ISO Regen performance Improvement Cheap!

[Superdad]

10 minutes ago, Quadman said:

FYI as reported on the audiomart forum different brands of cat 7 cable sound different from one another, go figure.

 

Well since it has been confirmed that these double-pair cheating adapters (they are really not PoE adapters in the traditional sense) don’t have transformers and as applied with this tweak are simply an easy means to try various Ethernet cables as DC cables, then it should come as no surprise that different topology and constrction of CAT cables yield different results.

 

Ultimately what needs to be considered and gotten to the bottom of is which wire construction and topology sounds best and why. 

[Superdad]

1 hour ago, charlesphoto said:

Anyway, expect to see audiophile Cat7/8 dc cables soon.

 

Well give that there is no transformers on the adapters, and assuming that nobody is getting an SQ bump from RJ45 jacks and cheap coax pigtails, it is clear that the entire hubbub is about the wire.  Gauge, shielding, and topology--that's what it always comes down to with cable.  

 

I'm sure you gents saw John S. post last week that he is laying out a small board with a DC-barrel plug and solder pads/holes so that folks will have an alternative to the crapy screw-terminal CCTV DC plugs.  Those will sure coming handy for people who want to experiment with making Ethernet wire-based DC cables.

[Superdad]

4 hours ago, lmitche said:

I would like to recommend that this thread be deleted. Short off that, it should be moved out of the Uptone sponsored section into the general topic section.  This discussion has little to do with the Uptone Audio Regen. I am sure Alex would be delighted with either solution.

 

While the bickering and bruised ego stuff I read last weekend made me sad, I see no reason for this thread to be deleted—there have been a lot of interesting experiments reported on and progress been made.  

 

However, notwithstanding the ISO REGEN in the title, the topic is really more broadly about what constitutes a high performance DC cable—and not just for UpTone products.  So I agree that the whole thread ought to be moved (fully intact) to one of the non-sponsored equipment forums.  @The Computer Audiophile, can this be done in such a way that everyone’s Following status and notifications don’t get messed up?

 

[I would not object to having the thread stay here in the UpTone forum, but I am a person who prefers harmony, and when words get ugly my instinct is to delete offensive posts.  But again, since this thread is not specifically about one of our products, I’d just as soon not have those moderator powers tempting me to clean up a mess. ]

 

Thanks,

—Alex C.

[Superdad]

1 hour ago, jabbr said:

The world is a very bizarre place when a DC power cable between an UpTone LPS1 and ISO Regen affects soundstage!!! Are we still talking about that?. @Superdad I ain’t goin’ to start ABXing pieces of copper wire ???

 

Hey, I didn’t start this!  The DIY tweakers just went there.    We include with the UltraCap supplies a 27-inch, 16awg coax cable that I have made in China 1,000 pieces at a time.  And we offer a 20-inch version of our heavy-duty Oyaide/Belden shielded star-quad (4x18awg).  Star-quad does lower inductance so technically beneficial.

 

But I personally have not had any time to do any DC cable “auditions,” so I make zero claims about any of this.  Maybe now that we are getting production caught up with orders I’ll have an evening or two to actually sit down and listen.  There is a small pile of items and comparisons I am overdue on.  

 

Ciao, 

—Alex C.

 

[Superdad]

21 minutes ago, RamDawg said:

I think you are jabbring about something you have not heard or experienced yet.

 

Well @jabbr does has had an ISO REGEN/UltraCap LPS-1.2 at his place for a few weeks, so it would not be hard for him to experiment with DC cables. (Though mushrooms on a sunny Sunday afternoon does sound like more fun than DC cable swapping. )

[Superdad]

13 minutes ago, jabbr said:

People can smoke the same exact mushroom and have a different experience each time.

 

Smoking them?  Dang, all these years I've been eating them!  

 

BTW, my wife's outstanding Hungarian mushroom soup elicits the same oohs and ahhs from guests every time.  No DBT needed.

[Superdad]

51 minutes ago, RamDawg said:

This troll has perched in our thread and aims to continue to make this his dumping ground.  He has had at least 1 offer to receive for free a DIY ethernet DC cable to try between his LPS 1.2 and ISO Regen (AC confirmed recent purchase).

 

Jonathan (@jabbr) is definitely not a troll.  He is a talented engineer who has shared a great deal here at CA and elsewhere.  He is skeptical but never closed minded.

[Superdad]

12 hours ago, Speedskater said:

Nope, you are writing about a product that has a poorly designed power supply.

 

What product are you referring to?  Surely not the ISO REGEN with its 5(!) LT3042 ultra-low noise/ultra-wideband regulators with properly done capacitor networks and layout.  

 

One aspect we are considering is that while the LT3042s perform better than any other integrated reg on the planet--and are likely not reacting to digital activity current draw spikes--the TI TPS7A4700 we use to regulate a clean 5VBUS output from the ISO REGEN (for DACs that need it) is not quite as immune to such.  We suspect that users whose DACs draw VBUS from the ISO REGEN are more likely to hear differences with lower-inductance DC cables.

Then again, I won't be surprised if a few people pop up and say they are either not using the 5VBUS from the ISO REGEN or are using some other VBUS injection after it--and still hear some cable difference.

 

Remember, what people are playing with with DIY DC cables is not just our UltraCap supplies and ISO REGENs, they are experiencing this with other power supplies and other devices.

 

--Alex C.

[Superdad]

4 hours ago, Ralf11 said:

engineer??

 

Yes as far as I am aware.  Don’t recall what he does professionally these days, but he does have a fair bit of engineering knowledge and experience (and that’s likely a severe understatement).  You can certainly ask him ( @jabbr).

[Superdad]

1 hour ago, jabbr said:

Are you saying that this is the reason it’s a factor with the LPS1.2 supply, because it has a buck converter? Does the leakage go around the solid state relays/bank switching MOSFETs?

 

The output of the UltraCap LPS-1.2 does not have any AC leakage--neither the high-impedance or low-impedance varieties--regardless of if it is charged by the included UpTone-branded 36W SMPS or someone's large LPS.

 

The Mean Well charger that came with the original LPS-1 did not have a "DC>ground shunt" to eliminate high-impedance leakage. Our UltraCap design has always completely blocked the path of traditional low-impedance leakage, but due to the small amount of capacitance of the transistors we use (not MOSFETs BTW, those would be worse) to alternate banks, some amount of high-impedance leakage can pass through.  Of course the easy solution to that was simply to never present any high-impedance leakage to the charging input of our UltraCap supplies.  John's completely effective ground-shunt trick is easy, but of course it made more sense for us to source an charger that already had that built in.  And we offer this new charger for $15 to those who would like it for their original LPS-1. (https://uptoneaudio.com/products/uptone-branded-7-5v-4-8a-36w-smps-with-internal-ground-shunt)

 

 

 

[Superdad]

5 minutes ago, jabbr said:

I ask because this thread is titled in a way that implies we are discussing the ISO Regen (and hence LPS 1.2) and DC cables that affect the sound of this specific pair, rather than potential issues with DC cables on other devices...

 

Yes, well some CA member started this thread (about using Ethernet cabling as a DC cable), and since I always thought it was more generic as a tweak I asked Chris to move it out of the UpTone sponsored forum (also in part because there was some ugly bickering that I did not want to bear responsibility for).  But the topic title remained the same, and the discussion wanders about.  Quite typical. 

[Superdad]

I am SO glad I had Chris move this thread out of the UpTone sponsored area where someone had started it.  Much stumbling about and numerous bizarre and misguided notions. Don’t know where to begin.  You’ll notice that John has steered clear too. 

Sorry guys...

[Superdad]

Or from that cheesy song from the band America"

"Oz never did give nothing to the tin man..."

[Superdad]

17 hours ago, genjamon said:

My first thought when I first swapped to the Kemets was "where'd my dynamics go?!".  Now, I'll acknowledge a certain smoothening of any sibilance in my system, and I can see how those with tinnitus and/or bright systems could benefit from the haze the Kemets added.  In Charles's case, using a LPS-1 and MicroRendu, which I am very familiar with, I wonder if it's not a better synergy.  And maybe the Kemets are providing more benefit over the LPS-1, but the LPS 1.2 may not need the Kemets due to better output regulation.  So, I'm not trying to say the Kemets are trash or anything, and I'm still interested in trying them when I put my LT3045 board in between LPS 1.2 and the UltraRendu.  But at this point, the Kemets are not going back into the system in this configuration - no way, no how.  I'm getting magic with the straight 12awg romex and will be happy to come back to this configuration if results are mixed with other experiments.

 

You are 100% correct and here is why:

Linear Technology LT3042/3045 regulators are completely unlike any other on the planet.  Their very high PSRR and ultra-low impedance across an extremely wide bandwidth mean they are “faster” that any capacitor.  So putting big caps on their output (even low ESR 560uF ones like Cornan is using everywhere) is totally the wrong thing to do.  There is no cap that has a lower impedance across a wide band than what comes right off the 3042/3045.  For any other regulator sure.  But for the LT3042/45, anything more than the required 10uF “stability” cap is technically going to reduce their performance.

 

And the above jives with your report about better results adding a Kemets to an original LPS-1.  That model utilized the 1 amp Ti TPS7A4700 regulator (the LT3045 did not even exist when we were doing the original UltraCap design).
 

[Superdad]

51 minutes ago, Abtr said:

Do you mean a 10uF cap is required for the Iso Regen, i.e., it is already build into it?

 

Not speaking of the ISO REGEN at all.  Just discussing the LT3042/LT3045 regulators--which do require a 10uF cap on their output.  

But yes, we do have an 10uF X7R cap on each of the five LT3042s used in the ISO REGEN--and of course right on the pins of the LT3045s in the UltraCap LPS-1.2. 

 

[Superdad]

46 minutes ago, mansr said:

Nonetheless, larger values of output capacitance do decrease peak output deviations during a load transient.

 

To the detriment of other performance parameters.  Believe what you wish, but how much time have you actually spent designing with the LT3042/45?  

 

Frankly, given the fact that you always post retorts within minutes of each of my posts, I think you just like to follow me around and be contrary to anything I say. ?

[Superdad]

57 minutes ago, sandyk said:

 The short answer is that you can't use a single LT3045 .5A PCB because of the current draw of your sMS-200.

It's specifications are : Power requirement
  Input Voltage : +6.5 ~ 12Vdc
  Power : Max 15W

 This means that with an 8V supply you would need to be able to supply almost 2 Amps.

 

While you are right about any 0.5A boards (though 1A PCBs with a pair of LT3045 are available), I can tell you that both the SOtM sMS-200 and sMS-200Ultra run fine from about 1A. We must have a couple hundred UltraCap users powering one of those two renderers with great results.  They are fine as long as they don't hang a big bus-powered USB hard drive off its extra USB port (whilst doing is discouraged--even by SOtM--for best sonics anyway).

[Superdad]

Moving Cornan's post from the ISO REGEN Listening Impressions thread as it belongs here not there:

 

I just could’nt take it anymore. I removed the copper tape from the dual Supra DAC cables.

 

 

Made a slight adjustment by adding some copper tape at the end of the cables covering the twisted drain wires.

 

 

A releif! The

magic is back! I don’t know if my version is a true JSSG360 or not but as is it is a 100% no go in my setup feeding the LPS-1.2. The semi-shielded Supra DAC is just miles better in terms of naturalness, airiness and music glow. The shielded version was just not alive & kicking like the semi-shielded Supra DAC.