Hi!
After removing the output transformers of DSC 2 (Beaglebone version) to install those of Ivan. One of the channels (the right) is not heard, I do not know problems of the IVAN transformer, I have tried both with the left channel and they work correctly. Any idea of where you could measure is to know where the problem is and try to fix it?
Thanks!
After removing the output transformers of DSC 2 (Beaglebone version) to install those of Ivan. One of the channels (the right) is not heard, I do not know problems of the IVAN transformer, I have tried both with the left channel and they work correctly. Any idea of where you could measure is to know where the problem is and try to fix it?
Thanks!
First post in this tread. I'm not an active participator on diyaudio, lurking mostly.
But in reply to the last post, maybe I will have one to sell, I'm still trying with the 2.6.2 version and building a second one.
Thing is, I build the 2.6.2 to the letter last month. Apparently I have been living in a cave. I recently discovered HQPlayer and I am in love with it, using it on my old Merging NADAC. DSD256 with poly sinc gauss long and ASMD7EC modulator. My NADAC never sounded this good!
Seeing the hardware page on the website of Jussi Laako I discovered the discrete DSC1.
I combined the DSC2.6.2 with the original Amanero board, bought at Audiophonics.fr. With the latest firmware 2006be it runs perfectly DSD512. The sound is very nice. But it does not beat my Merging stuff. Not at all. I do not understand why some people say that the DSC beats the ES sabre chips by a large marging. If so, then the sabre has been treated poorly with bad board design and clock. I cannot explain otherwise. Or maybe the superbly implemented Merging Ravenna interface is making the difference, I do not know.
Still, because I like the warmth of the DSC2.6.2 very much, there is a very natural feeling with the sound, I ordered the high end version of Amanero with better power supply and Musetec clocks. I hope that the transparency will notably improve. Because that is mostly lacking when compared with the NADAC.
Will keep you posted.
But in reply to the last post, maybe I will have one to sell, I'm still trying with the 2.6.2 version and building a second one.
Thing is, I build the 2.6.2 to the letter last month. Apparently I have been living in a cave. I recently discovered HQPlayer and I am in love with it, using it on my old Merging NADAC. DSD256 with poly sinc gauss long and ASMD7EC modulator. My NADAC never sounded this good!
Seeing the hardware page on the website of Jussi Laako I discovered the discrete DSC1.
I combined the DSC2.6.2 with the original Amanero board, bought at Audiophonics.fr. With the latest firmware 2006be it runs perfectly DSD512. The sound is very nice. But it does not beat my Merging stuff. Not at all. I do not understand why some people say that the DSC beats the ES sabre chips by a large marging. If so, then the sabre has been treated poorly with bad board design and clock. I cannot explain otherwise. Or maybe the superbly implemented Merging Ravenna interface is making the difference, I do not know.
Still, because I like the warmth of the DSC2.6.2 very much, there is a very natural feeling with the sound, I ordered the high end version of Amanero with better power supply and Musetec clocks. I hope that the transparency will notably improve. Because that is mostly lacking when compared with the NADAC.
Will keep you posted.
Last edited:
I've repaired a small change to the design in the meantime. I play with the balanced output, but it was not connected to ground with pin 1. That gave an unnecessary amount of noise, this galvanic separation between DAC and amplifier. With pin 1 now connected to the ground plane of the PCB, the problem is solved. Strange that this was not included in the 2.6.2 PCB design. Sound has improved!
Today I did quite a few measurements on the distortion, in order to find the best HQPlayer DSD modulator. It is precisely at low signal levels that the less good modulators show many high-frequency harmonics. Not much higher than -120dB, but still, the best modulators don't do this. I found that the best modulator in terms of distortion was HQPlayer AMSDM7 512+fs, both for DSD256 and DSD512. AMSDM7EC 512+fs is even better and with my current NUC it works on DSD256. So I now understand why at the puredsd.ru site the 512+fs modulator is being recommended.
Below is one of the measurements. I also compared to the NADAC. With the NADAC, the 512+fs modulator was good in terms of distortion, but gave too much high-frequency noise. The bulge you see above 50kHz got a lot higher. Must have something to do with the output filter of the NADAC. So for the NADAC, the best modulator on DSD256 is the ASDM7EC.
N.B. the increase in noise above 50kHz is a property of the sound card I am using (Focusrite Scarlett Solo 2nd gen). With nothing on the input (plug pulled) absolutely nothing changes to the yellow-colored noise floor. So the spectrum of my measurement only shows the 333Hz tone plus harmonics.
Today I did quite a few measurements on the distortion, in order to find the best HQPlayer DSD modulator. It is precisely at low signal levels that the less good modulators show many high-frequency harmonics. Not much higher than -120dB, but still, the best modulators don't do this. I found that the best modulator in terms of distortion was HQPlayer AMSDM7 512+fs, both for DSD256 and DSD512. AMSDM7EC 512+fs is even better and with my current NUC it works on DSD256. So I now understand why at the puredsd.ru site the 512+fs modulator is being recommended.
Below is one of the measurements. I also compared to the NADAC. With the NADAC, the 512+fs modulator was good in terms of distortion, but gave too much high-frequency noise. The bulge you see above 50kHz got a lot higher. Must have something to do with the output filter of the NADAC. So for the NADAC, the best modulator on DSD256 is the ASDM7EC.
N.B. the increase in noise above 50kHz is a property of the sound card I am using (Focusrite Scarlett Solo 2nd gen). With nothing on the input (plug pulled) absolutely nothing changes to the yellow-colored noise floor. So the spectrum of my measurement only shows the 333Hz tone plus harmonics.
Last edited:
The high end Amanero Board has arrived in the meantime. This is really a major improvement. So, this board will not be sold. It just sounds too good! Now transparancy, detail and 3D image competes with the NADAC. And what I like is the warm sound. Especially with strings this is very notable. I have open baffle speakers through which the original tone quality of the soundbox of cello and violin is very well preserved. And this pureDSD DAC can highlight this. Very nice!
Yes, I do not think that in your case the high end Amanero will be an improvement.
athomdiy,
Was it easy to make the olo111 Modulator board work?
Who did the soldering and who did the programming?
I need a solution with at least three channels and this is difficult because I am not willing to give up on the DSC2.
Olo111 could be a solution..
Do you know if the delay through the modulator is constant?
So if I use a pcm DAC for the subwoofer clocked by the DSC2 clock, the time difference between the main speakers (olo111-DSC2) and the subwoofer DAC is the same all the time?
Was it easy to make the olo111 Modulator board work?
Who did the soldering and who did the programming?
I need a solution with at least three channels and this is difficult because I am not willing to give up on the DSC2.
Olo111 could be a solution..
Do you know if the delay through the modulator is constant?
So if I use a pcm DAC for the subwoofer clocked by the DSC2 clock, the time difference between the main speakers (olo111-DSC2) and the subwoofer DAC is the same all the time?
Wow really great measurement for the DSC2 - also the NADAC is nice too.
I don't have the skills to build this DSC2 myself - can I buy yours
please send me a PM if possible.
I really have to react on this one. Merging NADAC implementation of Sabre cannot be compared to the Khadas toneboard. Completely different league. Having said this, the DSC2 with high end Amanero and the right power supply has a more natural sound that I like. But I would not like to call Merging NADAC 'artificial'. That does not do it justice. At all.
In the meantime I have been able to compare the high end amanero with Musetec clocks and it's own power supply with the Xing U30 which should be better then the beaglebone and original Amanero according to Pavel. I have to conclude that the Xing sounds very nice but the sound goes into the direction of the NADAC. The high end amanero sounds fuller, warmer. And not a bit less detail, so the glow in the sound doesn't cost you anything.
https://nl.aliexpress.com/item/1005...t_main.37.eb9c79d2I8H0Qk&gatewayAdapt=glo2nld version CCHD-957
https://nl.aliexpress.com/item/1005...t_main.32.eb9c79d2I8H0Qk&gatewayAdapt=glo2nld
https://nl.aliexpress.com/item/1005...t_main.37.eb9c79d2I8H0Qk&gatewayAdapt=glo2nld version CCHD-957
https://nl.aliexpress.com/item/1005...t_main.32.eb9c79d2I8H0Qk&gatewayAdapt=glo2nld