Well, we do not want convince you at all...this is a hobby...and the 1941 is an incredible machine meant for the DIY-community, no ? It is all about sharing impressions with others....and everybody has choices. I am thankful that Peter and Peter shared theirs.
The Mundorfs are settling in...so they got much more transparent, the full body is a litle less full, but still warm, very, very musical...with the help of the Peters and Soren, i would like to get all the ceramics out of the analog side of this DAC.
The Mundorfs are settling in...so they got much more transparent, the full body is a litle less full, but still warm, very, very musical...with the help of the Peters and Soren, i would like to get all the ceramics out of the analog side of this DAC.
Just don't remove any of the small 0402 size ceramics, they're very high speed decoupling and removing those will create signal integrity problems with the shift registers and clocking....
Or on the other hand, do what you want, I'll be happy to sell you a new board when you ruin the one you have....
Thanks for stepping in...that was what I was hoping for before doing stupid stuff...I marked the obvious candidates on the board with different colors...maybe you can give a hint, which colors are no-nos and which are OK to remove...
FETS — imgbb.com
I guess the 0402 you meant are the ones marked in green ?
What about those I marked in red, orange, pink and blue ?
FETS — imgbb.com
I guess the 0402 you meant are the ones marked in green ?
What about those I marked in red, orange, pink and blue ?
Last edited:
Don't ask me what to remove, I designed the board as I believe it should be designed for optimal performance....
I think everybody understands that and it is less meant that "…even Soren recommends to remove them..."...it is more to get some guidance trying stuff as no schematic is published.
But you gave some hints like in theory even the big caps are not necessary (they stay, dont worry), but there are ceramics around the Vref reg anyhow...I guess those are the red ones...if capacitance is needed around the regs, I would like to try alternative or get rid of them.
The small green ones I will not touch as you advised.
But you gave some hints like in theory even the big caps are not necessary (they stay, dont worry), but there are ceramics around the Vref reg anyhow...I guess those are the red ones...if capacitance is needed around the regs, I would like to try alternative or get rid of them.
The small green ones I will not touch as you advised.
I know. I love these guys that think you are smart enough to design the DAC but don't know how to make it perform well. They want to remove all the low inductance decoupling and replace it with snake oil crap electrolytic caps.
Well, we can go even further...just came to this link
randytsuch's audio page: Soekris R2R Dam Dac - Modding
Same lessons learned: Quality of the Vref power supply makes this DAC really sing....and he used LIPo cells for that !! The interesting thing is that Ian Canada has now a very nice Lipo cell PSU design which is sitting ready to use on my desk. I simply would need to know which is the right entry point on the shift registers.
randytsuch's audio page: Soekris R2R Dam Dac - Modding
Same lessons learned: Quality of the Vref power supply makes this DAC really sing....and he used LIPo cells for that !! The interesting thing is that Ian Canada has now a very nice Lipo cell PSU design which is sitting ready to use on my desk. I simply would need to know which is the right entry point on the shift registers.
No one really forces you guys to destroy the pretty power planes. Or do anything sacrilegious to the dac. Especially if you don't actually own one 
It is perfect out of the box. Even the original 1021 was perfect. The original filters were perfect too, followed closely by TNT's filters which alone can make your dac sound like an MSB Select. Or a DCS Vivaldi. As easy as pressing a button.
It is perfect out of the box. Even the original 1021 was perfect. The original filters were perfect too, followed closely by TNT's filters which alone can make your dac sound like an MSB Select. Or a DCS Vivaldi. As easy as pressing a button.
I guess the whole thing has been done a couple of times, just with the 1021, not the 1941 yet...
https://www.diyaudio.com/forums/ven...magnitude-24-bit-384-khz-112.html#post4607903
Maybe Rick joins our discussion...
Analog_SA, yes I find it funny as well...some people give you the impression that it is all about DIN(ot)Y vs. DIY...and I am very thankful to Soren that he gave us this wonderful product to play with...now lets start to play and not have philosophical discussions anymore based on assumptions.
So, if I get it right the shift registers here are VC595A...the printing is quiet small...which would be like this one https://www.mouser.de/datasheet/2/916/74LVC595A-1541851.pdf ...so 3.3V would be perfect for them...
Vcc at pin 16 and gnd at pin 8...will check this...
https://www.diyaudio.com/forums/ven...magnitude-24-bit-384-khz-112.html#post4607903
Maybe Rick joins our discussion...
Analog_SA, yes I find it funny as well...some people give you the impression that it is all about DIN(ot)Y vs. DIY...and I am very thankful to Soren that he gave us this wonderful product to play with...now lets start to play and not have philosophical discussions anymore based on assumptions.
So, if I get it right the shift registers here are VC595A...the printing is quiet small...which would be like this one https://www.mouser.de/datasheet/2/916/74LVC595A-1541851.pdf ...so 3.3V would be perfect for them...
Vcc at pin 16 and gnd at pin 8...will check this...
Last edited:
No one really forces you guys to destroy the pretty power planes. Or do anything sacrilegious to the dac. Especially if you don't actually own one
It is perfect out of the box. Even the original 1021 was perfect. The original filters were perfect too, followed closely by TNT's filters which alone can make your dac sound like an MSB Select. Or a DCS Vivaldi. As easy as pressing a button.
It is a DIY product, and it seems like that some can't keep their hands of their soldering iron, although the idea is that you supply you own power supply and your own output stage.... I would love some feedback on tube output buffers.
You just need to be aware that the vrefs need low impedance from DC to Mhz, and the vref supplies and decoupling have been carefully designed to supply that, using best fit parts. It should be said that I have a strong preference for SMT parts as otherwise I couldn't make them in Denmark at a reasonable price....
You need to be careful about not giving less experiences people the idea that you NEED to rip of all ceramics and add those magic caps, although somebody might want to do that to match their taste in capacitors....
Or you could change the vref voltage, the source for those are D29, which is a LM431, it have a divider, 10K0 - R597 to gnd and 5K23 - R595 to its output.
And as some will say, it's noisy cheap part !!, but it gets filtered....
But to reduce output voltage, the best is a resistor to gnd, the vref voltage have been selected to be best match to shift registers and r-2r resistors.
Ok, I traced the pin 8 and pin 16 with a milliohm meter...they are fed by the orange and pink caps in my little pocture, those caps again are in parallel to the caps we exchanged. So, I guess if you want get rid of ceramic sound, thoses would have to go as well.
This means as well that it should be quiet easy to apply battery Vref I guess...you simply get rid of the eight bigger caps and feed 3.3V in there...but measure the plus and minus orientation upfront, those flip every second cap...as randy stated...
Counting on Soekris if the stuff I am writing is not correct ...
By the way...I dont think that I messed anything up, but I am always a little worried when using a milliohm meter in a electronic circuit...even if it is unpowered...so, is there a way to measure if a shift register does not work properly anymore, e.g. by electrostatic damage or anything ?
It seems both Blitz and myself use valve, transformer coupled preamps. But these are not buffers. Imho, it is a lot harder to design a good sounding unity gain amp using valves, as usually this involves a cathode follower. Of course there is also the option of a step-down transformer at output which brings overall gain to unity, but this is bulky and expensive and probably not what most people want inside a dac.
If i had to attempt a one box dac solution it would be very similar to Lampizator's. For a SE dac this would be a single low mu DHT per channel, choke/transformer loaded with valve/choke rectification. For a balanced dac - pretty much the same, but around a differential amp.
You will hear a blown shift register, trust me
For such damage to occur would be an exceptionally rare event. As long as you don't use that meter to measure the coil resistance of your precious MCs all is well.
I agree, I build a classic differential pair with a ccs in the tail and an amorphous output transformer. Currently I use a pair of 801a, which gives me a larger voltage swing, so that I can use one tube stage less to drive my 300Bs.
But that was the choice for the Buffalo DAC...the Soekris has a much higher output and therefore, the 45 could be a great choice as it brings the whole output impedance down and brings the famous 45 sound.
Overall, I have build many DACs in the last 20 years as well with Borbely fet output stages and very sophisticated PSUs...but tubes sound much more alove and realistic. Fewer parts, better PSUs, no lytics anywhere.
But that was the choice for the Buffalo DAC...the Soekris has a much higher output and therefore, the 45 could be a great choice as it brings the whole output impedance down and brings the famous 45 sound.
Overall, I have build many DACs in the last 20 years as well with Borbely fet output stages and very sophisticated PSUs...but tubes sound much more alove and realistic. Fewer parts, better PSUs, no lytics anywhere.
You will hear a blown shift register, trust me
For such damage to occur would be an exceptionally rare event. As long as you don't use that meter to measure the coil resistance of your precious MCs all is well.
Ok...I thought I saw a little spark when I tried to do contact with my measurement needle....but the DACs plays and I cant hear anything strange...so itwould be more than just loosing in one channel some volume you think ? I have not measured output volume per channel today...yet.