ISO REGEN Listening Impressions (kicked off with some measurements)

[austinpop]

1 hour ago, scan80269 said:

I had a few hours of listening using the following setup with ISO REGEN added:

 

WIn10 i7-3770S AIO PC (HQPlayer & local library on SSD) <iFi Mercury USB cable> ISO REGEN (LPS-1 powered) <USPCB A>B hard adaper> Singxer SU-1 DDC (modified, LPS-1 powered) <ultra-slim 1.5' HDMI cable> Holo Audio Spring DAC <XLR cables> Jensen ISO-MAX PI-2XX <XLR cables> Auralic Taurus MkII headphone amp <Stefan Audio Art Endorphin balanced headphone cable> Sennheiser HD800

 

Spring DAC was in NOS mode.  Each LPS-1 fed by a modified Breeze Audio el-cheapo linear power supply.  No upsampling in HQPlayer.

 

Initial impressions:

 

WOW!!!  I almost could not recognize some of my favorite recordings!  

 

Firmer & more prominent bass, smoother treble, less sibilant, more articulate & expressive voices, more air around instruments, longer reverb tails, more liquid sounding, stronger dynamics along with improved details.

 

The sound is frankly hard to put into words.  A number of previously un-listenable recordings are big-time transformed!  The improvements are largest with Redbook 16/44.1K content, but even my HRx (24/176.4K) recordings sounded better like never before!

 

ISO REGEN (along with USPCB) is now my favorite USB grunge blocker.  I won't be surprised to hear ISO REGEN elevating the sound of USB over S/PDIF or AES interfaces for a wide variety of system setups.

 

Bravo, Alex and John!  ISO REGEN is another major winner!  

 

Back to more listening...

 

p.s. Alex, the ISO REGEN + modded SU-1 sounds like a killer combo for the Spring DAC.  Have you heard this setup yet?

 

p.p.s. That eye diagram for ISO REGEN is among the cleanest waveforms I've ever laid eyes on.  Awesome!

 

 

 

Great to read your impressions!

 

But - didn't you have an Aries-based setup? Did you switch to a direct-USB HQPlayer-based setup while I wasn't paying attention?  I'm curious how it sounds on a endpoint/renderer/NAA setup.

 

And - any comparison to previous USB decrapifiers - Intona, RUR, Regen, ...?

[austinpop]

35 minutes ago, Superdad said:

In fact, I'd venture to say that the impact of the ISO REGEN will greatest for those who simply have a standard computer running a standard USB cable to the ISO REGEN attached directly to their DAC with the included USPCB A>B Adapter.   

 

You're probably paying a lot more close attention to this, Alex, but is it me, or does it seem most of the initial glowing reports are from direct-attach users? I'm curious to hear how the endpoint crowd likes it - the mR/sMS-200/Aries gang! And also ISO-Regen compared to Intona+RUR/Regen.

[austinpop]

On 5/30/2017 at 10:54 PM, fpalm69 said:

Also,  will one LPS-1 power both the mRendu and the Iso Regen using a Y-cable?

Thanks!

 

Hopefully what I am about to say will echo @Superdad's views, but if not, he should correct me!

 

There are certainly applications where using a Y-cable to power 2 components from a single LPS-1 should work perfectly fine, with minimal loss of SQ. As long as the aggregate current demand of both components is within the 1.1A capability of the LPS-1, it should work. Obviously, the quality of the cable, the Y-adapter, etc will play a role, but overall, it's still a workable solution.

 

However...

 

The mR + ISO-Regen combo should not be powered by a single LPS-1. Why? because the ISO-R is an isolation device, with an electronic moat within it. Upstream of the moat, the circuit is powered by the USB device upstream of the ISO-R (in this case, an mR) using USB bus power. The ISO-R's PSU powers the circuits downstream of the moat, and provides the signal integrity, isolation, and clean bus power to the USB device downstream of it.

 

By powering the mR and ISO-R with the same LPS-1, you're effectively providing a common ground path around the isolation moat designed into the product. This is like turning the GI switch to off. Now, the audible degradation due to this is system-dependent. For some it may be minimal, for others, dramatic.

 

Interestingly, using a single LPS-1 to power both the ISO-R and something downstream (assuming the aforementioned constraints are met) should be just fine.

 

So - net, net - it's not about whether powering 2 devices with a single LPS-1 with Y-cable is bad. It's which 2 devices.

 

 

[austinpop]

Just now, tboooe said:

From my personal experience, using an LPS-1 to power the IsoRegen and my SU-1 did not sound as good as using separate power supplies.  I do not (yet) of a second LPS-1 but I did test powering the IsoRegen with a common external battery and it sounded better than sharing a single LPS-1 in my system.

 

I am not smart enough to understand the electrical implications of sharing an LPS-1 with the IsoRegen and SU-1.  Perhaps someone can help explain the implications?

 

Tommy,

 

Yes, I should have qualified my closing comment there. I didn't mean to say that there was no degradation at all from sharing vs. separate LPS-1s. Just that the degradation would be greater if powering the mR/IR combo.

[austinpop]

1 minute ago, BigAlMc said:

Will moving the MR back to it's own LPS-1 offset the possible drop in asking the SU-1 & IR to 'play nice' and share an LPS-1.....  Too confusing....

 

Trust thy ears!

[austinpop]

3 hours ago, Superdad said:

 

Hi Rajiv:

 

What you wrote is true.  Powering both a microRendu and an ISO REGEN from a single LPS-1 does defeat the galvanic isolation.  But there are some factors to consider (besides the expense of a second LPS-1; you know I don't  push such things to anyone, you all just keep buying them at a furious rate! ):

 

a) Powering off a single LPS-1 does not diminish the goodness greatness of all the ISO REGEN's other enhancements (Crystek 575 clock, fantastic signal integrity, etc.) which benefit the microRendu > DAC interface in the same way as with computers.

 

b) The microRendu was designed to be very quiet.  There is not much junk coming out on its ground plane, so sharing it to the other side of the ISO REGEN via the isolated and quiet ground of the LPS-1 may not be a big deal.

 

b) Since the LPS-1 blocks the path of leakage current loops between devices (it's like using a battery--only much better!), with a single LPS-1 you are not introducing any new leakage current or path.  What I leading to with this is that if one were to use separate supplies and only one could be an LPS-1, I would put it on the ISO REGEN.  

 

For example, if one used any standard (AC mains attached) LPS or SMPS for the ISO REGEN, while galvanic isolation would still be preserved, the PS powering the ISO REGEN will join the leakage current loops of the DAC/DDC, preamp, etc. downstream (remember, AC leakage current loops are an interaction between AC>DC supplies--LPS-1's don't join those).

 

Yet when using an ISO REGEN and separately powering the microRendu from a standard AC supply (even a noisy SMPS though I don't advocate such), the leakage from that supply really has no place to go.  The microRendu has Ethernet on one side, and a full blocking setup (the ISO REGEN/LPS-1) on the other so no path for the standard supply's leakage to loop with some other.

 

Of course these are all pretty easy comparisons to make and what counts is what you hear!

 

Have a great weekend everyone.  (I'll be spending mine testing 100 each of ISO REGEN and LPS-1 circuit boards; I'll multi-task and catch up on e-mail and get out everyone's $50 ISO REGEN rebates at last!)

 

Best,

--Alex C.

 

Hi Alex,

 

I think (?) we're in agreement. In fact, just in case anyone thought I was trying to dissuade people from sharing an LPS-1 between two components, let me point out I am doing this too - in a different context. Read my post here:

https://www.computeraudiophile.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-the-sms-200-and-microrendu/?do=findComment&comment=668661

 

I just wanted to point out that there are additional considerations when you power two devices across an isolation boundary with the same power supply. We advise the same caution about powering two FMCs with the same power supply, as the isolation you're trying to achieve is being defeated by the common PSU.

 

What I think both you and I are saying is that the audible impact of defeating this isolation may be vanishingly small in some systems.

 

Still, I stand behind my original comments. Use your ears to decide what matters, but be aware of what you're doing when you share a PSU across an isolation boundary.

 

[austinpop]

5 minutes ago, rickca said:

n particular, the upper registers of piano are remarkable.  It's easy to hear very subtle variations in the pianist's touch. Treble is really sparkling and clean when it should be.  

 

Nice, Rick! A ringing endorsement.

 

Without the ringing (as in, harshness, glaze)!

[austinpop]

In my experience - and this is purely empirical:

1. you want the component with the "best" clock closest to the DAC. Which has the better clock - the iFi or the ISO-R? I'm not going to presume to know.
2. you want the isolating supply (LPS-1) after the galvanic isolation

However, realize that these are only starting points. Don't assume anything. Experiment both ways and decide with your ears. Then report back, because it helps build the collective wisdom.

 

 

[austinpop]

25 minutes ago, Superdad said:

Don't let all these microRendu and other streamer users and their complicated chains distract you.  

 

I wonder who he could be talking about? What, me worry?  

[austinpop]

Audio existentialism.

 

If the tree falls, and no one hears it... they must be too busy listening to their ISO-Regen!

[austinpop]

Quick question regarding the red GI switch.

 

ON == GI off

I == GI on?

 

or vice versa?

[austinpop]

7 minutes ago, JohnSwenson said:

The switch shorts the upstream and downstream ground planes. So when it is ON it shorts the planes, which bypasses the galvanic isolation. When it is NOT on, the two ground planes are separate, ie galvanically isolated.

 

John S.

 

Makes engineering sense, John.

 

But you've got to see the humor and confusion when a GI switch set to ON means GI Off!

 

[austinpop]

7 minutes ago, Superdad said:

 

Yeah well it was not very easy to even find a single position vertical mount DIP switch that would fit.  Not stocked anywhere in the USA, I had to buy 1,000 of 'em direct from the manufacturer in China.  And they don't offer them without the word ON printed in the closed position. 

 

1 minute ago, JohnSwenson said:

We had no choice in the matter, that was the only switch we could find that would fit. We spent a huge amount of effort looking for a switch for this. The ON is built in to the switch, we couldn't change it.

 

To counter this we put the letters "ISO" at the switch position where the ISOlation is in effect. We didn't have room to write a paragraph explaining this, we hoped people would understand what ISO meant.

 

John S.

 

Thanks for clarifying, guys! 

 

I do sympathize with your dilemma, and well worth the tradeoff to keep overall cost commendably low.

 

Hmm - didn't realize the switch position says ISO. I'll have to take a closer look. I only saw an I, which I first mistook for 1 (one). 

 

ISO would make it much clearer, I agree.

[austinpop]

1 hour ago, Superdad said:

Great new review of the ISO REGEN is now up at Audiobacon.net.

 

https://audiobacon.net/2017/07/11/uptone-audio-iso-regen-review/

 

Thanks Jay!!

 

Congrats Alex!

 

Since we came to almost identical conclusions in @limniscate's and my joint review of the ISO-Regen with the SOtM Ultra "trifecta" over on this thread, I guess great minds do think alike.

 

 

[austinpop]

4 hours ago, Alexandre said:

Alex, can I use Iso regen with dragonfly black v1.5? The benefits is the same of the others DACs? 

 

I did that as a test in my system, powering the ISO-R with an LPS-1.

 

Works like a charm. 

 

Computer USB port > USPCB > ISO-Regen > DF Black.

 

sounds great too. 

[austinpop]

3 hours ago, thyname said:

Add more cost to an already inflated price? On a dc cable? I will pass

 

You could try making your own, based on John Swenson's excellent info on star quad cables here:

https://www.computeraudiophile.com/forums/topic/31554-diy-dc-power-cables/

 

I first made my own cables (using screw terminals, no soldering) from Canare 4S6 cable, and confirmed it improve SQ over the stock LPS-1 DC cable Alex includes.

 

Then Ghent offered his cables, which are equal or better in SQ to my home brews. They really are very reasonable, when you consider he uses Oyaide connectors.

 

Anyway, the 2 key points I wanted to make:

1. I now have 2 of his straight cables, and one of his Y-cables, and can attest they all sound excellent 

2. Amortize the extra $15ish for DHL faster shipping by ordering all the D.C. cables you think you might need in one batch.

[austinpop]

24 minutes ago, Always.Learning said:

I thought I would post some quick impressions before I send my ISO REGEN back to Alex for a replacement. (I've experienced "ticking" issues with the IR in the system and Alex assures me he will get the problem fixed and I'll have a new IR in a matter of weeks.)

 

I'm probably a little different than most in that I was inserting my IR into a system that already featured an SOtM Ultra stack. Rajiv (aka Austinpop) has done the same thing and reported very favorably on the IR. These are some quick and dirty impressions since I didn't really have the IR in my system very long and did not do a lot of comparisons with the IR in and out of the system. I should be able to report at greater length and about different configurations after I get my new IR and reinsert it into the system (sometime in September after it has burned in). 

 

At any rate, when I did my brief comparisons, I heard this:  When just listening with the SOtM Ultra stack in the system (see below for system details), I am thinking:  what a nice big soundstage, what great retrieval of ambience and depth, what airiness and clarity. Insert the IR, and the music is washing over me, I'm not really thinking, just engaging with the music. Head vs. heart. Also, the IR seems to increase both bass quantity and quality. With the IR in the system, it just feels "right." 

 

Here's the basic digital signal path and analog path for the comparison and impressions described above with the IR in the chain:

 

Late 2012 Mac Mini (with Uptone mods) in "bridged ethernet" mode and running Roon off SD card, powered by JS-2 and with external hard drive (powered by JS-2) on different circuit and in different room than rest of system  > BJC Cat6a ethernet > Paul Pang switch with TXCO clock (in same room as Mac Mini) powered by Jameco LPS > BJC Cat6a ethernet > long run (80 ft) of generic Cat6a ethernet > SOtM dcblCat7 ethernet > D-Link switch modified by SOtM (powered by LPS-1) > SOtM dcblCat7 ethernet > SMS-200 Ultra (powered by LPS-1) (reclocking the modified switch) > generic USB cable (for now) > ISO REGEN (powered by LPS-1) > USPCB > Tx-USB Ultra (powered by LPS-1)  >  Audience USB cable (for now) > Chord DAVE > HiFiMan HE 1000 headphones; or alternatively > High Fidelity Cable CT-1 Enhanced interconnect > Crayon CFA 1.2 integrated amp > Daedalus/WyWires speaker cables > Daedalus Athena speakers

 

Power:  Shunyata and Audience cables and conditioner; dedicated circuits with 10-gauge wire; HFC MC 0.5 Wave Guide

 

Vibration:  extensive use of Stillpoints and bamboo; Finite Elemente rack and isolation feet

 

Room treatment:  Stillpoints Aperture; A/V Roomservice acoustic room treatment; Acustica Applicata Diffusion Absorption Audio Device

 

Wow, that's great to hear! Not the ticking part - that's a bummer - but the fact that your experience adding the ISO-Regen to the SOtM Ultra chain agreed with my listening impressions.

 

You have an amazing system, I can only imagine how good it must be sounding now.

 

Just coincidentally, I had to remove the ISO-R from my chain, due to a glitch with an LPS-1 in my chain that I'm working through with Alex. I am really missing it. I always find that a good way to evaluate the benefit of a component is not just the improvement due to its addition, but after using it for a while, the impact of its removal.

 

Both the addition and removal of the component can tell you about the SQ benefits.

 

I hope you don't have to wait too long for your ISO-R!