No surprise ;-) I know that he would never get that to work despite the confident attitude.
What surprises me however, is that some reports seem to indicate that a DAM DAC is not immediately trounced by this DAC... and only 3 other persons has heard my last revision of filters which I think really ups the fidelity of the DAMs.... Maybe I release them - I'm thinking about it...
//
No, the DAM is not immediately destroyed at all...its more like two sisters are playing at the moment...I doubt that many people would be able to pick the right one in a blind test....
..., but my Soekris setup played now for years, is heavily modified from PSU to board to filters...
...while Andreas DAC is brandnew, running just 48hours, no mods and crystek clocks...so I would not be surprise if Andreas DAC will surpass the Soekris in two weeks as the PSU and all the regs are brandnew as well...but at the moment not yet.
I will as well exchange different xmos input board and will have a look into the cabling etc...I will for sure give Andrea's DAC all the love and attention it deserves...
...nevertheless both are fed by ultrabibs, so we have some parameters nearly the same...my learning from playing with different diodes and LC filters on the raw side of the Ultrabib has been nevertheless...its all audible. So, giving the best lyrics and best diodes makes sense...LC sound better than plain C-Filter etc...the usual things, but a reg still lets this through...
..., but my Soekris setup played now for years, is heavily modified from PSU to board to filters...
...while Andreas DAC is brandnew, running just 48hours, no mods and crystek clocks...so I would not be surprise if Andreas DAC will surpass the Soekris in two weeks as the PSU and all the regs are brandnew as well...but at the moment not yet.
I will as well exchange different xmos input board and will have a look into the cabling etc...I will for sure give Andrea's DAC all the love and attention it deserves...
...nevertheless both are fed by ultrabibs, so we have some parameters nearly the same...my learning from playing with different diodes and LC filters on the raw side of the Ultrabib has been nevertheless...its all audible. So, giving the best lyrics and best diodes makes sense...LC sound better than plain C-Filter etc...the usual things, but a reg still lets this through...
Ok, I guess I will do tomorrow some voltage tuning...I gave both, Vref and Vdd 5Volts in order to have max. voltage swing from the DAC....and fulfilling the 75% condition....Vdd=>0.75Vref....
On the one hand I feel its a bit too loud now for my system and on the other hand I read another condition here now(was not getting it the first time...to eager to get it going i guess):
"VDD power supply has to be equal or greater than 0.75 x Vref. For example, if Vref is powered by +/- 4.5VDC then VDD has to be powered by at least 3.4 VDC. In this case, even the clean side of the TWSAFB-LT (J18) has to be powered by the same 3.4 VDC."
Well...not sure why the fifo needs the same Vdd-Voltage as the dac as both are only connected over I2S, but if the master says so...this means basically as well vice versa this condition:
if i have a 3.3V at J18 fix (which i have), than my VDD for the DAC should not be 5V, but 3.3VDC as well which means Vref should be 4.4V or smaller. So, I will apply this tomorrow and see how this sounds...
On the one hand I feel its a bit too loud now for my system and on the other hand I read another condition here now(was not getting it the first time...to eager to get it going i guess):
"VDD power supply has to be equal or greater than 0.75 x Vref. For example, if Vref is powered by +/- 4.5VDC then VDD has to be powered by at least 3.4 VDC. In this case, even the clean side of the TWSAFB-LT (J18) has to be powered by the same 3.4 VDC."
Well...not sure why the fifo needs the same Vdd-Voltage as the dac as both are only connected over I2S, but if the master says so...this means basically as well vice versa this condition:
if i have a 3.3V at J18 fix (which i have), than my VDD for the DAC should not be 5V, but 3.3VDC as well which means Vref should be 4.4V or smaller. So, I will apply this tomorrow and see how this sounds...
D
Deleted member 537459
The XMOS implementation of your 1941 board will be hard to emulate. It works using the dac/fifo clock and this avoids having close but dissimilar clock sources all over the board. The clock source may not be the best but it is well distributed.
Thanks for the info...Well, in theory I could actually emulating this using the I2soverusb and Andreas clocks, but not with 5.x/6.x, but with 22/24 as the i2soverusb can use external masterclock...but than i go to less premium crystals (which i got as well here)...not sure if its worth though...
I think its great that Andrea is back ! it makes so much more sense to communicate over the forum so everyone can learn...he is on vacation right now in the US but still answers emails...which is simply outstanding support.
so I share his input on j18 = Vdd condition:
As the Ultrabib in their normal form go only down to 4.7v I cant go to 3.3V to match the 3.3V i got at the J18 (from my diyinhk ultralownoise reg). I asked Salas which parts to change to go furhüther down, but this is not as trivial as just changing a resistor...
So, Andrea responded in the night and confirmed that going up in voltage at J18 to 4.7V is fine, it wont smoke anything. So, the Vdd=J18 condition gets fulfilled (maybe i just connect a second wire to the Vdd+reg of the DAC instead a second reg and see if this works...the ultrabib has enough reserves).
I think its great that Andrea is back ! it makes so much more sense to communicate over the forum so everyone can learn...he is on vacation right now in the US but still answers emails...which is simply outstanding support.
so I share his input on j18 = Vdd condition:
As the Ultrabib in their normal form go only down to 4.7v I cant go to 3.3V to match the 3.3V i got at the J18 (from my diyinhk ultralownoise reg). I asked Salas which parts to change to go furhüther down, but this is not as trivial as just changing a resistor...
So, Andrea responded in the night and confirmed that going up in voltage at J18 to 4.7V is fine, it wont smoke anything. So, the Vdd=J18 condition gets fulfilled (maybe i just connect a second wire to the Vdd+reg of the DAC instead a second reg and see if this works...the ultrabib has enough reserves).
Last edited:
I am not sure...did Merlin talked about a modification of an ultrabib or was this a completely different regulator which was better than Reflektor-D and BiB ?
For reference I have been running at 3.3v from day one with good volume condition....enjoying the reports Blitz
I will experiment with 3.3V for sure as well...found 4 Reflektor-D boards with kits in my shelf...and found as well the comments of Salas what to change to get the ultrabib down to 3.3V...just will take a week or two...have ordered as well some 3045 kits which are configurable and we ll see (ON the Soekris the Ultrabib ate the 3045 for lunch).
and in the meantime everything burns in at 4.8V...I have just given Vdd-connection together with J18 with a cable, so no individual reg for J18 anymore to have J18=Vdd always...
...And ordered today the parts for the battery supply (ouch)...
and in the meantime everything burns in at 4.8V...I have just given Vdd-connection together with J18 with a cable, so no individual reg for J18 anymore to have J18=Vdd always...
...And ordered today the parts for the battery supply (ouch)...
BIB,Reflektor -D,UltraBIB
When you have the Fifo Lite with the Drixo 5-6 MHz Connected to the DAC TWSDAC-LT you would have a set which in my opinion will exceed the Soekris,I think FIFO and TWSDAC-LT DAC are inseparable.
It looks a bit like PPS caps decoupling switches on the dac lite from photo, can anyone confirm? a nice touch if so.
@Blitz , are you using the calibration feature for the DAC?
Even without the clocks it is interesting the soekris DAC holds up so well.
For the ultrabib it would probably help for this app to increase C2 if using the stock value. I also found the UES caps included in the kit are an unusually poor (harsh) sounding cap, thought it could be related to bipolar construction (notice higher leakage spec compared to similar caps).
I have yet to replace them in my bibs, I want to see if a 33u polyester can work for C3 output cap with a small series resistor to get ESR in the desired range, always good to avoid lytics where possible I think.
@Blitz , are you using the calibration feature for the DAC?
Even without the clocks it is interesting the soekris DAC holds up so well.
For the ultrabib it would probably help for this app to increase C2 if using the stock value. I also found the UES caps included in the kit are an unusually poor (harsh) sounding cap, thought it could be related to bipolar construction (notice higher leakage spec compared to similar caps).
I have yet to replace them in my bibs, I want to see if a 33u polyester can work for C3 output cap with a small series resistor to get ESR in the desired range, always good to avoid lytics where possible I think.
Yes, mine is calibrated...
And I replace most of the caps to Nichicon Superthrough which is the best sounding lytic for me...the smaller 50uF is not available, so i am looking there for alternatives as i use silmic there or something else, but not top...for Vref i use lt4320 ideal bridge which sounds more resolving than any normal rectifiers by a margin...
And I replace most of the caps to Nichicon Superthrough which is the best sounding lytic for me...the smaller 50uF is not available, so i am looking there for alternatives as i use silmic there or something else, but not top...for Vref i use lt4320 ideal bridge which sounds more resolving than any normal rectifiers by a margin...
LC-supply, Lt4320, germanium diodes, removed about 100ceramic caps, removed lytics and soldered ultrabib directly in, got rid of computer-caps and brought Mundorf smooth foil lytics for xrossovers in, TNT filters, Directly heated triode tube output stage (801a) etcetc...many of the learning I will give to Andreas DAC as well, but I like to do this step by step, it is like using spices when you cook...its not always the same recipe needed...