Separate DACs for each halfwave ?

Status
Not open for further replies.
Jump to Latest
Bernhard

Bernhard

Member
Joined 2002
#1
Hi,

is there a DIY project with separate DACs for each halfwave ?

My Technics SL-P770 got that IMHO.

2 x PCM56 per channel, each fed from a different data line...

Noise floor & distortions with -60dB signal are about 6 dB better compared to any of my Philips players. :cannotbe:

Also it claims to be a 18bit DAC while PCM56 is 16bit.

2 bits won from splitting ?

Any chance to get schematic of this player ?


Bernhard
 
#2
what about midscale error? PCM56 says:

Initial Bipolar Zero Error (Bit 1 “on” and all other bits “off”)
is the deviation from 0V out and is factory-trimmed to
typically ±30mV at +25°C.
Click to expand...

so we take one DAC with worst case +30mV at midscale, and one with -30mV at midscale. This amount of crossover distortion has the big advantage we can go class c in the power amp without doing significant more degrading the THD.

You are sure the 2 DACs ar not in balanced or parallel confiduration?
 
#4
Bernhard said:
4 different data lines coming from a Technics >40pin smd ic.

Outperforms 1541 S1, 6 dB better distortion & noise.
Click to expand...

Very typical of you, Bernhard.
Don't take me wrong but LISTEN to the TDA1541 S1.
Whatever it measures, the TDA1541 (in all its versions) gives the best sound I've heard coming from a CD IMHO.
Put a good clock on both players, a good analog stage and compare.
:angel:
 
#6
Just working through this. Minimum FSV for the PCM56 is +5 volts. Say you level shift one of them from -5 to 0 and the other sits from 0 to 5. With a bipolar output stage to recombine the outputs, you could get effectively 17-bit @ +10 volt resolution. It's not 32 bits resolution. MSB voltage value = 5 volts + 16 bits resolution left to resolve 5 volts... right back where you started.
I don't think you would gain any higher resolution from this topology unless you fix the output voltage requirements for each scenario.
In addition, unless you're creating an A/D -> D/A processing pair, you'll have to find a way to drive a 32-bit signal digital signal.
--
Danny
 
#7
Bernhard, 1541 is technically enormously obsolete part and if you are in search for good distortion specs do not waste your time with it. Almost any recent unit is better. Did you know, there are even chips like this (see below)? And you can have that for the price of ‘41/S1.

Pedja
 

Attachments

  • -60.gif
    -60.gif
    13.3 KB · Views: 1,356
#8
Pedja said:
Bernhard, 1541 is technically enormously obsolete part and if you are in search for good distortion specs do not waste your time with it. Almost any recent unit is better. Did you know, there are even chips like this (see below)? And you can have that for the price of ‘41/S1.

Pedja
Click to expand...

Pedja,

which one is this ? Multobit ?
 
#13
“An NE5534 op amp is recommended for the I/V circuit to obtain the specified performance. Dynamic performance such as the gain bandwidth, settling time, and slew rate of the op amp affects the audio dynamic performance of the I/V section.”

No idea what they wanted to say by the second sentence. Makes you read the datasheets as a secondary literature.

Btw, I don’t think you will prove they are wrong.
 
#15
Pedja said:
It is PCM1792, but PCM1794 looks the same. You must be careful with implementation because, as per manufacturer’s recommendation, to achieve this level of performance it is necessary to use NE5534 as an I/V converter.
Click to expand...

This has been discussed sometime ago on a thread on this forum.
For me, BB engineers are (were?) audiophiles.
TI engineers are just engineers.:clown:
NE5534?:bawling:
 
#20
till said:
horowitz/hill says the opamp is critical for perforance as the DACs output capacitance does a phase shift with the feedback resistor and so fast settling opamps are needed
Click to expand...
Actually it is the input capacitance that creates pole and makes the phase shift with feedback resistor. In the case of usual voltage feedback opamps you can correct this using feedback capacitor. In the case of current feedback opamps it happens at pretty high frequencies.

But this hardly solves I/V opamp problems.

Pedja
 
Status
Not open for further replies.
Top Bottom