Signalyst DSC1

Mshow DSC1 DAC decoder DSD Assembled Compatible with Amanero and our XMOS usb | eBay

Unbalanced, close to MIska's original design but they changed the muting circuit and maybe other small details.

I cannot comment on the transformer question. There are plenty of ways to convert balanced signals to unbalanced - the Twisted Pear Balsie module is simple, effective and provides good output drive for instance, at the expense of more complication.

If you order the DSC Balanced kit let us know how you get on with it.
 
Perhaps noise are coming from the logic gate. I might try to put a ferrite bead after each resistor. There are various beads from Murata. I think it worth a try.
I already did this with a сhinese non-balanced board. The problem of noise has remained.
 

Attachments

  • IMG_0004.jpg
    IMG_0004.jpg
    224.3 KB · Views: 1,124
Last edited:
Some informations interesting to think about some ways to improve electronics from murata site.
chapter01-p1_img0015.ashx


Wire could be considered like a noise source of spatial conduction.
Transformer is also considered like a similar case.

Some feature about sowter transformer :
Mumetal core for minimal harmonic distortion. Internal electrostatic shield for high common mode noise rejection. Mumetal can for magnetic shielding.

i think mumetal shield is an other step to upgrade transformer.

An externally hosted image should be here but it was not working when we last tested it.
 
I have to buy an unbalanced version of DSC1-DAC because the seller was ran out of balanced board, very sad
What I did here is I soldered all components but not IV stage, and take the signal out before IV stage for measurement.

I measured output without playing dsd file; that noise is coming from computer via usb power supply through Amanero board.

mtRPs5x.jpg



I don't think I will finish iv stage of dsc1 dac because it's confirmed that there are noises while playing dsd file. So do you know any iv converter that is good enough for this dac (the signal very small 0,2mVpp with impedance change from 470 ohm to 15K ohm)? Or can I use transformer for isolating high frequency noises, if yes, what is the ratio of that transformer 1:1, 1:5 or other?
Thank you for your patience in reading my issues!
 
I measured output without playing dsd file; that noise is coming from computer via usb power supply through Amanero board.

Power the Amanero with a better power supply and use it with an isolator;

Free Amanero Isolator bare PCB Twin pack - DIYINHK

or even better use an isolator/reclocker like the TP Hermes/Cronus.

Alternatively, there are USB boards that include isolation and reclocking, such as the JLSounds I2SoverUSB product.
 
They use flip flops and different resistor values

Yes, you can use other weightings than unity weighting used on DSC1 to gain different filtering characteristics. But it increases jitter sensitivity to some extent. So it is always trade-off between different attributes. With discrete resistors you can experiement with different weightings. I used resistor networks so it is not possible with my board, reason why I chose network is both size and the properties. Since most of the networks are designed to be used as bus termination networks, they have somewhat favorable switching characteristics. SMD networks would be better than the through-hole ones I used. I just wanted to use as few SMD components as possible.

AFAIK, most DAC chips use unity weighted elements.
 
Do you have any capture of this noise on oscilloscope ?
This noise is not permanent. It's hard to see with an oscilloscope. The noise is similar to the correlation, it appears on quiet levels of music. The higher the DSD frequency, the more noise. I digitized for example a quiet section of music from the DSD256 in wav. https://yadi.sk/d/_AiM2Z3x3R8tiP
 
Yes, you can use other weightings than unity weighting used on DSC1 to gain different filtering characteristics. But it increases jitter sensitivity to some extent. So it is always trade-off between different attributes. With discrete resistors you can experiement with different weightings. I used resistor networks so it is not possible with my board, reason why I chose network is both size and the properties. Since most of the networks are designed to be used as bus termination networks, they have somewhat favorable switching characteristics. SMD networks would be better than the through-hole ones I used. I just wanted to use as few SMD components as possible.

AFAIK, most DAC chips use unity weighted elements.

Yes you are correct WRT unity weighted elements. A good example is Sabre DACs.

I have a question for you Miska - would it not be better to use differential OP flip flops for a balanced design?
I'm wondering if the bulk of non linearities are due to inherent non RTZ nature of DSD stream. As such I would think the fastest dif OP logic would give best results. Something like 16 x Potato 74G74 dual FF should work well.

Otherwise is it possible to integrate RTZ coding?

Terry