DAC AD1853 with discrete output

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
You have good taste, then.

Now I feel like watching DUNE.

Maybe to do opAMPs on output - LT1469 in i-v and LT1122 in filter (and made on output small class A amplifier as optional)? :)
 

Attachments

  • Cxema.jpg
    Cxema.jpg
    304.8 KB · Views: 420
I don't like LME49860 either, only LME49990 in high gain with LME49990 buffer, 4x total.

They say LT1363 sounds better in Class-A bias, I haven't tried that yet.

I will order LT1469 and LT1122 later. :)

PCB for AD1853+discrete (not my).
I draw 2 layers PCB for AD1853+opAMPs, then will draw 2 layers PCB for discrete output. :)
 

Attachments

  • AD1853 SemiVT.jpg
    AD1853 SemiVT.jpg
    763.2 KB · Views: 415
Not enough information about the sound of discrete output: "the sound of discrete output differs from the traditional opAMPs" and all. Someone like - someone hates.

I will link you to a Swiss paper on discrete tomorrow or next day, on phone at the moment.


and now we rebuilding AD844 and LT1363 on discrete elements... :eek::D

What? :snowman2: !!


I think, that Jean Hiraga's Le Monstre 8W will spread this AMP. ;)

I want to build a DIY speaker amp, at the moment I only have "JLH", what do you think of Class-D? Which is the best?
 
I will link you to a Swiss paper on discrete tomorrow or next day, on phone at the moment.

aaaa, not understood. :)


What? :snowman2: !!
opAMP build on discrete elements : D



I want to build a DIY speaker amp, at the moment I only have "JLH", what do you think of Class-D? Which is the best?

Class D for car's subwoofer only. JLH much worse than Jean Hiraga's Le Monstre 8W and Super Class A 20/30W. I almost smashed JLH 1996 after made and heard Hiraga. :):D
 
What does a current out DAC need?
Low Input resistance of the I/V (although a switched capacitor output stage as the AD1853 is not that sensitive, R2R DACS are much more picky). This impedance should be as flat as possible over a broad frequency range, as the DAC spits out, inherently, a high amount of HF components.

The shown discrete Output stage is NOT an OP Amp style stage, but common base open loop.
It has a low input impedance, which rises way later in the frequency band than a op amp I/V (which relies only to a high Loop gain, and it decreases quickly with rising frequency.

I suggest you to leave the discrete output stage as it is (at least from a topology view). You may look to increase the bias current through VT4-8, it gets better when their bias current is >> max current swing of AD1853 (AFAIR +/-3mA).
I'd bias them at least to 10mA.


Anyway, common base stages are the way to go for DAC I/V.
 
What does a current out DAC need?
as current is >> max current swing of AD1853 (AFAIR +/-3mA).Low Input resistance of the I/V (although a switched capacitor output stage as the AD1853 is not that sensitive, R2R DACS are much more picky). This impedance should be as flat as possible over a broad frequency range, as the DAC spits out, inherently, a high amount of HF components.

The shown discrete Output stage is NOT an OP Amp style stage, but common base open loop.
It has a low input impedance, which rises way later in the frequency band than a op amp I/V (which relies only to a high Loop gain, and it decreases quickly with rising frequency.

I suggest you to leave the discrete output stage as it is (at least from a topology view). You may look to increase the bias current through VT4-8, it gets better when their bi
I'd bias them at least to 10mA.


Anyway, common base stages are the way to go for DAC I/V.
DAC will be not work on low impendance, 1 mA maximum output current. 10 mA bias can damage BJT (300 mW maximum).
I think too - build DAC with opAMPs (and to make discrete output for interest).
About R2R - do you have good scheme on PCM1702 (it's alone multibit, that I can to get in my city (PCM1702U-K))?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.