Hey,
OT:HiFi-friend Bardis had a Monica TDA1545 NOS Dac he wasn´t to impressed of. So I redesigned the DAC part with advices from T.
Anyway, Andy Evans mentioned the small DHT double triode 3A5/DCC90 some years ago. So I decided to try it when designing a buffer for Bardis redesigned NOS-Dac. As he has a lowZ input preamp a hybrid configuration was chosen. A Slagle bifilar transformer would be great but then Zout would be in the ballpark of 6kohm and loaded with at least 20kohm. The heater CCS should later be exchange for discrete components.
There where no sign of microphony. The only thing made was that the tubeholders where mounted on a metal plate that then was mounted to the chassis with extremely soft silicone cushions.
The DAC was shown and demonstrated at Hififorum´s DIY-show in Gothenburg some week ago. Great success!
Pictures: Schematic, inside view of Bardis soon to be ready DAC, the readybuilt DAC playing at the show in Gothenburg.
OT:HiFi-friend Bardis had a Monica TDA1545 NOS Dac he wasn´t to impressed of. So I redesigned the DAC part with advices from T.
Anyway, Andy Evans mentioned the small DHT double triode 3A5/DCC90 some years ago. So I decided to try it when designing a buffer for Bardis redesigned NOS-Dac. As he has a lowZ input preamp a hybrid configuration was chosen. A Slagle bifilar transformer would be great but then Zout would be in the ballpark of 6kohm and loaded with at least 20kohm. The heater CCS should later be exchange for discrete components.
There where no sign of microphony. The only thing made was that the tubeholders where mounted on a metal plate that then was mounted to the chassis with extremely soft silicone cushions.
The DAC was shown and demonstrated at Hififorum´s DIY-show in Gothenburg some week ago. Great success!
Pictures: Schematic, inside view of Bardis soon to be ready DAC, the readybuilt DAC playing at the show in Gothenburg.
Attachments
![BARDIS3A52[1].png](/community/data/attachments/152/152612-30bd11b8c7fc1f33c7837b56c269d1c7.jpg?hash=ML0RuMf8Hz)
For those who can accept 5kohm Zout this will work. Andy Evans uses 126C bifilar with good results together with 26 that has a Ri that is double that of 3A5.
This one I haven´t tried though.
Dave Slagle can surely build a bifilar that will be even better. Somewhat higher inductance should then be asked for.
This one I haven´t tried though.
Dave Slagle can surely build a bifilar that will be even better. Somewhat higher inductance should then be asked for.
Attachments
Clever design, and cool implementation. I'll bet it sounds good..
One of these days I will have to investigate NOS dacs.. (Long ago I heard a 47 LABS dac in conjunction with their Shigaraki - I remember it was rather nice actually.)
One of these days I will have to investigate NOS dacs.. (Long ago I heard a 47 LABS dac in conjunction with their Shigaraki - I remember it was rather nice actually.)
Last edited:
Thanks Kevin,
Have to admit I haven´t listened to it yet, so I rely on the impressions from other visitors to the show, as I didn´t make it there. Will try to visit Bardis to see if any modifications can be done later this summer. There is probaly room for improvments in the heater and bias area.
Tubes are good for DACs as they easily deals with the +2,5 offset.
Have to admit I haven´t listened to it yet, so I rely on the impressions from other visitors to the show, as I didn´t make it there. Will try to visit Bardis to see if any modifications can be done later this summer. There is probaly room for improvments in the heater and bias area.
Tubes are good for DACs as they easily deals with the +2,5 offset.
Reviving this old thread as I have a TDA1541a and some 3a5's - wondering how the choice of 100 ohms into the tubes was derived.
R7 is Riv in this case?
R7 is Riv in this case?
