output stage for dac

[Deleted]

I still think around Q5 can be improved

Leo
Quite likely, but as it's only an emitter follower (if bipolar) the audible improvement may be hard to pick, compared to what you have already achieved! ;D
Alex

[leo]

Maybe I should try a valve stage, get Miguel interested

[pagan]

Some info for ouput stages here
www.head-fi.org/forums/f6/diy-analog-stage-zhalou-other-v-out-dacs-276297/

They start talking about the pcm**** devices after a few pages

There are some more, but i need to find them again

found another
www.diyaudio.com/forums/showthread.php?s=&threadid=6121

[leo]

Thanks for those, I'll have a read now

[jonclancy]

leo said:

Maybe I should try a valve stage, get Miguel interested

Lampization? ;D

[leo]

jonclancy said:

leo said:

Maybe I should try a valve stage, get Miguel interested

Lampization? ;D

It did cross my mind

[pagan]

leo said:

jonclancy said:

Lampization? ;D

It did cross my mind

"Lampization"
Does this mean valve?

If yes.
Just use Fet's and increase 2nd harmonic untill it sounds like a valve.
At least you can do it at 1/10 of the voltage

[leo]

pagan said:

leo said:

It did cross my mind

"Lampization"
Does this mean valve?

If yes.
Just use Fet's and increase 2nd harmonic untill it sounds like a valve.
At least you can do it at 1/10 of the voltage

Sorry Allan, I was just having a little joke, people who know me already know I've tried a valve output stage in one of the TDA1541A based dacs I built a while ago, its not really the sort of sound I want now although back then it was interesting to try it out

[Deleted]

Just use Fet's and increase 2nd harmonic untill it sounds like a valve.
At least you can do it at 1/10 of the voltage

[leo]

Decided to have a look at two totally different dacs outputs on the scope, signal used was a 1k 0dBFS Square wave cd

Notice the pre- post ringing on the second one

[pagan]

Leo
You might have seen this
www.lcaudio.com/images/ZapFilter/zf2dia.gif

are you any good at reverse engineering?
and maybe, try to improve.

[pagan]

leo said:

pagan said:

"Lampization"
Does this mean valve?

If yes.
Just use Fet's and increase 2nd harmonic untill it sounds like a valve.
At least you can do it at 1/10 of the voltage

Sorry Allan, I was just having a little joke, people who know me already know I've tried a valve output stage in one of the TDA1541A based dacs I built a while ago, its not really the sort of sound I want now although back then it was interesting to try it out

A while ago, ay, valves, ay. what turn of the century?
SandyK, remembers

I know, I Know delayed responce, but better late than never

[leo]

pagan said:

leo said:

Sorry Allan, I was just having a little joke, people who know me already know I've tried a valve output stage in one of the TDA1541A based dacs I built a while ago, its not really the sort of sound I want now although back then it was interesting to try it out

A while ago, ay, valves, ay. what turn of the century?
SandyK, remembers

I know, I Know delayed responce, but better late than never

Heres parts of one I was making, the sound was actually not bad at all, I may do something else valve based a little later for the laugh

[leo]

pagan said:

Leo
You might have seen this
www.lcaudio.com/images/ZapFilter/zf2dia.gif

are you any good at reverse engineering?
and maybe, try to improve.

Interesting, looking at the schematic it appears it can be set for voltage out dacs and current output

[pagan]

leo said:

pagan said:

Leo
You might have seen this
www.lcaudio.com/images/ZapFilter/zf2dia.gif

are you any good at reverse engineering?
and maybe, try to improve.

Interesting, looking at the schematic it appears it can be set for voltage out dacs and current output

Leo
I'm using one at present on a voltage output dac, wm8740 dual diff.

Solder the bubbles to set for either voltage out dac or current dac.
also either balanced or single ended inputs.
No caps on in or out, think it uses a servo. also onboard reg.
It's a pain in the but dam things nearly all SMT

[pagan]

leo said:

Decided to have a look at two totally different dacs outputs on the scope, signal used was a 1k 0dBFS Square wave cd

Notice the pre- post ringing on the second one

If thats input signal and output signal
it's not good

[leo]

I can't see any servo on the schematic, so what sort of price do these things cost? not cheap I'd imagine.
I could do a layout for through hole but its going to be big, could try SMD but they are always a pain in the arse to etch out yourself

[leo]

pagan said:

leo said:

Decided to have a look at two totally different dacs outputs on the scope, signal used was a 1k 0dBFS Square wave cd

Notice the pre- post ringing on the second one

If thats input signal and output signal
it's not good

You tried looking at yours? you'd be surprised what OS does The first pic is the output of the AYA, perfectly flat, not bad for the ye old TDA1541A

[pagan]

Leo

the TDA1541A just sound very musical, seems to be a character of the dac.
Add a tube (personal preference) for flavor.

Hee Hee

Like adding salt into the wound

[leo]

The second one if of the PCM dac with its OS and upsampling

[leo]

Do you have a scope Allan?

[Deleted]

If thats input signal and output signal
it's not good

Allan
The CRO photos were the outputs of 2 different DACs . Leo was using a test CD. The one with the ripples was typical of earlier generation CD players.

Leo
Which DACs were the outputs from, and more details please of type of DAC and technology used . Did you try both with a higher frequency test tone ? (Test disc permitting)
Alex

[pagan]

leo said:

I can't see any servo on the schematic, so what sort of price do these things cost? not cheap I'd imagine.
I could do a layout for through hole but its going to be big, could try SMD but they are always a pain in the arse to etch out yourself

I saw another schematic or someone discussing it on diyaudio IIRC,
A dc servo was mentioned, but I thought the opamp was for an onboard reg.?
I'll see if I can find it.

It's not cheap
www.lcaudio.com/index.php?page=6
but it is good, but there is always better.
I use it as a reference output stage, to compare others.

please no SMT, it's a pain in the but.
Thought hole better, try different component easlier.
I'm having enought difficulty reading normal components let alone smt.

Besides someone mentioned putting a ccs in it.
need to find the post again.


Yes I have a scope, but an old 20Mhz one

[leo]

The first was on the output of the AYA dac, TDA1541A S2 common base I/V output stage, it also uses compensation circuitry to deal with NOS

The second one with pre and post ringing is of the output stage of the much newer PCM dac, if I add some filtering on the dacs output it should improve things BUT adding this filtering usually hurts the sound IMHO

The PCM is the one I'm still working on, I'm only limited by the output section so far so this is something thats giving me the biggest headache
I've tried the recommended circuit in the datasheet using op-amps. After trying the simple discretes I found them to have the best potential sound wise but are harder to get right

[leo]

pagan said:

leo said:

I can't see any servo on the schematic, so what sort of price do these things cost? not cheap I'd imagine.
I could do a layout for through hole but its going to be big, could try SMD but they are always a pain in the arse to etch out yourself

I saw another schematic or someone discussing it on diyaudio IIRC,
A dc servo was mentioned, but I thought the opamp was for an onboard reg.?
I'll see if I can find it.

It's not cheap
www.lcaudio.com/index.php?page=6
but it is good, but there is always better.
I use it as a reference output stage, to compare others.

please no SMT, it's a pain in the but.
Thought hole better, try different component easlier.
I'm having enought difficulty reading normal components let alone smt.

Besides someone mentioned putting a ccs in it.
need to find the post again.

If its a fet based input op-amp you can see on the board I'd imagine it will be the servo.
One thing we have to watch with Servo's is where exactly they are fitted, some people inject their output into the amps input pair transistors which is a bad way IMO, it feeds the noise into the circuit and also the type of op-amp used has an influence on the sound so it solves one problem of eliminating signal DC blocking caps but adds another problem.
A badly implemented servo can have more sonic signature than properly DC biased coupling caps

I'll try and find a high res pic of the Zapfilter, I would like to know how things are implemented before buying such a thing