UltraCap™ LPS-1 launch thread!

[gstew]

Greg in Mississippi

[gstew]

haha, yes I suppose it's a strange question but an issue I faced with my VR-LIO. The switching of it's ultra-capacitors is quite loud in a quiet environment but luckily infrequent. With the LPS-1 switching every 4 seconds at 1A, it becomes a strange but important question..

 

I thought I saw an announcement from Vinnie about an upgrade in future units to use quiet switching devices that would be available (for a price) for existing units.... see info here:

 

Announcing new LIO “SS” upgrade (Silent Switching)

 

That is the solution!

 

Greg in Mississippi

[gstew]

For the life of me I can't figure out why or how a power supply would need to "break in."

 

Oh, I don't know... maybe capacitors, resistors, dialectrics in the circuit board and semiconductors?

 

I know it is controversial, but in my experience, all electronics take some time to break in. How long will vary depending on the components involved, how audible on where the unit is in the chain, but all go through that process.

 

Greg in Mississippi

[gstew]

Sensitive, sensitive. I'm not saying that the LPS-1 doesn't break in. I'm merely stating that as a power supply I don't imagine that there will be an audible difference as things settle within the box. A cumulative performance within the LPS-1, itself, sure, but being audible is what doesn't make sense to me.

 

Guilty as charged. I was expecting responses that started with knock-offs phrases from over-educated/under-experienced EE-types such as:

 

That is impossible...

 

What is the mechanism...

 

Show us your measurements...

 

Ok, now that I'm over that, I will go out on a limb and say that for those getting the LPS-1 to power a Regen or a uRendu, I expect the change in sonics due to component break-in will not be a large effect. Why?

 

1. Most of the components that are exposed to the output side are behind great regulator. That doesn't mean the effects of those changes can't be heard through a regulator, but...

 

2. Both the Regen and uRendu ALSO have on-board regulators, in the most critical locations the same unit as used on the output of the LPS-1, if I remember the relevant posts correctly.

 

My experience is that having a minimal set of components before and even smaller set after great regulators feeding other great regulators in those downstream units will minimize the sonic effect of component break-in. I'll be VERY curious to hear what others report.

 

As suggested by John and Alex, I expect the impact of eliminating AC leakage current paths for the powered unit to swamp break-in sonics.

 

But my situation is a bit different than most others here... I'm getting a couple of LPS-1's to try in several locations in my DIY'd digital streams...

 

1. Powering an SDTrans384 SD-card player, with the option to try both on that unit with one powering just the clocks.

 

2. Powering a Kali R-Pi I2S reclocker and the associated DAC, and in situations where only one LPS-1 is needed for the reclocker/DAC combo (some of the DACs I use on the Pi are powered through the Kali reclocker), to also power the LMC optical networking interface upstream of the Pi.

 

3. Using both configured as a +- supply to power a Soekris DAM DAC (that has been shorn of it's active op amp output components).

 

I suspect LPS-1 component break-in might be more audible in these applications.

 

I'm sure I'll find other places to try them over time.

 

Greg in Mississippi

[gstew]

<SNIP>Regards break-in/burn-in. Such terrible sounding words to a manufacture--we don't want anything breaking or burning! How about "seasoning?"

 

But seriously. From my 3-4 decades in audio, including a LOT of time in the building/refining of capacitors, cables, amps, speakers (with my Hovland buddies), and a lots of comparing of passive and active parts, I am certainly not a burn-in denier. I could even rank the rate at which I think parts and materials burn-in/smooth-out. From fastest to slowest (and somewhere once I even had the audacity to put a hour estimation figure next to each):

 

Resistors (less than an hour)

Vacuum tubes

Polypropylene film capacitors in signal path

Analog cables (I can't remember which I thought take longer, speaker or interconnect cables, but any cable with Teflon dielectric will take much longer, and that's usually interconnects)

Teflon film capacitors in signal path (200+ hours)

 

Power supply break-in? I suppose, but I really can't say that I've noticed it much.

 

The UltraCap LPS-1? I certainly have not noticed any change, and I don't know that John has either. Then again, we are always focused on testing and listening, but not so much on break-in. And as Greg pointed out, the numerous (sonic) benefits of our crazy new supply are what people are going to notice. We'll leave it to you folks to decide if the piece changes over time. <SNIP>

 

Our experiences with those types of components are roughly similar. But then there are two classes of components you didn't mention that I have heard break-in, oops, season, electrolytic and ceramic capacitors.

 

In my experience, electrolytics are generally fairly quick to 'season'... 2-3 days or so. That may be entirely different for the Supercapacitors in the LPS-1, but I do remember seeing some electrolytics there too.

 

Of course, there is that special sub-class of electrolytics, Black Gates, of which I still have a good stock and use in some of my DIY builds. They take 2-3 weeks of seasoning AND if they are off for more than a few days, need a few days playing to come back on song. Pain in the B*TT things, if they just didn't work so well in certain situations!

 

Then there are ceramic caps. A long time ago with a computer music setup many galaxies away, I did an extreme modification to the digital portion of a Juli@ sound card which was feeding a DIY'd I2S-connected DAC. I added 100uf ceramic caps across each power feed of each digitial chip on the card (20+ total along with some regulator upgrades). The results were a marked increase in dynamics, bass impact and definition, and overall resolution... and a harshness that took 2 weeks to diminish and another 2 weeks to mostly disappear. Of course some of that could have been me becoming attuned to the new sound... who knows, how to tell?

 

I did do two more rounds of mods on that Juli@, first replacing all of the 100uf ceramic caps with 47uf Oscons, which sweetened the highs and upper mids compared to the first mod, but lost some of the marked dynamics and bass impact. This took less than a week to settle in. Then I went back to the 100uf ceramics, but on only the 'deep digital' chips... PCI interface, controller, microprocessors, etc... and used a combination of the Oscons and stacked pairs of 1uf PPS film SMD caps on the I2S/S/PDIF signal generating chip (AK4114) and on the modified Dexa regulators... that preserved the dynamics and bass impact of the 1st version with close to the sweetness of the 2nd... and again those new 100uf ceramics took 2+2 weeks to finally season.

 

DIY'ing and modifying is a great way to experience and begin to categorize how different parts 'season'... I can highly recommend it!

 

Hi guys (good to see you here Greg!; call me sometime, it's been too long).

<SNIP>

By the way, we are right on schedule (maybe even a day ahead!) for shipping the first 150 units next Friday, October 14th as promised.

<SNIP>

 

Yah, it has been too long... I will ping you for some good times to call, BUT only after I've received a shipment notice of my LPS-1s... don't want to slow them down, I'd have ~148 other fanatic audio enthusiasts VERY ANGRY at me if I did that!

 

Later!

 

Greg in Mississippi

[gstew]

So a related question... If I had a 12v power supply that could provide 3+ amps continously, would there be any reason not to feed 2 LPS-1s with it?

 

Just curious... I'm checking some supplies I have here, not sure ANY will do that, but just in case.

 

THX!

 

Greg in Mississippi

[gstew]

Thank you Superdad. I am also interested in a raspberry pi stack with a Kali reclocker board and piano 2.1 dac

 

https://www.allo.com/sparky/bundle-kali-piano-2-1.html

 

I have been told that Kali needs 150mA , Sparky needs about 2A only on boot , about 1.7A while working.

Sparky is their name for the single board computer running the stack-the raspberry pi equivalent.

 

Do you have any plans for a 'jumbo LPS-1'?

 

Skypickle,

 

I have good experience with using the Kali on a Raspberry Pi to feed several DACs. You'll find you get MUCH better sound quality by using separate supplies as much as possible, with a minimum of 2, 1 for the Pi and 1 for the Kali/DAC combo.

 

While I am currently generally using a DIY'd linear supply to power the R-Pi's in my setups, a good friend suggested this SMPS Adapter as sounding good powering the Pi: QVS 2.5Amp 5.1v Switching Power Supply for Raspberry Pi B with Built-in 4ft Micro-USB Cable ARUSB-2.5A - Micro Center .

 

I have one of these and in a quick check, it sounded pretty good too, though I haven't yet decided to use it instead of my DIY supply. BUT it is in the same realm of SQ and is a lot less expensive than my DIY'd supply's parts alone.

 

That friend discusses this more in this post: Hifiberry DAC+ Pro - HW mods anybody? - Page 10 - diyAudio

 

I sometimes use an LPS-1 to power the Kali and 1 or 2 more to power the DAC (depending on which setup I'm using at the time) and get better SQ out of that setup as opposed to the good DIYd'd linear supplies I generally use there. So I can recommend their use here.

 

But don't try to power it all with 1 supply, whether you use an uprated LPS-1 or something else. 2 supplies is a minimum for acceptable sound from this type of setup, IMHO.

 

Greg in Mississippi

 

P.S. I should add that I consider the Kali or the more expensive, but more flexible and capable Ian FIFO to be a MUST when using an R-Pi to connect to an I2S-input DAC. Done right, that setup can be pretty decent!