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

[Cornan]

 

So, my idea for this, the remaining 'floating shield' is to shunt it to the Star-ground point once I get it up and running again, and listen for further SQ.

 

Another thing I would like to shunt to there would be the point before the filter and the GND isolation circuit computer-side.

 

Note that if you're using high-speed data lines un-shielded, the lines will be a source of EMI/RFI in the air. This could also affect your SQ, if your not DAC, other devices in the audio chain.

 

A star ground is great, but I really think that it just great if it is only one path to safety ground...and that is also the tricky part.

Otherwise I would recommend to drain the shield to a external grounding box which will serve as a "end-path" for the noise, instead of shunting it into the safety ground and possibly "pollute" other gears.

 

High speed data requires shielded cables...so I am not afraid that would happen to me!

 

 

 

Sent from my Nexus 7 using Computer Audiophile mobile app

[YashN]

A star ground is great, but I really think that it just great if it is only one path to safety ground...and that is also the tricky part.

 

Not sure what you mean by that.

Otherwise I would recommend to drain the shield to a external grounding box which will serve as a "end-path" for the noise

 

Same thing I said above: it is a sub-component of my AC Filter box. But actually there are two sub-components planned, one for chassis (same principle as default chassis inter-link or similar principle to Tripoint), one for signal (similar principle to Entreq).

[YashN]

Even more difficult to design a WAF-filter. I would say it is mission impossible. I have already decided. My next gf have to be a audio neard!

 

If you're WSS (Wife-Selection Skill) is high, you don't need to fear the WAF.

 

Cool pics!

[Cornan]

My Ethernet cable ends are metal I just tried one not sure of the sonic impact yet. I have to listen more.

Even if I have seen a lot of similarities between USB and Ethernet lately I am not sure you will gain as much just covering the shield on the Ethernet plugs. I have ordered parts to assemble my own cables with unshielded screw terminal RJ45 plugs and silver plated unshielded wires which I will connect to pin 2 & 4 and 3 & 6 (twisted pairs on a distance) inspired by @scan80269 experiments. This will be more similar to a USB cable, but I am still waiting for the parts plus I also need to connect FMCs between my router and streamer for best results. I will initially power the downhill FMC with my Kingrex uPowers linear regulated 5v output (and the unregulated 7,5v output powering Regen).

More of this later when I got everything together.

 

Sent from my Nexus 7 using Computer Audiophile mobile app

[Cornan]

If you're WSS (Wife-Selection Skill) is high, you don't need to fear the WAF.

 

Cool pics!

I'll surely need a lot of WSS-practise! ? I had a few but not none ever cared for good SQ even if they all loved music (last part always included in my personal "WSS plan"). How you can love music and not Hi-Fi at the same time is a mystery to me! ?

 

Sent from my Nexus 7 using Computer Audiophile mobile app

[lmitche]

Holy cow! I just covered all my outer USB barrels with electrical tape and it all works and sounds f***** amazing! The clarity was improved a lot by this simple tweak. My neck hair agrees with me! A must try!

 

Sent from my Nexus 7 using Computer Audiophile mobile app

Hi Cornan, thanks for this, I forgot to tape the barrels in my latest cable iteration. Done now, and another nice SQ boost!

[lmitche]

So, basically, in the iDSD Nano, since the internal battery always powers the DAC and the Clock - great job, iFi! - the only thing which was actually powered by the Computer USB port was the USB Receiver.

 

Hi Yashn, last week Eurodriver and Jud were here and with 3 lps1s on hand we tested any number of configurations with the IFI microIDSD DACs.

 

After testing power injection via the Aqvox cable, we settled on the internal battery as the best way to power the IFI. It could not be beat.

 

We also tested any number of USB solutions, and in the end definitely found my hacked $6 Chinese split cable to be the best sounding that I own here. Jud has a custom Mapleshade USB cable that sounded even better, which unfortunately left with Jud.

 

I just added a thin layer of scotch tape to the barrel of the data end of my split cable for a very nice SQ bump.

 

Thanks to you and Cornan for the inspiration.

 

Larry

[Cornan]

Hi Yashn, last week Eurodriver and Jud were here and with 3 lps1s on hand we tested any number of configurations with the IFI microIDSD DACs.

 

After testing power injection via the Aqvox cable, we settled on the internal battery as the best way to power the IFI. It could not be beat.

 

We also tested any number of USB solutions, and in the end definitely found my hacked $6 Chinese split cable to be the best sounding that I own here. Jud has a custom Mapleshade USB cable that sounded even better, which unfortunately left with Jud.

 

I just added a thin layer of scotch tape to the barrel of the data end of my split cable for a very nice SQ bump.

 

Thanks to you and Cornan for the inspiration.

 

Larry

You are welcome Larry!

I am a bit envious of Jud's Mapleshades cables. I find their AC cables extremely interesting...which I know Jud ownes as well. Many things that I personally believe in are incorporated into their design.

 

Sent from my Nexus 7 using Computer Audiophile mobile app

[Cornan]

Actually, due to the fact that GND now sound great in my 100% unshielded setup I think I will perform a still not realized idea which is to "by-pass" the USB Regen with my GND wire with a single GND wire straight from Aries Mini to my DAC/HPA not going through the Regen (powered by unregulated battery). I guess that could potensially shorten the GND path and be a good thing. I will let you know if it is as soon as my gf lets me "play with my toys" (her words)!

 

Sent from my Nexus 7 using Computer Audiophile mobile app

[austinpop]

@Cornan

 

With these latest revelations, could you perhaps summarize everything you recommend on the USB chain? I have lost track!

[Cornan]

@Cornan

 

With these latest revelations, could you perhaps summarize everything you recommend on the USB chain? I have lost track!

 

Well, here is a list. I will not go into details about DAC requirements:

 

1. Unshielded twisted wires. Gladly with unbleached cotton or silk sleevings. Preferably with a loose fit.

2. No 5v wire or clean injection with linear regulated battery or ultracaps.

3. If using 100% unshielded ICs, ac mains cables and a Tech IT keep the GND wire. If you use shielded wires somewere in the chain lift the GND.

4. Cover the USB barrels with electrical tape.

5. Keep the GND path as short as possible anywere in the chain

6. Power all DC devices with unregulated batteries or ultracap for isolation to the mains.

7. Possibly (ie theory) use tech IT for unshielded chain and balanced power supply for shielded chain.

8. Take great care of your safety ground path. Make it great and make it short.

9. Use external grounding to clean dirty shielding paths.

 

Hmmm...I think that should do for starters! [emoji4]

 

 

Sent from my iPhone using Computer Audiophile

[YashN]

Hi Yashn, last week Eurodriver and Jud were here and with 3 lps1s on hand we tested any number of configurations with the IFI microIDSD DACs.

 

After testing power injection via the Aqvox cable, we settled on the internal battery as the best way to power the IFI. It could not be beat.

 

I know, it's very, very cool to have a fully battery-powered DAC, and indeed I get to do that, and with GND fully disconnected it sounds fantastic.

 

However, when both the USB receiver section and the analogue sections are powered by the battery, with high-res and high volume, I can tell you my iDSD Nano cannot last very long.

 

It probably also has to do with the USB power which cannot really fully charge the internal battery to satisfaction.

I just added a thin layer of scotch tape to the barrel of the data end of my split cable for a very nice SQ bump.

Thanks to you and Cornan for the inspiration.

I have yet to try it myself.

 

Thinking that the USB port power injection is really a weak link. Ideally I would have liked a separate power input and the ability to totally disconnect the ground connection to the jack here potentially...

[YashN]

Thanks to you and Cornan for the inspiration.

 

You're welcome, thanks for your participation and reports

[Jud]

I know, it's very, very cool to have a fully battery-powered DAC, and indeed I get to do that, and with GND fully disconnected it sounds fantastic.

 

However, when both the USB receiver section and the analogue sections are powered by the battery, with high-res and high volume, I can tell you my iDSD Nano cannot last very long.

 

It probably also has to do with the USB power which cannot really fully charge the internal battery to satisfaction.

 

I have yet to try it myself.

 

Thinking that the USB port power injection is really a weak link. Ideally I would have liked a separate power input and the ability to totally disconnect the ground connection to the jack here potentially...

 

The micro-iDSD will run up to 10 hours on a charge, though after about 2 hours it will start drawing power through USB.

 

 

Sent from my iPhone using Computer Audiophile

[YashN]

About the barrels: if it's metal, some shielding with tape or with the usual injection of rubber or what not helps with capacitive coupling, but, that thing, if it's made of metal and still connected to ground (even though you disconnected the shield), is an antenna or emitter as well...

 

So I was thinking today, of actually re-shielding it with Aluminium.

[YashN]

The micro-iDSD will run up to 10 hours on a charge, though after about 2 hours it will start drawing power through USB.

That's the same claimed "10 hours" for the iDSD Nano, but if the power draw starts after 1 or 2 hours through the computer USB port, it's not a full 10-hour of battery autonomy for me because as soon as it really starts drawing power from the computer the SQ will drop greatly.

 

It looks to me that the 10 hours is something really optimistic and perhaps taking into account only the clock and DAC chip power autonomy, or really, really low volume and low resolution.

 

So, it seems to me as I mentioned previously, that if I could also do the main power to it from another bigger battery, that could be helpful, perhaps trickle-charged by a LRP.

 

Of course, an LPS-1 or the similar Vinnie Rossi solution would be ideal here.

[YashN]

Pics, guerilla-style testing:

 

The green LED is post filter and pseudo-termination of the computer-side USB port. Adele's "25" on vinyl on a Technics Direct Drive turntable makes an appearance below

 

Showing the third switch computer-side:

 

 

The two switches for the jack-shield can be removed, the proof has been made for me, so I will re-use them for further tests in the workshop/lab.

 

The switch on the filter is nice for immediate A/B although I am standing very near the speakers, and actually nearer to the right one when toggling it. Of course, the idea of an Arduino IR detection and Apple Remote is in my mind, but the effect is immediately noticeable though.

 

The second little circuit on the right of the LED + Switch circuit is the ground isolation.

Note how none of my power cables each side run alongside the data lines. They actually are at right angles (as much as this guerilla testing will allow). The isolated GND is away from the data line as well.

In the foreground is the SMPS iPhone charger, which despite its SMPS status appears to be doing quite a stellar job.

 

Doesn't mean I don't want to replace it with a cleaner power supply.

 

In the far background, you can detect a guerilla-style ball-and-cup arrangement with green marbles, which works to seismically isolate the DAC in its usual position.

Wanted to watch some show, but my girlfriend wants to listen to her collection. She says she wants to listen to all the 'big voices' now, even if she doesn't necessarily like them.

 

We listened to Sting - Fragile and The Eagles - Hotel California earlier, songs we have heard countless times, right? We still heard new performance details in both.

[YashN]

Now can you see why I've been saying that everything sounds great to listen too, regardless of subject material?

One thing I notice is that because micro detail retrieval is so high all of the atmosphere and production effects are much more prominent... the sound is much wetter, which in turn gives the illusion of softer, smoother, forgiving presentation... but with miles more emotion.

Things that used to be prominent are now not so and things previously unheard or not prominent are now completely 'there'... it sounds like different equipment.

Does that sound like what you are now hearing, compared to how it was before?

 

Here is some aural feedback from August 2015 when I had disconnected GND in my setup:

 

This morning, the sound is super clear and powerful (great dynamics and soundstage) and, best thing of all: when I tested for the high-pitched metallic hiss, it sounds like it's gone!

 

So, yes, I've been trying not to go back, but the practicality of charging so far has kept me back, and more recently iFi's warnings put a temporary damper on things, but I knew I had to make that GND as clean as possible for the DAC.

 

Another interesting extract from that post:

 

Another thing: I have seen reports of someone saying the Nano's circuit is floating inside the chassis. In other words, there is no internal star-grounding at the chassis, and one guy does a mod (among other mods) to the Nano to change the grounding scheme and get fantastic results.

I wonder if I could do the same.

 

However, I believe there was an issue with one of the modded units from that guy, so not sure it was done in the proper way.

 

The subsequent post from the same day has just a little feedback:

 

One thing I am noticing this morning while testing: that GND disconnection sure sounds very good

 

Which brings me back to my observations regarding how an Entreq box works (the one catering for signal 'grounding') and how my rudimentary external signal 'grounding' was working as well.

 

In a boxed circuit, like a DAC, it appears a power Ground is connected to the signal ground and that causes a lot of SQ issues.

 

When we use an external grounding box, are we successfully providing an alternate lower impedance route other than any internal signal + power ground paths for the problematic currents to go to?

 

It does seem it is the case.

 

A corollary question would be: if we succeed in externalising that route, it appears this is working while the internal route is also still physically there. Is it possible to remove that internal physical link in some instances? Would that have a better effect on SQ? Any at all? Any detrimental side effects to our circuits, like the extreme of having a malfunctioning one?

[Cornan]

About the barrels: if it's metal, some shielding with tape or with the usual injection of rubber or what not helps with capacitive coupling, but, that thing, if it's made of metal and still connected to ground (even though you disconnected the shield), is an antenna or emitter as well...

 

So I was thinking today, of actually re-shielding it with Aluminium.

 

Using electrical tape will not cure the problem ofcourse, it is just a simple tweak to make bad slightly better. Auminium foil would probably work as an experiment as well.

Please let me know how that turns out if you decide to try it!

 

Sent from my Nexus 7 using Computer Audiophile mobile app

[r_w]

I'm not sure, but probably our USB paths are now nearing optimal.

 

;-)

 

Here is some aural feedback from August 2015 when I had disconnected GND in my setup:

 

 

 

So, yes, I've been trying not to go back, but the practicality of charging so far has kept me back, and more recently iFi's warnings put a temporary damper on things, but I knew I had to make that GND as clean as possible for the DAC.

 

Another interesting extract from that post:

 

 

 

However, I believe there was an issue with one of the modded units from that guy, so not sure it was done in the proper way.

 

The subsequent post from the same day has just a little feedback:

 

 

 

Which brings me back to my observations regarding how an Entreq box works (the one catering for signal 'grounding') and how my rudimentary external signal 'grounding' was working as well.

 

In a boxed circuit, like a DAC, it appears a power Ground is connected to the signal ground and that causes a lot of SQ issues.

 

When we use an external grounding box, are we successfully providing an alternate lower impedance route other than any internal signal + power ground paths for the problematic currents to go to?

 

It does seem it is the case.

 

A corollary question would be: if we succeed in externalising that route, it appears this is working while the internal route is also still physically there. Is it possible to remove that internal physical link in some instances? Would that have a better effect on SQ? Any at all? Any detrimental side effects to our circuits, like the extreme of having a malfunctioning one?

[Theobetley]

Well, here is a list. I will not go into details about DAC requirements:

 

1. Unshielded twisted wires. Gladly with unbleached cotton or silk sleevings. Preferably with a loose fit.

2. No 5v wire or clean injection with linear regulated battery or ultracaps.

3. If using 100% unshielded ICs, ac mains cables and a Tech IT keep the GND wire. If you use shielded wires somewere in the chain lift the GND.

4. Cover the USB barrels with electrical tape.

5. Keep the GND path as short as possible anywere in the chain

6. Power all DC devices with unregulated batteries or ultracap for isolation to the mains.

7. Possibly (ie theory) use tech IT for unshielded chain and balanced power supply for shielded chain.

8. Take great care of your safety ground path. Make it great and make it short.

9. Use external grounding to clean dirty shielding paths.

 

Hmmm...I think that should do for starters! [emoji4]

 

 

Sent from my iPhone using Computer Audiophile

Just a question to all because you never can know with certainty about these things. What if the electrical tape on usb barrels damps acoustic feedback? I'm sure this contributes to the sonic improvement. Relative to the aluminum shielding idea I had this on my usb short cable running from the uRendu to my dac and when I replaced with black electrical tape I got a big sq pop. Maybe its just YMMV.

[lmitche]

You are welcome Larry!

I am a bit envious of Jud's Mapleshades cables. I find their AC cables extremely interesting...which I know Jud ownes as well. Many things that I personally believe in are incorporated into their design.

 

Sent from my Nexus 7 using Computer Audiophile mobile app

Hi Cornan,

 

Jud is definitely on to something with the Mapleshade / Omega Mikro cables. My Omega Mikro interconnects arrive tomorrow!

 

I have spent some hours on the phone with the two principals this week. Stay tuned, we are just at the beginning of this story.

 

Larry

[Cornan]

Just a question to all because you never can know with certainty about these things. What if the electrical tape on usb barrels damps acoustic feedback? I'm sure this contributes to the sonic improvement. Relative to the aluminum shielding idea I had this on my usb short cable running from the uRendu to my dac and when I replaced with black electrical tape I got a big sq pop. Maybe its just YMMV.

 

I am sure the electrical tape also could possibly dampen micro vibrations...but I am also equally sure that it is just a minor part of the whole picture. I am quite use to listen to how shielding and GND is affecting SQ since I have been eveluated this for years. The electrical tape improves the sound similar to GND vs no GND and shielded vs unshielded. It "clears the fog" and brings music to life so to speak. My guess (yes, it is still a guess) is that the isolation from the connected device ground is the overall gain with this tweak. I also guess that you would get pretty much the same result isolating the music source + DAC USB A and B female sockets.

Anyway, I will eveluate this further before I will say I know for sure!

[Cornan]

My Omega Mikro interconnects arrive tomorrow!

 

Great choise Larry! Enjoy and do not forget to post your impression of the Omega Mikro!

 

I have spent some hours on the phone with the two principals this week. Stay tuned, we are just at the beginning of this story.

 

Very intreresting! Just do not leave me standing on needles for too long!

[Cornan]

Got my special order from ATL Hifi today. It took me a couple of hours to carefully put my DC blocker trap filter together with my thick unshielded AC wire. The 11awg wires was a pain to fit into the internal screw terminal blocks, but finaly ended with success. Pretty happy with the result!

 

 

Plugged it pre my Peaktech 2240 tech IT. Not a huge difference in the beginning...maybe a little darker. After starting to listen to previously "difficult" tracks like for example Agnes Obels "Fuel to Fire" from the brilliant album "Aventine" I started to realize some great advantages.

 

 

This track can sound a bit like it is recorded inside a closet. With the DC blocker "the room" felt less closed than before. A blacker background. A sightly nuanced low bass performance maybe? Warmer mid-range?

Anyway, a pretty good start. Even if it is not a night and day change it will remain in my main system. Still early days, and I can tell it improves by the time I am writing this! Kudos to @Middy for finding it and for Alex at ATL Hi-Fi for managing my special order in a very short time and for no extra charge. I am happy! ?

 

A Merry Christmas to everyone! ??

 

 

 

 

Sent from my Nexus 7 using Computer Audiophile mobile app