UltraCap™ LPS-1 launch thread!

[romaz]

3-days ahead of schedule. My assistant is really earning her bonus!

 

Because of the shear volume, the paperwork, and a glitch in my ShipStation software, only USA-bound units are going out today. I am now going back and forth with their support department (great shipping system, but zero telephone support and that slows down the back and forth to issue resolution).

 

Should begin getting all the internationals out tomorrow.

 

Hope you first recipients enjoy it. Will be interesting to see what questions and comments arise.

 

Thanks and regards,

 

--Alex C.

 

Kudos to you and your assistant, Alex! Because I am in California, hopefully, my unit will arrive tomorrow.

[romaz]

From the LPS-1 User Guide: "it is normal for the case of the LPS-1 to become very warm – perhaps even hot – under full load."

 

Inspired by this, i invested in a pair of heatsinks: 100 x 100 x 25 mm for the LPS-1 and 75 x 46 x 33 mm for the microrendu.

 

Just trying to maintain my cool.

 

[ATTACH=CONFIG]29705[/ATTACH] [ATTACH=CONFIG]29706[/ATTACH]

 

I have a custom solid aluminum heatsink now that fits my mR perfectly. You can have one crafted for you for only a few dollars by the following company to perfectly fit the mR (and LPS-1) and it does a nice job cooling the unit.

 

0.601" **NEW** - HeatsinkUSA

 

For those whose units run very hot, consider using a conductive thermal paste like Artic Silver 5 between the mR (or LPS-1) and the heat sink to improve heat dissipation even more effectively.

 

Arctic Silver Incorporated - Arctic Silver 5

[romaz]

Is heatsinkUSA your company, romaz? And what a shame you did not send the first unit to the first customer ;-)

 

No, it's not my company. I found other companies but this one was the least expensive and they do good work. They also have different models depending on which aesthetic you prefer. As the LPS-1 will be black, it would be nice to get one anodized in black but that will probably increase the price significantly.

[romaz]

Does the LPS-1 have feet attached underneath ?

 

If not, perhaps you could fit 4 self adhesive rubber feet underneath it to improve the air flow around it if you are concerned about it getting too warm when it is used at full power.

John would not design a unit that gets overly warm, as that would decrease it's longevity.

 

I think heat is often the cause of premature death of many electronic components and so simple and inexpensive ways to dissipate heat from a component is never a bad thing. In my case, I also like the way they look.

[romaz]

My LPS-1 arrived earlier today. I'm possibly the first one. I've been listening for several hours straight.

 

First test was to compare the included Meanwell feeder against the iFi 9V/2A, HDPlex and a 9V battery supply as feeders. No difference that I can tell.

 

In my system, with my mR feeding my Chord DAVE which then directly powers both Omega ALNICO speakers and headphones (Abyss, HE-1000, TH-900), I have previously used the iFi 9V, HDPlex 9V and a 9V battery supply to power my mR. iFi and HDPlex sounded equally good. 9V battery supply sounded only slightly better. This suggested to me that leakage current from the iFi or HDPlex were not impacting my SQ as much as I expected.

 

Installing the LPS-1 was extremely easy. I set the output to 7V and it was basically plug and play from there. Even after several hours, it ran warm but never hot to the touch. My mR is now being powered by a 7V supply for the very first time and is running cooler than it ever has, slightly cooler than the LPS-1. Audio gear is plugged into a dedicated 20A line and my Meanwell is plugged into a separate circuit so no worries at all about backwash noise, even if it's minimal.

 

How did my mR sound with the LPS-1 compare against my 9V battery supply? The improvement was immediately apparent. Bass was immediately louder and so I had to turn down my subwoofer a couple of notches and so obviously, this difference was not subtle. Bass is also tighter and better defined. In fact, everything is better defined. Vocals are clearer. There is sparkle in the treble that I feared could be fatiguing but has not been. The presentation for sure is more dimensional. There is more air around voices and instruments. The perception of depth is improved. Perhaps the most impressive improvement is the enhanced dynamic contrasts, especially impressive with microdynamics, which Bob Katz defines as music's rhythmic expression, integrity or bounce. There is a more effortless agility to the sound. Loud to soft and soft to loud is just impressively good. Attack is more incisive which then can stop on a dime. Simply amazingly tight control. I suspect this is the benefit of a low output impedance. As these stark improvements are noted in contrast to a 9V battery supply, once again, at least in my system, leakage current seems to play less of a role and low output impedance is perhaps what is more important.

 

If low output impedance is the largest contributer to the improvement I am seeing, it leaves me to wonder how this could be improved further, if at all. I have on order a shorter 12-inch DC cable made of UP-OCC grade silver (6N purity) and terminated with Oyaide barrell connectors. If this makes a difference, I may consider ditching the barrell connectors at some point and hardwiring LPS-1 to mR using this cable since all connectors affect impedance.

 

How will LPS-1 compare against the best linear power supplies that plug into the wall? John already indicated that the LPS-1 improves upon the JS-2. Next week, my Paul Hynes custom SR7 finally arrives. This unit was ordered 6 months ago (Paul Hynes works slow) and is the finest LPSU Paul knows how to build. Ripple noise is <5uV and output impedance from DC to 100kHz is <3 milliohms. Since John has not yet been able to measure output impedance of the LPS-1, this should give me some reference of just how good it is. This unit incorporates Furutech's newest NCF IEC inlet, annealed silver DC wiring from the transformer onward and including the umbilical. Paul is also using his ultra low impedance SL connector (<1 milliohm) to connect the umbilical to the PSU chassis. This umbilical will also incorporate an Oyaide barrell connector on the mR end. Fuse has been upgraded to the new Audio Magic Beeswax SHD. If the LPS-1 can outperform this PSU, I will be suitably convinced that I cannot do better than the LPS-1 and will happily repurpose my Paul Hynes unit. Based on what I am hearing with the LPS-1, this wil be tough to beat. Nice job, John and Alex!

 

As a second test, I tried the LPS-1 on my TP-Link FMC (receiving module). This module was already being powered by a 5V battery supply so ground noise from leakage current was not a concern. With an Aurender N10, these FMCs made a significant difference. With the mR, it was subtle at best but I kept them in the chain since I already owened them and since they didn't appear to be doing any harm. I didn't expect the LPS-1 to make a difference here but somehow, there is an improvement although no where as large as with the mR. Once again, this can't be a leakage current issue since I had been already using a battery supply. Input impedance should have no bearing here. It could be the LPS-1's ultra low ripple noise compared to the noise generated by my battery supply's internal switching regulator. I have to convince myself that this improvement is real as it is definitely more subtle. If it is real, it's probably not enough of a difference to justify buying another LPS-1 but if the Paul Hynes is somehow better with the mR, then I may repurpose the LPS-1 to FMC duty. Interesting times ahead. All I can say is "finally!"

 

If mR in NAA mode using Roon/HQPlayer improved upon the Aurender N10 by about 20% with my Chord DAVE, I would estimate improvement with LPS-1 to be at least equivalent if not more compared to what mR brings by itself. Just amazing. The mR has taken on new life.

[romaz]

The Regen does not galvanically isolate. Here is what John Swenson had to say about this:

 

http://www.computeraudiophile.com/f6-dac-digital-analog-conversion/universal-serial-bus-industry-standard-cables-connectors-and-communications-protocols-between-computers-and-electronic-devices-isolator-advice-needed-23475/#post401322

 

Alex,

 

I've been thinking about getting a galvanic isolation device for a while. But, I've concluded that, unless I'm missing something (entirely possible if not likely!) my Regen and Paul Pang USB V.2 card together already do this.

 

Uptone's Regen website says, "The USB REGEN's secondary function is that it disconnects the computer's noisy 5 volt bus power coming down the USB cable, and provides clean 5VBUS on its output--for DACs that need it...." To me, that sounds like the 5V lines going into my DAC are isolated from the computer USB power. No use of the term "galvanic isolation" on the website, but it sure sounds like that's what is happening with respect to 5V bus.

 

As for the digital data, the Paul Pang USB card has a "Hand made silver wire audio grade digital output transformer for TCXO." Pang's website is not crystal clear, but I read this to mean that the USB data lines to my DAC are effectively isolated from the computer USB bus.

 

I would appreciate your (and/or) fellow forum member's) take on this.

 

Rich

[romaz]

Don't know that I agree with that assessment. Drives are quite efficient these days and most 2.5" hard drives draw average of 1 watt at idle, with peaks read/write of perhaps 5-6 watts.

The LPS-1 is about 7.5W supply. I have powered a spare 500GB drive from it.

Here's an interesting observation, Alex. I am currently powering my 4TB Samsung EVO SSD with your LPS-1 and it has made a small but nice improvement as far as SQ. For playback, I have absolutely no problems. I was recently forced to reformat this SSD and so I had to transfer more than 2TB of music back onto this SSD from my NAS where I have my collection backed up. In the process, I found that with sustained writes to the SSD, the LPS-1 is unable to adequately power it. This occurs after every 500-1000MB of data transferred forcing me to wait about 15 seconds for the blinking red light on the LPS-1 to turn green again. From there, I can resume transferring data only to have the process interrupted again because of a blinking red light on the LPS-1. Just to be clear, my LPS-1 is not powering anything else and is being fed by the Mean Well supply that came with it.

 

As I look at the Samsung's spec sheet, Samsung claims that their 4TB 950EVO consumes 3.1 watts during reads and 3.6 watts during writes. At 5V, your LPS-1 is supposed to be able to continuously supply 5.5 watts according to your 1.1A spec and so either this SSD is consuming above that number or else the LPS-1 has trouble sustaining 1.1A.

 

Just to be sure, when I power this same SSD with a 5V/2A supply, I have no problems at all with these file transfers.

[romaz]

I looked up the specs for the 4TB Samsung 850 EVO ... @romaz, you said 950 EVO but I'm pretty sure you mean 850 EVO, right?

 

Anyway, the 4TB version has the following specs. The note about actual power consumption is a bit of a mystery.

Active Reads: 3.7 W (max)

Active Writes: 4.4 W (max)

Idle: 0.5 W (max)

 

Note actual power consumption may vary depending on system hardware and configuration

 

 

Yes, my error. I meant 850 EVO.

[romaz]

Okay, well that settles it--his particular SSD is drawing over 1.1A during heavy read/write sessions.

 

Also, I just looked at the Samsung spec page for his 4TB SSD, and it says average draw is 3.6W and max is 5.6W in burst mode. That burst mode consumption is just enough to put it over the LPS-1's limit.

 

Honestly, the vast majority of SSD's consume well under what Roy's 4TB monster does. Heck, I have big spinning hard drives that consume less!

You're absolutely right! It's interesting how Samsung has spec sheets posted in different places that offer different values but on their website, it states a draw of up to 5.6w in burst mode and indeed, this is just beyond the LPS-1's stated limit. This is obviously what is happening.

[romaz]

Romaz, if you change the SSD for a 4 or 5 TB 2.5 inch hard disk you might be surprised with the SQ increase.

This certainly could be true as I haven't tested too many different SSDs against spinning hard drives but in my direct A/B of a Lacie Thunderbolt drive with stock Seagate 2.5in 500GB HD vs an equivalent Lacie Thunderbolt drive but with the stock HD replaced by a 1TB Samsung 850 Pro SSD, I could hear no difference at all which left me with the impression that it didn't matter. The only argument I came up with in going with an SSD over a hard drive was the absence of mechanical noise and the potential for better reliability.