input choke filter on a relatively routine regulated supply 317? etc. Fine.

 

It's way, way more than that...

 

Build one.

 

Listen (especially if you can compare to a DAC which by default is USB-bus powered in a direct connection to a computer).

 

I did. I did use a 317 myself as this is what I had on hand, but his original reg is much better (and there are probably much better ones nowadays).

 

In my build I did several iterations post-reg until I was happy with the results - problem here is the 317 is not very good so you end up battling non-linearities because of capacitors and their interactions with that Reg.

[sandyk]

12 minutes ago, YashN said:

 

What's up?

 

Tell me about it. Been looking for some new cool gear to do multi-track recordings in my studio. Can you believe one manufacturer went from a Linear PSU to an SMPS in a new model of a multi-track interface and thinks they resolved an overheating issue like that?

 

So now, this is quite a common issue I have seen on the production side of things lately: various SMPS used in different pieces of equipment, even some new(ish) synths.

 

Mind-boggling, really.

 

But they aren't using cheap generic Asian crap. It's mainly the PSU area that governs how a file will sound.

 In a few minutes I have to head for Sydney for the day, but can go into depth with you via a PM later if you like.

[Ralf11]

3 minutes ago, sandyk said:

It's mainly the PSU area that governs how a file will sound.

 

 

 

?

[YashN]

5 minutes ago, sandyk said:

 

But they aren't using cheap generic Asian crap. It's mainly the PSU area that governs how a file will sound.

 In a few minutes I have to head for Sydney for the day, but can go into depth with you via a PM later if you like.

 

iPhones have been manufactured in Asia too, so there shouldn't be any definite conclusion as to the country of origin, it's more about the nature of the PSU itself.

 

I'll take a Simple Linear over an SMPS any day for audio, a Swenson Linear over any other Linears. Still haven't experimented with Salas-type PSUs though, that would be next on my list of things to do (too many to mention).

 

Hope Sydney's is great as Summer's here.

[Cornan]

8 minutes ago, YashN said:

 

iPhones have been manufactured in Asia too, so there shouldn't be any definite conclusion as to the country of origin, it's more about the nature of the PSU itself.

 

I'll take a Simple Linear over an SMPS any day for audio, a Swenson Linear over any other Linears. Still haven't experimented with Salas-type PSUs though, that would be next on my list of things to do (too many to mention).

 

Hope Sydney's is great as Summer's here.

 

Allow me to interupt! Floating SMPS is a truly great feeder suppply for a linear regulator. It is not a great PSU on its own, but a floating SMPS is ONLY concured by a floating and balanced AC-AC feeder supply in my world. Nothing else is better.

[YashN]

1 minute ago, Cornan said:

 

Allow me to interupt! Floating SMPS is a truly great feeder suppply for a linear regulator. It is not a great PSU on its own, but a floating SMPS is ONLY concured by a floating and balanced AC-AC feeder supply in my world. Nothing else is better.

 

That would be the output-grounding trick by John? Worth a try for sure.

 

Does it totally interrupt leakage currents? AC-side pollution, etc...?

[Cornan]

8 minutes ago, YashN said:

 

That would be the output-grounding trick by John? Worth a try for sure.

 

Does it totally interrupt leakage currents? AC-side pollution, etc...?

 

Not sure what you mean by output-grounding trick by JS?

Anyway, the balanced toroidal ac-ac transformers into dual 5A LT3045s is surely better than into Gophert csp-3205II. But, not by much! Small diffetencies in the bass region and dynamics. That’s all! 

[YashN]

1 minute ago, Cornan said:

 

Not sure what you mean by output-grounding trick by JS?

Anyway, the balanced toroidal ac-ac transformers into dual 5A LT3045s is surely better than into Gophert csp-3205II. But, not by much! Small diffetencies in the bass region and dynamics. That’s all! 

 

I thought that 'floating SMPS' was actually using John's output-grounding trick on an existing SMPS.

 

Doesn't seem like it is judging by your response? Would it be one where this grounding is already implemented? No idea, I don't care much about SMPSes anymore

 

I may still give the output-grounding a try and listen in my setup, especially at the DAC power side: this is where I have several configurations I can test and listen to.

 

However, it might be a while until I get time and will to do this.

[Cornan]

2 minutes ago, YashN said:

 

I thought that 'floating SMPS' was actually using John's output-grounding trick on an existing SMPS.

 

Doesn't seem like it is judging by your response? Would it be one where this grounding is already implemented? No idea, I don't care much about SMPSes anymore

 

I may still give the output-grounding a try and listen in my setup, especially at the DAC power side: this is where I have several configurations I can test and listen to.

 

However, it might be a while until I get time and will to do this.

 

I can tell you this. If you have a floating ground and it is not connected to a network

device do not care about JSGT (John Swenson’s ground tweak). If it you have a grounded PSU try it everywhere! ?

[Cornan]

@YashN I forgot to tell you that I missed your presence here at CA, You're a bit like politics. You just hate when they are neutral! ?

[jabbr]

1 hour ago, YashN said:

Build one.

 

Listen (especially if you can compare to a DAC which by default is USB-bus powered in a direct connection to a computer).

Err ... I have ... do you have a different schematic than the one I’ve seen? What I’ve seen was nice solid straightforward input choke filtered PSU. Oh perhaps not a 317 was it an LT1084 or similar? All those devices can be made to work well.

[mansr]

2 hours ago, YashN said:

Knowing Digital circuitry doesn't make one an automatic expert in analogue circuits.

High-speed (1 GHz and up) digital is very much an analogue affair. Go read the PCIe spec if you don't believe me.

[sandyk]

2 hours ago, YashN said:

 

I thought that 'floating SMPS' was actually using John's output-grounding trick on an existing SMPS.

 

Doesn't seem like it is judging by your response? Would it be one where this grounding is already implemented? No idea, I don't care much about SMPSes anymore

 

I may still give the output-grounding a try and listen in my setup, especially at the DAC power side: this is where I have several configurations I can test and listen to.

 

However, it might be a while until I get time and will to do this.

The grounding trick gave a small improvement via  Coax SPDIF with my STB which of course uses an SMPS.

[sandyk]

3 hours ago, YashN said:

 

iPhones have been manufactured in Asia too, so there shouldn't be any definite conclusion as to the country of origin, it's more about the nature of the PSU itself.

 

I'll take a Simple Linear over an SMPS any day for audio, a Swenson Linear over any other Linears. Still haven't experimented with Salas-type PSUs though, that would be next on my list of things to do (too many to mention).

 

Hope Sydney's is great as Summer's here.

The main requirement for any PSU is a very low and FLAT output impedance from almost DC to the MHZ region.If the use of Low ESR electro's sees the impedance much lower at say 100khz than midrange, then there will be an apparent increase in HF detail.

This is especially obvious when using Voltage regulators such as the ultra low noise LT3045 etc.when powering BOTH Analogue or Digital devices.  USB Memory for example.

[fas42]

5 hours ago, YashN said:

In fact, I recently removed it and despite all the other filtering and customisation/isolation between the computer and the external USB DAC, the holographic, detailed, 'they-are-here' effect is gone.

 

The system still sounds good, but I know it's several levels below its potential.

 

Yes, that's how it works - getting a high level of quality requires every element of the playback chain to be performing adequately; and if you remove, disable a single key area of that chain then the loss can be dramatic. It's a plateau of SQ that one is chasing - and the edges of that plateau are very steep ...

[Speedskater]

5 hours ago, YashN said:

On 3/31/2018 at 3:26 PM, Speedskater said:

When 'marce' posts engineering facts, I would put money on them being correct. ........................

You'd lose a lot of money, unfortunately.

Knowing Digital circuitry doesn't make one an automatic expert in analogue circuits.

My money stays on 'marce'. Digital circuitry is just a small part of his engineering knowledge.

[fas42]

One also has to be an expert on what distortion and noise factors are most relevant, important for the human hearing system ...

[marce]

On 6/21/2018 at 8:02 PM, YashN said:

 

You'd lose a lot of money, unfortunately.

 

Knowing Digital circuitry doesn't make one an automatic expert in analogue circuits.

 

Computer Audiophillia is about Mixed signal circuits and EMC. The issues we get are almost always where assumptions about Digital and Analogue are wrongly adopted or when the Digital viewpoint of robustness of transmission doesn't take into account any intermediate D/A and all the analogue issues that can accompany transmission in the analogue domain.

 

By now, attentive people reading here should know more about both USB and any intermediate A/D and D/A steps from reading a file on an HDD or elsewhere through to the DAC then onto Speakers. There are many such steps, each can be damaging to sound.

 

All we get from this supposed 'expert' still polluting this thread is how he thinks things should work.

 

Better stick to digital if that's all he knows about.

 

Cut it out, you don#t know what I work on and in fact do a lot more mixed signal, analogue, Rf and microwave than just digital...

As to your insult, no you wouldn't loose money, so stop the insults please, at least I am an expert in my field and what I work on is all pretty much high end electronics and is all mixed signal, in fact at the moment its microwave and digital.

 

[YashN]

On 6/21/2018 at 8:11 PM, Speedskater said:

My money stays on 'marce'. Digital circuitry is just a small part of his engineering knowledge.

 

Really, eh? Well, bet the farm then.

 

"To be totally correct"...

 

You lose.

 

Guess who corrected Speedskater on this very same subject years ago here.

 

[YashN]

On 6/21/2018 at 4:56 PM, Cornan said:

@YashN I forgot to tell you that I missed your presence here at CA, You're a bit like politics. You just hate when they are neutral! ?

 

Thanks man,

 

What's up? Many new audio experiments?

 

Been enjoying the fruits of my labour.

 

Can you believe much gear on the production side of music is badly designed as well? Crazy, huh?

[Speedskater]

On 6/28/2019 at 12:17 PM, YashN said:

Really, eh? Well, bet the farm then.

"To be totally correct"...

You lose.

Guess who corrected Speedskater on this very same subject years ago here.

Not a correction, just a clarification.

And your problem with either of our posts is what, exactly?

[Cornan]

On 6/28/2019 at 6:19 PM, YashN said:

 

Thanks man,

 

What's up? Many new audio experiments?

 

Been enjoying the fruits of my labour.

 

Can you believe much gear on the production side of music is badly designed as well? Crazy, huh?

 

Great thanks! I have a very needed brake from this hobby right now, so not much new here. Just enjoying the fruits of previous labour really.

This hobby is a rabbit hole of improvements if you really start digging. I've stopped for a while, but expect myself to start digging again in the near future. I still have plans. Right now my other hobby photography is in the spot light. Enjoying every minute of it! 📸😊