ECdesigns

[realDHT]

10 hours ago, tapatrick said:

Possibly, let’s see how this unfolds. I would first prefer to see multiple reports of listening tests that have come to the same conclusion that I have. It’s a helpful way to get round the problems of reading online claims. - even my own. So far we have 1 or 2 which is not enough to come to a definitive conclusion.
 

The EC combo is remarkable and I hope that this is kept in mind.  

Thanks for for sharing your listening experiments and impressions with the ecd combo. I have a U192 and fractal DAC on order and hope that the delivery time will not be super long. Will post my report when I have it. 

[realDHT]

U192 and fractal DAC arrived 2 days ago  2 weeks after I ordered it, so faster than the 3-4 weeks estimation given in the webstore (I am in Europe). I first let the music server and ecd combo play for a full day without amplification, for some break-in. Then it was time for A/B comparisons with my other DAC, which I believe is a fairly nice ESS9018 implementation with Amanero USB input and transformer output stage, it is described here http://www.audiodesignguide.com/DAC32/index2.html  

 

Source: DIY music server similar to Nenon projects (Ryzen 3700x CPU, Apacer ECC memory, ASUS rog Strix x470 motherboard, Optane 900P for OS, External usb HDD for music, JCAT Femto USB card. Windows Server 2019, Audiophile Optimizer, Process Lasso, JRiver player). All external and internal server components powered with separate linear regulated rails, including CPU and ATX voltages.  

 

5V power to the ECD fractal DAC via an iFi iUSBpower (the iFi was fed 9V from a Salas UltraBib shunt regulator). 

5V power and signal to the U192 via JCAT Femto USBcard (USB card fed with 5V Salas UltraBib shunt regulator),

 

Amplifier: DIY electrostatic headphone amplifier (DACs connected directly to this amplifier, volume via JRiver internal volume DSP)

Headphones: STAX SR-007 mk2 electrostatics. 

 

Impressions:

The ECD combo sounds warmer, smoother, more relaxed, more effortless than my 9018 DAC. I find acoustic instruments to sound more real with the ECD. On most recordings, the difference is not subtle, it is very easy to hear. For example I percieved much more life-like rendering of harpsichord (J. S. Bach: Concertos for Harpsichord and Strings, The English Concert/Trevor Pinnock, Archiv Produktion, 1982) and a huge improvement in percussion, higher frequencies in particular (Eva Cassidy, Still not ready for the good times, Imagine, 2002). 

Voices also sound better with ECD, no harshness and a touch more presence. ECD is more resolving overall.

 

On one recording (Alison Krauss, Live, 2002), I found the difference between both DACs to be much less apparent, and I had difficulties to pick a clear preferrence on that one. Found that a bit strange. The Krauss recording is 24 bit / 88.2 kHz, while the others are redbook format, could be part of the reason maybe? 

 

I will definately keep this one.         

[realDHT]

7 minutes ago, Ben75 said:

Hi realDHT,

 

Great to see you trying the ECD combo and to read your feedbacks.

Something you could certainly improve is to use a shunt resistor for managing the volume (like the SVC 24 from ECD) as it seems that Jriver volume control does not maintain a bit perfect playback.

 

Enjoy the listening! 🙂

Hi Ben75, thanks for the suggestion!

You are correct that I'm not playing bitperfect now. I remember EC designs opinion is that bitperfect is important. Will probably try shunt resistor later to see if I can hear the improvement. 

[realDHT]

Less than a week since I recieved my ecd combo, I already started doing some modifications 😁 (bye-bye warranty). 

 

Since the enclosures are not shielded, except from the aluminium bottom plate, I tested placing them close to my amplifier transformers and could hear very noticable hum/interference pickup. I shielded the rest of the enclosures with 0.2 mm copper foil. This worked wonders, I can no longer provoke any interference pickup from the amp. At the same time I also installed foam vibration damping material under the circuit boards. 

 

   

[realDHT]

13 minutes ago, bodiebill said:

 

Looks neat!

 

I would be interested to know how ECD owners place their devices. Stacking? Weights? Keeping distance between certain devices?

I was thinking to stack my U192+UPL96+SVC24 together using something like sugru for added weight and keep them at some distance from the DA96. Is that a good idea?

 

I place my 2 devices closely side by side with the aluminium plates in contact. This is only because I find they need the same ground, otherwise I get ground loop hum. This only happens with my iFi power supply, not with the standard one from EC. Probably its a good idea to keep some distance from the DA96 as you say. Had not heard about sugru, checked it out and it looks like a very useful product!  

[realDHT]

34 minutes ago, hopkins said:

I also keep the DA96 at a distance from the other units. I think the U192 emits a little more noise than the UPL. Otherwise, if you stack them, I would put the DA96 on top, just below the SVC - perhaps a small aluminum plate on top of each unit would work instead of applying copper shielding inside ? @realDHT would that be beneficial in your opinion ? 

 

Sugru does look like a nice option 👍

I guess a top alu plate could be beneficial. If stacking units, also each bottom plate would shield cross-talk between units, this was probably the designers intent. With my shielding tweak, one needs to be careful so that no piece of foil can fall off and reach circuits.

 

 

       

[realDHT]

20 hours ago, bodiebill said:

Just received the Paul Hynes SR4T so I tried out powering the UPL as per screenshot. Without the (red) SBooster Vbus2, i.e. when the UPL receives power from the NUC, it works fine. However, when adding the Vbus2 and powering the usb hub separately with 5VDC from the SR4T, the Silicon Labs CP2010x device (which is needed for the UPL) is not recognized.

 

I will try whether it works with the U192 later...

 

 

 

Hope you will get it working and looking forward to hear what difference it makes to the U192 sound.

[realDHT]

I have experimented a little with upsampling using the U192/DA96. I played a few redbook recordings and compared the original 44.1 kHz with upsamling in Jriver to 88.2 kHz. I used the SoX resampling algorithm which is supposed to be the best. Since I don't have an attenuator yet, i'm using digital volume control so the chain is not bitperfect anyway. 

 

I found that with the upsampling, much of the magic was gone. It still sounded nice, but not as special as before.              

[realDHT]

I checked the Hifiman He-6 spec and the sensitivity is low at 83.5 dB, but I think the real  problem here is that it has low impedance, 50 Ohm. The DA96 has 375 Ohm output impedance. 600 Ohm headphones might work well directly, but I think EC modified their DA96 to half output impedance when driving 600 Ohm headphones.   

[realDHT]

There have been previous reports in the thread of various power supplies and sources together with the ECD combo. I made some experiments of these things yesterday and I write here my personal impressions:

 

1. Impact of the powersupply to the DA96 DAC

Stock supply VS very clean regulated supply (iFi iUSBpower fed by Salas shuntreg)  

 

2. Impact of the source

old server VS new server

 

Old server is an Intel I3 system with standard quality hardware, using PicoPSU (DC-DC switchmode converter) power, fed from a standard quality Velleman 13.8V LPSU. New server is an AMD Ryzen system with selected hardware (JCAT USB card, Optane drive etc) using dedicated high quality low-noice LPSU rails for all components. Both servers use the same OS/Player/Settings/Software optimizations.

 

Some month ago when I made experiment 2 with my prevous DAC (Amanero/ESS9018 DAC), there was a very big difference in sound quality between the servers, where the old server sounded almost muddy when compared with the new. This time comparing them using the ECD combo, I found most of the difference is gone! Both servers now sound great. Almost disappointing to me, since I spent much money and building effort on the new server :) Oh well, the new server still sounds somewhat better to me, but the difference is now much more subtle. I would say I can hear bigger difference doing up- or downsampling, or even changing the dither algorithm.

 

With experiment 1, there was bigger difference. With the stock power supply, I found the music a little more "grey" and somewhat restricted in dynamics. I would say the improvement using high quality power to the DA96 is worth the effort and extra cost.    

 

Takehome message: Power supply quality seems to matter more for the DA96, than for the U192.

There seems to be some merit to ECDesigns claims of source independence. Not complete source independence for sure, but much more so than my prevous DAC, that is very clear.         

[realDHT]

3 hours ago, bodiebill said:

 

Thanks for the report! Interesting conclusion, as some have claimed the opposite: that it matters more for the U192 than for the DA96. I can not (as yet) judge from my own experience, but hope to try things out before long...

Yup, I did expect the U192 to be more sensitive since the oscillators are in that box. However I could not test upgrading the power to U192 directly (I tried using the iFi iUSBPower, but for some reason it did not work with the U192). My conclusion comes from assuming that the new server provides much cleaner power to the U192 than the old server. 

[realDHT]

21 minutes ago, bodiebill said:

 

Just wondering: how did you test upgrading the power to U192 indirectly?

Because my new server has high quality 5V LPSU that feeds the U192 via USB, and in the old server USB 5V comes from a switchmode picuPSU. Therefore I assume that comparing the different servers is indirectly comparing good vs bad quality USB power. But its not an accurate comparison. It is possible that the PicoPSU 5V is much better than I think, and then the U192 might still be sensitive for a bad power supply. 

[realDHT]

11 hours ago, Huubster said:

The SR4 is on my shortlist now, but I'll start with un-doing my LPS mods first!

 

 

Thanks for the report! 

 

May I ask what were the LPS mods you did that you will undo?

 

Was the U192 the new updated version?

 

 

[realDHT]

2 hours ago, Huubster said:

 

My first mod was removing the internal wiring of the ground cable. So I decoupled the ground wire in the LPS case from the casing itself (My wallmount has no grounding either, so I don't miss it). That was a major step up, I'm pretty sure that modification stays, but I will definitely do a re-check.

 

The latter mod was a very simple one too, and could only have, conform my logic, a positive impact, but I had my doubts from the first notes I heard. I removed the on/off switch from the circuit. Internally AC is coming in, wired to the on/off switch, and from the on/off switch wired to the circuit board. I changed that by routing the wiring from incoming chassis to the circuit board directly. I had to fiddle around with and bow those metal lips on the outer ends of the cabling to make it fit on the circuit board connectors. Maybe they don't make proper contact now, causing the deteriorated sound.. So I was actually thinking this morning to remove those horrible nasty metal lips and solder the contacts instead. I still believe that removing the on/off switch cannot make the sound worse right?!

 

And yes, the U192 is the updated version.

Ok, tnx.

I agree that your bypass of the switch should not make things worse unless a bad contact is introduced. Bypassing of the power switch should not make much difference to sound anyway IMHO. 

      

[realDHT]

16 minutes ago, bodiebill said:

 

Happy to try. Actually (as I believe I reported somewhere earlier in this thread) with the U192 I compared a Ghent Y cable with external clean linear power (6VDC linear converted by LR3045 to 5VDC) to a Lush USB cable without separate PS. Surprisingly, the latter easily won. However, I assume the USB cable topology is less important for the UPL (as in that case it does not carry audio data but just occasional control signals) so I should definitely try the same again but with the UPL. A bit short on time these days, but I will report back once done...

If you have the possibility to raise the input voltage to 7VDC or similar, the LR3045 will improve in performance so could perhaps make an audible improvement.  

  

[realDHT]

6 hours ago, seeteeyou said:

so far DA96ETF didn't seem to be too sensitive to its source of power 

 

I do not agree with this however. In my system there was quite a big improvement when I exchanged the stock DAC supply to a very good one. A friend of mine experienced the same in his system. There has also been some other earlier in the thread that experienced the same.

 

 

[realDHT]

I for sure agree with that the ECD combo sounds great already with stock power supplys. And sadly some may be scared off by reading that so many are doing changes in power supplys etc, thinking there must be something wrong in the first place.. I remember seeing a discussion about the Taiko Extreme music server where someone doubted the qualities of the Extreme since so many seemed to start tweaking their system even more after buying it :)   

[realDHT]

20 minutes ago, Jacob said:

I think this is totaly not defenetive - every one has a different quality power lines , so what good for some does not has to be the same for others.

Agreed, like with all other listening impressions, there are many variables involved.

[realDHT]

11 hours ago, mevdinc said:

Wow, 'better than any active loudspeakers' is a huge claim, to say the least.
Obviously, you haven't heard any of the top range active ATCs. I have heard both the Kii Three and Dutch and Dutch 8C side by side and preferred the sound of 8Cs. Neither can get anywhere near my active ATCs and not surprising as my speakers cost around £65K.

I am sure Fractal DAC is quite special as almost all the user experiences in this very topic indicate. And it seems PowerDAC will be very special too, but I would have to hear it myself. It will be very difficult to find a matching passive speaker to mate with it to beat most top range active speakers. If I am not mistaken, that's why ECdesigns is/will be providing a speaker system to go with it.

In the meantime, I will see if I can find a way to try out the Fractal DAC with my system.

Happy listening.

'Better' is always very subjective and in the end comes down to personal preference and system matching. 

I agree that active speakers is a very good solution, the best for many (I use active dipole speakers myself). 

On the other hand, I would guess the powerdac will be spectacular with for example the best single element Lowther speakers..

 

 

 

[realDHT]

6 hours ago, Norton said:

Thanks.  Yes, I’ve never understood why the DAC96 needs a different version of a passive vol control.

 

It is because the DA96 has a 375 ohm resistive output. Usually attenuators contain a high value series resistor which would then add to the DA96 output impedance, making it way too high and affecting the sound. The SVC is a shunt attenuator, so it has near zero series resistance, the attenuator resistors only shunt to ground. At your selected volume setting, the DA96 output resistor will form a voltage divider with the selected shunt resistor in the attenuator. 

[realDHT]

I made an usb cable for external power to the U192 this evening. The cable also made it easy to measure the current consumption. The U192 draws around 130 mA playing 24/96 music, wanted to know since extensive modifications are planned 😁   

[realDHT]

9 hours ago, hopkins said:

The FractalDAC has not generated that much interest, outside of our small fan club here, which is a shame. There is not a single professional review of their products outside of the hifi-advice review of the Mosaic UV dating back a few years. Obviously, the ever changing product line has not helped, nor has the recent EU-only shipping policy.

 

One would think some people in the audio business would still be curious to find out more, but it seems most are content peddling lesser (and often more expensive) brands, as a result of having editorial policies largely driven by advertising revenue. 

 

There's really not that much innovation in audio today, with a lot of products offering pretty much the same features/quality. 

 

 

Very true, but I am not sure if EC Designs actually wants a lot of attention and reviews etc. 

 

I remember reading from the reviewer in one of those rare reviews published, that ECD were not that interested to provide the new products for review , although the reviewer asked for it (and the reviewer was very positive in the review). The units look completely hand made, the soldering and everything, at least the ones I have. So with a much larger exposure and customer group, production would have to be automated/outsourced, so would change the company more than they might want. 

 

I guess this would also enforce longer product cycles, which could be a better or worse for customers depending on your viewpoint.    

[realDHT]

3 hours ago, hopkins said:

 

They actually were interested in getting professional reviews, but I think there were some issues in that case with a prototype version being passed on to a reviewer when it should not have. 

Aha, then we are doing a contribution by keeping this discussion thread alive and kicking. I had never heard about EC Designs before seing this thread some months ago, and now I am a dedicated fan 😄 

[realDHT]

I was going to try external clean power to my U192 and compare the difference to using the default usb bus power. Also moving the same 5V PSU between the U192 and the DA96 to get an impression which unit would benefit the most.

 

My experiment ended up badly straight away, since I accidentally took the wrong cable from my multirail LPSU and put in 9V instead of the correct 5V. I realized this when the U192 died some minute later :( 

It turned out the internal regulators on the circuit board were fried, but luckily (or probably due to good design) no other parts. 

 

The good part of the story is that I managed to resurrect the U192 removing the fried components and replacing them with external regulators I had. So now I have stationary external power to the U192 and it sounds very good :) However it is not pretty or practical and I will have to redo it to a nicer more permanent solution. So I could not do the A/B comparison i Intended, but I do think it sounds even better now than it did with the computer bus power.

 

WIth this new setup I did do a new A/B comparison for the DA96 PSU between stock PSU and a Salas shuntregulated supply. Now I did not think the difference was that big, a little more relaxed and smoother sound with the shuntregulation, but not a big difference. 

[realDHT]

8 hours ago, tims said:

I have the UPL combo'ed up with the DA96 DAC and I was wondering how many of you have the U192 as well as the UPL. 

 

To me the fractal DAC is a must have but those of you with both the  UPL and the U192;  as the DAC only has one input do you have a second DAC to run both without having to change ElectroTos cable each time you want to use each of them or perhaps use some other configuration?

 

It's too bad that the DA96 only has one ElectroTos input but maybe an extra input would degrade the SQ in some way?  

 

I have only the U192, but perhaps for you who use both it could work to use a RCA splitter cable? Just an idea, you would need to confirm with EC Designs that it is safe..