[johnjen]

So was going to finish my 800 mod write up first, but this sorta jumped out at me that I could at least start this series.
And it is a long series, unless I cut it short and just post the conclusions, but the sequencing and unfolding of this story is flexible, based upon reactions and my desire to pour this much effort into all of this.
Because as it looks now I'll probably be posting this in like 6 or 7 parts if I break up the sections listed below, which are already long and there is more to add as I refine all of this.

So, on with these results of my investigations.

Why DO power cables and such make ANY difference to our headphone gear?

Yet another experiment focusing on power delivery and some not generally understood factors to consider.


Part A Why Would I Want To Do This In The First Place?
Part 1 Theory and Expectations Based Upon My Initial Research
Part 2 Measured Results and Observations.
Part 3 Analysis and Conclusions.


Part A Why Would I Want To Do This In The first Place?
So first off, just what IS this all about, let alone why take the time and spend actual $$ to investigate a topic that is the subject of such consternation?

And I kept waiting to see if anyone else would venture down this path, a path that seemed kinda obvious to me, one that would help shed some light on this rather polarizing subject.
I mean I figured the Sound Science guys would have jumped on this already, just to get to the bottom of this.
But instead it’s an Audiophool who is jumping down this rabbit hole.
Imagine that!

But I WAS truly puzzled why changes to the power feed would, or even could, make much if any, let alone such a profound change, to the SQ.
And in this case using gear that uses such small amounts of current, these sorts of effects just shouldn’t be that big a deal.
At least that is what the prevailing theories were informing everyone, was the case.
But that wasn’t what I observed, nor was it as simple as it first appeared.

Because after a while everytime I experimented I honed in on a combination of factors that made an obvious change to the overall SQ, but it wasn’t just change for changes sake.

No, indeed.
These steps resulted in ‘Better’ SQ and in ways that were very obvious to me.
They also helped me to define just what it meant to be ‘better’, vs. just a change, which is a critical factor in any investigation such as this.
And these changes provided enough motivation such that I kept track of theses changes over time and took notes in a journal as the hours accumulated.
And especially, all the attempts which resulted in no change, at all, were very helpful, in that they showed me what factors weren’t important, nor made any difference at all.
They saved me a whole bunch of time.

And after a while patterns emerged, certain types of SQ changes became describable and most importantly scaleable.
Meaning they were additive and after a while what was being subtracted from the acoustic presentation became as important as those, that were being added/changed to the experience.

This curiosity drove me to dive deeper into this end of the pool to see what I could see.
As such this is the second set of posts (The ASCC test results were a set, as was the investigation into fuses), that examines and addresses this issue and provides further context for the findings I’m presenting.

Electrifying Thoughts…

So for a while now I’ve been pondering the specifics of how our power supplies, the LPS variety not the SMPS type, actually couple with the live electrical grid, you know, the 120Vac 60Hz variety (US standard) and then how that power is actually transferred into its ‘reservoir’ to then be used by the active downstream load.

And I figured using my ‘spare’ amp , a Schiit 1st generation Mjolnir, which draws such a small amount of current, that cables and such really shouldn’t make much if any real difference to SQ.
Yet they did and still do.

And using it as a test device with its tiny amount of demand for power, would be a ‘worst case’ test scenario, to see if there were any measurable results that would be useful in my investigations.
Besides, it was available and I wasn’t planning on destroying the amp, even though I would have to modify it to be able to take the necessary measurements.

So the plan was formed and I knew it would require some cross corroboration from several different angles to zero in on what was going on.
And so I looked into fuses, and also how much current was available at the duplex receptacle (ASCC tests), and this new set of experiments looks at how the amp wants to be fed current.

And I’ll tie all of these different aspects together and see what the results are all pointing to.

End part A

Next up Part 1 Theory and Expectations Based Upon my Initial Research.

JJ

 

[johnjen]

A system status report.

Yesterday I finished cooking all of my cables except my AES cables, which I'll get to in the next couple of days.
They were re-installed with cleaned up contacts and with an oil change.
I also have ≈ 250hrs on my 'B' set of 800's and I figure by now I have over 1Khrs on my way over priced SR Black duplex receptacles.

The SQ has yet again taken a sizeable step up, everywhere and by every criteria I use to evaluate them.
And this is on my PWD dac, not my JggyB

When it arrives, I figure the combination of it and all of these newly introduced tweaks will take the SQ up yet another step.

Right now, as I type this it's hard to pin point the specific changes but collectively the overall acoustic presentation is Whole-istic, which is similar to HE (Hear Everything), but with even further reduced tLFF (the Listener Fatigue) which results in more acoustic power with less volume which makes for even stronger SDSG (SuperDuperSuperGlue) effects, (like seemingly falling asleep in my chair, until the album ends).

It could also be described by saying every 'voice' is whole and complete and doesn't bleed into any other 'voice' in either harmonic content, nor in time.
This is highlighted when duets are sung and both voices are similar in tonal structure, and especially when the background vocals are recessed.
This is where each voice is clearly evident and any voice can be followed with ease.
This is a level of focus and differentiation that makes the music compelling, and like HE (Hear Everything) where both everything heard 'belongs' to it's 'parent', and there is no acoustic energy that doesn't belong to that same parent.
IOW No smearing thru time of any aspect of each and every 'voice', nor interference or overlap between any of them as well.

It's like when you clean a window and do a perfect job such that the glass disappears, and there is no distortion, so what you see is shown with perfect clarity.
It's kinda like that.

JJ

 

[johnjen]

So my JggyB is back and in my system.
And during the time, while using my PWD dac, several other changes to my system setup were also used, along with several different means of delivering the digital audio signal, and I noticed an interesting pattern emerge.

Each time I took a step 'down' in gear (the PWD, using USB, vs SPDIF vs AES etc.) the changes weren't 'that bad', meaning, seemingly the SQ didn't deteriorate all that much and tunes were still enjoyable and engaging, still T3 and HB&W and such, and they stayed that way.
But, when a step up was implemented, the changes were much more noticeable, but only after a while (10-20 minutes), and only then did those highly sought after nuances and refinements start to standout.
I went thru these sorts of shifts several times as different equipment and signal paths were employed and it was after that 10-20 minute 're-adjustment' period where those "nuances and refinements" were 'remembered'.

What this says to me is that, blind A/B testing has a fundamental imbalance/flaw inherently embedded within that testing modality.
And my guess is that this 10-20 minute 'delay' before the 'good stuff' is perceived again, is also a function of the ability of our internal auditory processing, and is not solely a matter of the equipment itself 'settling in'.
Which in turn implies that (relatively) quick A/B switching is inherently hampered by this disparity, which in turn would tend to favor the step down in SQ vs. the step up simply because the step up takes much more time to become evident.

This, in my mind helps to explain why the use of blind A/B testing is so often confusing and misleading.
Because if the changes that a step up vs step down need, are different (and the basic assumption all along has been they are equal), then the results are by definition 'tainted'.
And even if both the up and down steps were treated appropriately (to better determine which was which) 2 things would happen.
#1 the time before a sufficient determination could be made (10-20 minutes) would greatly lengthen the test which would effectively kill our short term memory (a key aspect of the test).
#2 and it would essentially 'tag' the 'better' choice simply because after that 10-20 minute 're-adjustment' period additional "nuances and refinements" would be give themselves away.
And this is predicated upon our ability to hear and recognize these differences of "nuances and refinements" to begin with.

For me anyway these 'quick and dirty' sorts of tests seem to always have a weakness of some sort or other.
And as long as you are aware of them and take them into account, they can yield results of value, but when these foibles are either ignored or are simply not known, so that they aren't taken into account, they can, all to easily be a source of problematic results.

JJ

 

[gefski]

Great write-up, JJ. Reminds me of my recent "step back" to first few days of new Yggy-B (which I described as like a "good" $200 dac) after 3 yrs of almost daily listening to Yggy-A.

And quickie a/b listening, as you point out, doesn't get us to the nuances that make the music "alive". It seems to be useful for people just starting out, as it shines a light on frequency response variations and aberrations (too bright, etc.).

 

[johnjen]

Thanks gefski.
It always bugged me that this generally help insistence of A/B-x'ng was the De-facto way to know what is 'better' from what is not.
Granted it can let you know fairly quickly if there are differences, but taking it to the next level of which is 'better' just always felt skewed.

And right now my JggyB is blossoming, HMMAIAA (Hearing My Music As If Anew Again) and it's only got ≈ 60hrs so far.
So SDSG is in full effect, even though it's time for the dogs to get their midnight snacks.

Even their use of their doggy super powers, their "Power Stare", isn't sufficient to break thru my SDSG.
hahahahahahahahaha

JJ

 

[gefski]

You're a real disciplinarian!

 

[johnjen]

I gotta crack that whip because if I don't, Bear (the Chesapeake), will choose to ignore me even more than he already does.
Not to mention Punkin, (the older lab who knows how to play the game) will completely ignore us, UNLESS there is food involved.

Maintaining my Alpha status is a never ending challenge with smart dogs.
They both know they can do, what they want, how they want, whenever they want.
And since they are food motivated we still have an edge, which I have learned to use, and in ways, that I never imagined would be needed.

Bwwwhahahahhhahahaha.

JJ

 

[Lakson]

Interesting comments. For me the question is: Do we like to be decived? Is the beautiful lie better than than the plain reality? The answer lies within
ABX test in terms of small differences brings confusion and fatige, if nothing else. It's quite far from enjoyment we can have in this hobby.

 

[johnjen]

Good points, all.

And then there is the perspective that precedes, "Do we like to be deceived?"
That being, are we aware we are being deceived and that we are doing the deceiving to ourselves?

This, in part, is where we 'blindly' rely upon 'experts' to tell us what is, and is not true.
All the while, the answers DO reside within.
But it does take effort and perseverance to seek them out, in contrast to 'blindly' accepting the edicts of 'experts'.

Of course some topics require experience with, and a sufficient understanding of, the details and factors involved, which is where the perseverance applies.

And of course this isn't to imply that the 'experts' are wrong per se, but it is to say it should remain our 'goal' to verify claims with our own direct experience.
In this way we learn what does and doesn't 'ring true', because we do have direct personal experience to support our opinions and perspectives.

Just my additional 2.5¢

JJ

 

[johnjen]

So, a status report on my system.

The JggyB is blossoming, like crazy, and only has ≈125hrs thus far.

It has eclipsed my Jggy on tracks I'm very familiar with, and in surprising ways at that.
Like the soundstage has 'expanded' in all 4 dimensions (x-y-z-time), with greater definition, focus and REALNESS, and not in simple terms like bigger or deeper etc.
But in terms of being able to hear the entire soundstage, as it adds its reverberant harmonics to each of the 'voices' as they excite the room.

The bass is much better, but hasn't quite equaled the PWD, quite yet.
But is significantly 'better' than my original Jggy in definition, extension, and tonality, and being able to focus upon any bass line and follow it with ease.

There is a degree of refinement to the top end I have NEVER heard before.
Cymbals, massed violins, vocal plosives such as 'P's T's & S's have added delineation and focus and tonality that is compelling, where before they were just 'there'.
Think, hearing the rivets on the outside edge a cymbal, as distinct sources of added sound to the cymbal.

Talk about moisture on the breath, as it's about to drip on the floor…

I could go on, but I think I'll just kick back and be amazed at the degree and types of improvements, with more to come as it continues to 'settle in'.

This upgrade is WAY kewl.

JJ
ps as an 'acid' test I played those tracks that can all to easily act like dull razor blades inside my ears.
It was music instead and was engaging at that.

 

[gefski]

Reading around, to your impressions and others, it appears that the upgraded Yggys "ripen" faster than new Yggy-Bs. My new unit certainly wasn't a rewarding listen (compared to my sold Yggy-A) at 100ish hours; rather, it needed 400, then the switch flipped (like buying 10 avacodos, waiting a few days, then they all ripen in the same hour, Conni has to make a huge bowl of guacamole, and we have to eat it NOW).

In both cases, the rewards are significant.

 

[johnjen]

I have been keeping a log of changes to my systems SQ, as I make changes to the setup and gear for over 5 years now.
And I have observed certain time frames that have repeatedly demonstrated periods of 'blossoming' and other periods of TD (toilet dump) episodes.
And granted these repetitive cycles don't always occur but they have frequently enough, that it doesn't surprise me when they do.

And having the system's SQ blossom in the 125-150hr accumulation window, is one of the more common patterns.
Another is blossom action at the 325-375 time window, which can be followed by a TD at ≈400-460hr accumulation window (which thankfully is much more rare).

Of late, due to the ongoing reduction of the number of choke points in my system, many of these patterns have either not manifested at all, or have done so but rather weakly.
Except for this latest JggyB change to the system.
In this case the blossom action was rather pronounced and a most welcome change at that.

And if the SQ continues to follow previous patterns, the JggyB should REALLY blossom, big time, after the 460hr window passes.

So I'll continue to monitor (as in listen to my music) as it accumulates hrs and more fully settles in.

JJ

 

[Currawong]

I'm curious to find out more about your HD800 mods. I gave up on them because it seemed the more I modded them, the more precise, but more boring they became to listen with. I never tried a hard-wire mod though.

 

[Tand2016]

I sent my Rednet 16 R to SOtM for modifications some time ago. When it was done I figured it would be interesting to modify my 2012 i7 Mac mini also.

The Rednet and Mac mini has been modified with 2 sCLK-EX clock boards, both placed internaly in the Rednet. It had space to both since I have modified it with a LPS-1.2 from UpTone Audio and could remove the original PSUs.

The first sCLK-EX card is conneced with 2 tabs to the Rednet and 1 tab to a modified D-Link switch I also have on order, the 4 tab is still unused. The second sCLK-EX card is connected with all 4 clocks to my Mac mini. eABS-200 has been used both in the Rednet and the mini.

The clock cables to my mini will be 40 cm, maybe I will try some shorter later on.

Looking forward to receive them in a couple of weeks and go back from usb to AOIP again ....

 

[johnjen]

I sent my Rednet 16 R to SOtM for modifications some time ago. When it was done I figured it would be interesting to modify my 2012 i7 Mac mini also.

The Rednet and Mac mini has been modified with 2 sCLK-EX clock boards, both placed internaly in the Rednet. It had space to both since I have modified it with a LPS-1.2 from UpTone Audio and could remove the original PSUs.

The first sCLK-EX card is conneced with 2 tabs to the Rednet and 1 tab to a modified D-Link switch I also have on order, the 4 tab is still unused. The second sCLK-EX card is connected with all 4 clocks to my Mac mini. eABS-200 has been used both in the Rednet and the mini.

The clock cables to my mini will be 40 cm, maybe I will try some shorter later on.

Looking forward to receive them in a couple of weeks and go back from usb to AOIP again ....

snip

Wow That is a HUGE undertaking.
Dealing with RF is (or can be) a hall of mirrors, or be wonderful.
What do you figure you'll notice in terms of SQ?

And it looks like you patched right into the output of the existing clocks (as a guess), but how did you disconnect the original clock?

And that is a really good job of equalizing the word clock inputs.
Nice implementation, nice job all the way around!

I'd really like to hear more especially as the parts come in and the full system is dialed in.
I figure if yer gunna push the envelope, do it up.

JJ