Any good TDA1541A DAC kit?

[s3tup]

Ok, i've got proof-of-concept circuit working, for back-syncing with single WM8804 in simultaneous transmit/receive mode, HW configged.
SPDIF Back Sync

It will work the moment i'll add XTAL with MCLK divider to get BCLK/LRCLK clocks, and feed them to both DF (TDA for NOS) and WM8804.
I'll check the solution for 96kHz this evening (max speed of DIR9001 i use for generating clocks which i feed to the WM).

44.1/48 works flawlessly.

[gaetan8888]

Originally Posted by ThorstenL

Hi,

Can you provide proof of this? Well, proof of the contrary is in March 2012 Stereophile.

A DAC I designed with passive I/V and Tube gain stages is tested by JA in Stereophile March 2012 and shows appx. -135dB FFT Noisefloor with 24-Bit Signals (I am sure you can work backwards to the ENOB from this easily).

So tubes DO deliver 16 Bit resolution just fine (and yet another of your claims has been illustrated to be incorrect).

Ciao T

Hello Thorsten

I've take a look today at the Stereophile web site and I can't find those tests.

Can you tell me, please, where they are in the Stereophile web site ?

Thank you

Bye

Gaetan

[SoNic_real_one]

-135dB noise floor is bull. It's because shorted outputs probably and that is just the ADC noise. Anyway, irrelevant.
One of the relevant/important measurements is THD+N at 0dB and -60dB. That shows the actual audio resolution of a certain device.

[ThorstenL]

Hi,

Originally Posted by gaetan8888

I've take a look today at the Stereophile web site and I can't find those tests.

Can you tell me, please, where they are in the Stereophile web site ?

They are in the March issue (print).

Ciao T

[ThorstenL]

Hi,

Originally Posted by SoNic_real_one

-135dB noise floor is bull. It's because shorted outputs probably and that is just the ADC noise. Anyway, irrelevant.

I wrote -135dB FFT Noise Floor. So what you write is Bull...

If you paid attention in the relevant classes you know how these measurements work and you can work backwards from this to ENOB.

Originally Posted by SoNic_real_one

One of the relevant/important measurements is THD+N at 0dB and -60dB. That shows the actual audio resolution of a certain device.

THD itself is meaningless, Noise is derivative of the above. So more bull.

Ciao T

[s3tup]

Thorsten, have you tried different passive filtering options prior to I/V convertion?
The issue with most I/V stages with feedback (and others too) is that their input impedance rises with a frequency, failing at voltage compliance and other troubles for HF noise. Moreover, HF noise ain't good for anything.

How about inductor to I/V with matching cap to GND (or thru matching resistor, to match input impedance of I/V, to GND)? Kinda speaker's crossover with flat impedance for DAC, and frequency splitting to GND and I/V?

By the way, my back-sync works with 96kHz too. I don't have 192kHz source to test it, and there is almost no soundcards with 192kHz input.

[abraxalito]

Originally Posted by ThorstenL

I wrote -135dB FFT Noise Floor.

Yeah but unless SoNic is clairvoyant, he's not going to be able to work out the 22kHz bandwidth noise unless you tell us the effective bin bandwidth. Does the original article divulge this important snippet?

Number of bins, no. of averages, sample rate and the ****-factor arising from the windowing function pretty please?

[ThorstenL]

Hi,

Originally Posted by s3tup

Thorsten, have you tried different passive filtering options prior to I/V convertion?

I normally incorporate passive filtering (1st Order) with the passive I/V, similar to the capacitor in parallel with the I/V resistor in Op-Amp I/V.

For Op-Amp I/V I usually use CFB Op-Amp's, or if not I use VFB Op-Amp's I use the Wurcer Trick (cap from -in to ground) and Op-Amp's that are not unity gain stable. You might call that "pre-filtering", though it actually works differently.

Ciao T

[ThorstenL]

Hi,

Originally Posted by abraxalito

Yeah but unless SoNic is clairvoyant, he's not going to be able to work out the 22kHz bandwidth noise unless you tell us the effective bin bandwidth. Does the original article divulge this important snippet?

Indirectly.

Stereophile always measures the same way over decades now and you can use this data for comparison and work out the correction factors.

In fact, our friend SoNic illustrated not only that he does not understand the difference between noisefloor and FFT noisefloor.

He also illustrated that he never looked at the article referenced and simply declared something to be bull-.... based on his lack of understanding basic electronics and his prejudices against tubes, instead of making an intelligent contribution to the thread or instead of providing proof that my assertions where wrong.

Ciao T

[abraxalito]

Originally Posted by s3tup

Thorsten, have you tried different passive filtering options prior to I/V convertion?

As a relative noob to the I/V malarky, I'll just mention what I'm playing with at the moment as a kind of counterbalance to Thorsten.

I flirted with cap to ground but didn't much like the sound on TDA1543 or TDA1545. I'm not really sure why this would tend to flatten the sound stage, but that was the effect. On TDA1543 the effect was mitigated by using an external current source (set the internal bias to zero). I decided to take the dual, which is a series inductor (after the I/V resistor) and that's been giving me more listening pleasure of late. But I'm doing this in differential mode, with digital inversion and more than one DAC chip.

The series inductor I chose was one with a high SRF (above 200MHz) intended for use as a CM choke. I agree that HF noise is the big bug-bear and DACs create a lot - especially CMOS DACs like the TDA1545.