ECdesigns

[tapatrick]

15 hours ago, realDHT said:

Stay tuned 😁

Also curious to hear about your mods to the PD and if they make any significant difference.

 

I had lengthy explanations from John about how he has implemented clean power and uncomplicated reclocking in the PD (with a single 200mhz clock) resulting in significantly reduced jitter. Given how much research he has done I'm not sure this can be improved much.

 

Andrea Mori doesn't supply a 200mhz clock as far as I know. (I have a 24mhz clock from Andrea gathering dust, now unused which I was going to try in the Fractal DAC).

[matthias]

35 minutes ago, tapatrick said:

with a single 200mhz clock

 

Please, can you shed some light on why such a HF clock is needed in the PowerDAC?

Thanks

 

Matt

[realDHT]

9 hours ago, yogibear said:

 Same here with my high dB FRs. Plan to add woofers which will be 103 dB, so will be very scary. But am not so sure / confident to mod PD.

I only think it is a real problem in very unusual DSP/ASP compensated setups like mine, but can be somewhat scary/annoying I agree

 

 Although I do not think John would recommend DSP volume, I have not really noticed any SQ degradation using 64 bit volume control in JRiver/HQplayer etc, so in my opinion, PD set for full volume and using 64 bit software volume is a viable alternative too.

 

BTW I also like the combination of good tube amps and high sensitivity FR drivers, I can imagine those go well with the ECD dacs :)  

 

 

[Qhwoeprktiyns]

12 minutes ago, matthias said:

 

Please, can you shed some light on why such a HF clock is needed in the PowerDAC?

Thanks

 

Matt

 

My understanding is that the precision of the master clock in the powerDAC is sufficient. The critical aspect, as you can see from John's explanations, is how the clock signals are used! I would not start obsessing about this.

 

If @realDHT, for example, has the skills and desire to try another clock he will be able to give his opinion after. 

[realDHT]

31 minutes ago, tapatrick said:

Also curious to hear about your mods to the PD and if they make any significant difference.

 

I had lengthy explanations from John about how he has implemented clean power and uncomplicated reclocking in the PD (with a single 200mhz clock) resulting in significantly reduced jitter. Given how much research he has done I'm not sure this can be improved much.

 

Andrea Mori doesn't supply a 200mhz clock as far as I know. (I have a 24mhz clock from Andrea gathering dust, now unused which I was going to try in the Fractal DAC).

Yes it is indeed not easy to improve upon something that such designers like the ECD brothers have made :) Although it should not be impossible when removing the cost constraints that companys always have. Actually the oscillator in the PD is 24 MHz, I was also surprised that I got opportunity to use my leftover 24 MHz Driscoll. Yep I will write more about what I did and some listeing impressions.

[tapatrick]

15 minutes ago, matthias said:

 

Please, can you shed some light on why such a HF clock is needed in the PowerDAC?

Thanks

 

Matt

 

[tapatrick]

9 minutes ago, hopkins said:

I would not start obsessing about this.

Heavens! no one around here would do that :)

[yogibear]

1 hour ago, realDHT said:

I have not really noticed any SQ degradation using 64 bit volume control in JRiver/HQplayer etc, so in my opinion, PD set for full volume and using 64 bit software volume is a viable alternative too.

 

BTW I also like the combination of good tube amps and high sensitivity FR drivers, I can imagine those go well with the ECD dacs :)  

 

 

 

Actually tried something very similar today. JRiver 64 bit acts as a virtual sound card on my laptop. PD set to full volume at 9. Tube amp volume half way. SQ is another level now. Staggering I would say......

 

Music Volume controlled through my phone at the software level on the laptop.

[matthias]

2 hours ago, yogibear said:

Actually tried something very similar today. JRiver 64 bit acts as a virtual sound card on my laptop. PD set to full volume at 9. Tube amp volume half way. SQ is another level now. Staggering I would say......

 

So DVC of JRiver is superior to VC of the PD?

 

Matt

[yogibear]

Just now, matthias said:

 

So DVC of JRiver is superior to VC of the PD?

 

Matt

No, it would not be fair to conclude. I guess it should depend on the input sensitivity and the gain / power of the amp used. 

[realDHT]

The Powerdac-r consists of 2 circuit boards. The lower board is the DAPI, then the DAC board is stacked on top of it with connectors. I did not touch the DAC board, only the DAPI board and externals (enclosure, added attenuator, added masterclock etc) was modified This is how the original DAPI board looks.  

[realDHT]

Then desoldering power supply components (capacitors, 3.3V LDO regulator) to prepare for separate shuntregulators for 5V and 3.3V.

 

Then desoldering the onboard oscillator to prepare for a small squarer board instead.

 

Then, since I will change from a vertical to a horisontal enclosure, I needed to rotate the toslink reciever module 90 degrees and mounted it underside the board instead.

 

Also removed the volume/display module since I will use my shunt attenuator instead. 

 

 

[realDHT]

Very short wires (5 mm) soldered to the GND and CLK signal pads, that will go into the squarer board and hold it in place.  

[realDHT]

The squarer board (by A. Mori) soldered to the DAPI board.

 

The 50 ohm ufl-SMA coax (ca 10 cm lenght) will link the squarer board to the Driscoll sinewave oscillator. 

[Qhwoeprktiyns]

How are you going to compare your changes? Do you have another "stock" powerDAC? 

[realDHT]

The external sine oscillator (from A. Mori) in its shield box.

 

I vibration-isolated the crystal from the Driscoll board using a 'sandwich' of memory foam  and small copper plate. Electrically connected with thin wires 

 

Also sandwiched the driscoll board in the shieldbox between memory foam layers below and above the board (mechanical and acoustic damping).

 

The power wires going out from the side wall will be fed with a 17 VDC salas shunt regulator. 

 

     

 

[realDHT]

Everything except for the top DAC board is in place now. 

 

I could re-use the same enclosure, same shunt attenuator and most of the regulators from the fractal mod project. Just had to re-position everything and add an extra r-core transformer (for the external clock).

 

 

 

     

 

[realDHT]

Even the copper shield between the stacked boards that I made for the fractal mod, fitted perfectly so I could re-use that one too.

 

Perhaps a bit obsessive 😁 but the idea is to reduce potential crosstalk between the boards, and add some weight on the memory foam damping feet that the board stack is resting on.

 

 

[realDHT]

Finally the DAC board is added and the attenuator and back panel RCA contacts connected. 

 

The output impedance of the PD is much lower than the fractal DAC (approx 30R compared to approx 300R), so the attenuator that fitted perfect for the fractal, will now reach high volume much earlier, but it still works well, I can use at least 15 steps until I reach too high volume in my system (total range is 48 steps).

 

 

[realDHT]

30 minutes ago, hopkins said:

How are you going to compare your changes? Do you have another "stock" powerDAC? 

 

Friend and forum member @rowli has an stock PDAC so we have made some listening comparisons, and will do more at some time ahead.  

[Qhwoeprktiyns]

5 minutes ago, realDHT said:

 

Friend and forum member @rowli has an stock PDAC so we have made some listening comparisons, and will do more at some time ahead.  

 

Perhaps you could do together some blind tests on source immunity as well? It would be nice to get some additional opinions on this topic. 

[realDHT]

Since I did not have the patience to to the mods one at a time and compare each, I could not really say if differences is due to the clockmod or the powersupply mods. Since rowli lives in a city quite far from mine we don't have possibility to meet and listen more than a few times per year.

 

To summarize, the main mods:

 

24 MHz onboard oscillator exchanged to very low phase-noice 24MHz Driscoll sine oscillator + squarer board

 

4 separate low-noise, low output impedance power supplies (Salas shuntregulators, separate tranformer windings, all supplies choke input configuration, SiC diodes, high quality capacitors) for: 

* Squarer board (3.3VDC)   

* Sine oscillator (17VDC)

* PDAC supply 3.3 VDC

* PDAC supply 5VDC

 

 

[realDHT]

13 minutes ago, hopkins said:

 

Perhaps you could do together some blind tests on source immunity as well? It would be nice to get some additional opinions on this topic. 

Yep we did not test that specifically so far, but my feeling/subjective impression is that source immunity is impressive. At Rowli's place we used a standard office-style PC as source and it still sounded spectacular..   

 

I'm interested in this but can't test it at home since (as far as I know) I have only 1 toslink-capable source (my music server).

I would need to get a usb-toslink converter or similar to be able to test.  

[realDHT]

So, on visit in Rowli's town, I brought my mod-PDAC to do some listening comparison with the stock-PDAC in rowli's system. 

 

The system ( @rowlican correct me if I'm wrong):

 

Speakers: Linkwitz Lx521 4-way active open-baffle with ASP

 

Amplifiers:

6 channels NCore NC400 for Bass, Lower midrange, Tweeter

2 channels Single-ended 300B tube amps for upper midrange

 

Source: Standard PC located far from rest of system, using 10m toslink cable

HQPlayer, using software volume for volume matching (both PDACs set at max hardware volume)

 

Good room treated with diffusors etc.  

 

 

Comparisons were somewhat complicated by the fact that switching between the DACs took longer time than usual. 

 

I tried a few rounds of blind testing for fun where Rowli did not tell which DAC he connected. 

 

My impressions and first opinions:

mod-PDAC: Somewhat more clear and open with a little higher resolution. I perceived for example cymbals stand out more and had a more real "metal sound" (even if the stock PDAC and fractal DAC also excels in this compared with other DACs I have had). However also a little less warm sounding with female voices on the tracks we listened to.

 

Stock-PDAC: More warmth and "body" on the female voices which I really liked, but not as well resolved and realistic cymbals. 

 

I didn't hear differences in soundstaging etc. 

 

As I understood it, Rowli sort of had the opposite impression, that my mod-DAC was the warmer sounding :) he might want to comment on this. All in all, I believe there is audible differences but I would need more comparison time to define what differs, and do more blind testing. Rowli's speaker system is awsome but my headphone system has IMO even higher resolution that might showcase the differences more clearly. We will see.   

 

     

 

[matthias]

@realDHT

The standard PowerDAC-R has a real big heatsink, however your modification seems not to need one, at least not such a big one.

 

Please, can you shed some light on this?

 

Thanks

 

Matt