here is my DAC

[A 8]

Since takashi just built his own I think he would appriciate any help to fix it......before he will consider switching it for something else😱

-Takashi, Your signals look bad so it should sound bad. Unless this is how a non os dac should look something must be wrong.

Can you show the schematics?

 

[takashi]

hi all,
here is the circuit diagram of the dac, supply voltage is 5V

the resistor of TDA1543 will be no problem, because i use the CRO and fine tune those resistor, the coupling cap is 100uF
and i am building a active lowpass filter and test can it improve the high frequency sine wave.

 

[tschrama]

The signals look typical for a NOS DAC, except the HF noise. The noisy garbage in the signals may be related to the power supply. I hope you can show us the schematic to tackle the problem.

Regrads.

Thijs

 

[takashi]

may not be the power supply problem
i have use 3 different LM1085-5 for ad1892, 3X74ahct74 and 2Xtda1543 seperately.
and each regulator, i have a inductor and capacitor to do filtering
and for each IC, i have E.cap and SMD cap for decoupling.

do any one have test of hf sine output from tda1543?

 

[takashi]

one more thing, i cant hear any bad sound, everything seems(listen) normal.
so have any one have use a CRO to test the output of TDA1543/NOS DA?

thanks

 

[takashi]

I have downloaded the filter pro program from TI and help to cal. a low pass filter, the program will show the curve of the filter, but it will not show the curve below 1Khz!
so any recommended program to cal. low pass filter?

i have follow the circuit and make a 3 pole filter, 40k cut off, but the bass have very bad sound after i add the filter. and cannot hear the high freq improvement.

i think i need to use a CRO again~

thanks

 

[tschrama]

I have tested my NOS DACs extensively with custom made test signals on a self-made test CD with Goldwave and a CD-recordeble. Using the apropiate frequency (997Hz?) I get a perfect square wave without ringing or HF noise on my cheap 20MHz analog Osciloscope. The TDA1543 is noisy, so a -90dB signal is allmost totally lost in wideband noise.

If it's not your supply, try elliminating some components and see what happend. For instance only one TDA1543, no reclocking, or both.

Goodluck,

Thijs

 

[leo529]

takashi, the DAC that i show really quite expensive, but the quality is not bad

about your DAC, have you try remove all reclock circuit? or just reclock the SDATA?

 

[takashi]

hi, tschrama,
i have remove the the reclk circuit, but result same
may be i need to try other method......
thanks~

 

[leo529]

CS8412

Why don't try the cs8412/14, many people ar using it, and get e good result on it.

 

[takashi]

there is the photo of 3 - order LPF

 

[takashi]

here is the 1k hz sine

 

[takashi]

and it is 20hz

 

[takashi]

100hz sine wave

 

[takashi]

greatly improvment of 10 k sinz wave

 

[takashi]

20k sine wave

 

[takashi]

and finally, i find out what is the fault~

the 3 resistor of pin 6, 7, 8 of TDA1543 need fine tune~
and the signal is now output from the tda1543 is about 2.5Vp-p

so the singal is fine now(without any filter)


and after add the active filter, the signal is very good~

and why i dont use cs8414? i just want to try something new!
i dont want to always follow the standard! if no one to try, they DIY player will have lee thing to play~HAHA

 

[Peter Daniel]

and why i dont use cs8414? i just want to try something new!
i dont want to always follow the standard! if no one to try, they DIY player will have lee thing to play~HAHA

It is good to try something new. But it would be even better if you tried the old stuff too and can share the results ( differences) with DIY community. Then your efforts are really worth something😉

 

[takashi]

yes~ i agree!

so i have done a phototype of cs8414+tda1543
i compare the i2s with AD1892, i find that AD1892 and DIR1703 have a more accurate Bclk, LRclk!
i think it is because of the oscillator that AD1892 and DIR1703 used.

 

[Circlotron]

one more thing, i cant hear any bad sound, everything seems(listen) normal. so have any one have use a CRO to test the output of TDA1543/NOS DA? thanks

Isn't that an interesting (and very honest ) observation! The signal looks more jumbled and messy than a teenager's bedroom and yet it sounds just fine! So, are we tweaking things that actually don't matter, just chasing after nice looking waveforms that don't make the sound any better?

/Circlotron - wishes he had an external dac too so he can see what all the fuss is about.