TI PCM1792 vs AnalogDevices AD1955.

Status
Not open for further replies.
Jump to Latest
#21
I don't have a good capacity to analize the sound result.

I use for this porpose my friends with lot of experiences.

The value given at the end of the webpage is a general value of all the parameters.

Give some description of the value given is not a very utility.
 
#22
What's interesting is highend DAC makers are using the AD1955 like the Lavry Black DA10. Lavry did a lot of analysis of all the chips and picked the AD1955. He has a lot of info on his website forum pages.

Few Dac makers seem to be using the TI/BB 179x chips.

Just a thought to consider.

-David
 
#23
CD Players and DAC's that use PCM179X (not including mine, of course):

SimAudio Moon Supernova (1798)
PS Audio Digital Link III (1798)
RAKK DAC (1794)
Channel Islands VDA-2 (1794)
Atoll CD 100 Mk3 (1794)
Bel Canto DAC3 (1792)

Well, that's what I came up with in a few minutes. Pretty good sampling though.
 
#32
analog_sa said:
I'm just saying not particularly suitable as it's difficult for a tube to pretend to be a dead short.

If a dac chip feels happy with a few hundred ohms load, great.
Click to expand...


Very high transconductance triodes and pentodes actually work reasonably well for this application when used in grounded grid connection. Tubes I would recommend would include the 5842/417A, (triode) or 7788, 6688, D3A, and C3A or G pentodes in triode connection. I would recommend the use of a bias servo or some similar means to get the operating point just right. (effective input resistance can be well below 50 ohms depending on tube chosen.)

You can also use a very small value resistor for I/V conversion (I used 10 ohms) and follow it with any of these types in a common cathode configuration for very low noise and high gain.

I've tried both ground grid and common cathode as well as designed & built a tube based op-amp for this purpose. This didn't work that well, but it did work. (Two cascade 12AX7A for gain driving a 12AU7A CF) The problem was insufficient output current with acceptable linearity for the feedback resistor value I wanted to use.)

Currently I am using THS4131 monolithics for dual differential I/V conversion into 500 ohm line to line transformers for balanced to unbalanced (or balanced) outputs and IMO this is the best performance sonically I have yet attained. Rest of system is all tube.

Previous efforts sounded very nice, but just a little too "tubey." (Dac 1 and Sony CDP790 both with resistive I/V into common cathode stages.)
 
#33
Interesting setup you have there, Kevin.
IVY is modified to match the impedance of the transformer, right?
which types are you using?

I finalized the CODs yesterday, right now are connected to USB source with passive conversion.. quite good the result, but I expect an improvement by using an active conversion.. hope to complete the IVYs and supply soon 🙂

Vale
 
#35
Thanks for the link, any clue about competitor A and B names?


ess9008 is strongly acclaimed as top DAC, no doubt about it.
Surely I'll get myself a buffalo dac (sabre based) when available again on twisted pear's web site.

cia
Vale
 
#36
Valeriano said:
Interesting setup you have there, Kevin.
IVY is modified to match the impedance of the transformer, right?
which types are you using?

I finalized the CODs yesterday, right now are connected to USB source with passive conversion.. quite good the result, but I expect an improvement by using an active conversion.. hope to complete the IVYs and supply soon 🙂

Vale
Click to expand...

Hi Vale,
The THS4131 will drive 50 ohms or less directly so I have just connected the outputs directly. I had originally thought to use some small buffer resistors, but these are not necessary. No misbehavior visible on the scope. The 4131 is configured to null common mode dc errors so effectively there is no common mode dc error or dc present on their outputs. This gives me a very simple signal path, just a single THS4131, two very closely matched resistors (feedback resistors matched to >0.025%) and transformer per channel, and best of all it is easily configured for balanced and unbalanced outputs and provides full galvanic isolation between the dac and the line stage.. (No ground loops are possible.) It also sounds a lot better than the LM4562 balanced to unbalanced output stage which itself is not shabby, but perhaps not as good as the 4131..
 
#37
Hi Kevin

it looks good.
Basically you are 'only' using the THS for I/V conversion on the IVY, and the output directly attached to the transformer (one per channel), which also performs bal to unbal conversion.

may I ask you which type of transformer is used for this application?

thanks for the infos.
Ciao
Vale
 
Status
Not open for further replies.
Top Bottom