Project non oversampling DAC

Project non oversampling dac

Long time that i have been looking for a diy dac , until know I used a sony ES505 and the best pioneer cd player , both are excellent value for money , and having tried many outboard dac on both machines from sonic frontiers ect , never under the impression that they improved the sound .So I never got further then the stage of gathering info .And some clients of mine have Wadia equipment and those didn't impress either .

Then a few months ago while surfing on the net and searching for dac information ,because this is one of these things i allways wanted to build I stumble on Galavotti's home page and I read about his project of building a non oversampling dac , also there is a link to Kusonoki's
Theory article that appeared in a Japanese magazine known as MJ , I was very impressed , this is the first time that I felt real progress has been made .The kusunogi article explained what was wrong with normal dacs and suggested a simpler approach for the quest of audio nirvana fullfilment.So i searched the net for more info on this subject and found quite a few , the polish design from Lesha and
other similar designs by Thorsten Loesch and Andrea Ciuoffoli, All very similar but with there own personal touch .

Why very similar ?

Because all have dedicated powersupplys and most either use tda1543 ( cheap and not as good as the tda1541) or the tda1541.

the connections of the tda1541 are std see philips databook .no hocus opocus there. and the I/V conversion is also std and and i tried without filter and most use a filter ,the tube output is a must for serious DIY fans .

then there is the question of input receiver and the majority uses crystal cs8412 so i used it too.

A bit of research and a month later , I draw schematic v1 and approach a few customers about this project and they like it, since my last project was with stephane ( the outer limit ) I give him a call and ask him to look the project through as he has lots of experience with digital circuits and pcb design for critical RF applications .The revised powersupply is all courtesy of him , he managed to use just one transformer for all ( it does have 5 different windings )
And with 1 extra transformer it is also fully capable of supplying current to a Philips vam 1254 high end drive mechanism .

We added a CS8412cp for the people who wish to use it as a stand alone dac .

The circuit : click here for schematic of dac

Very straight forward :
All powersupplies have current sources for optimal shielding against each other , 2 ground planes are used : digital and analog .
Elna cerafine & philips elcos used , everywhere .It can be used as a stand alone dac , also .
With a pulse input transformer from Lundahl for optimum signal transmission quality .
Followed by the CS8412 , this all goes straight to the TDA1541 from Philips which still is the reference 16 bit dac , there the signal is sampled in top quality mkp capacitors matched to 1%
( this is very important as timing is critical ) the 470pf is a silver mica used for the clock , also 1% selected .(allthough the value is not critical !!!! )
In the analog outputstage no filtering !!!!!!!
The current to voltage conversion is done by simple dale 1w 68 ohm resistors , 2 in parallel to avoid clipping of the dac outputchip .the analog signal is amplified by a tube amp : a 5842 with current sources as to have optimum quality , (this is almost as good as anode chokes )
And a relay mutes the tube stage when it is powered up .
The tubes are biased with negative bias , I just cant tell you guys enough how much better this is .

The power supply's :

Pcb A : powersupply for the dac board
everywhere 3A regulators are used : LM185 and LM337 , one source 7vac winding does the job , voltage is doubled and doubled again when a higher voltage is needed , the multiple regulators filter out all riplle and most importantly isolate the supplies in ac from each other . for the dac and input receiver we need :
+5v 57mA 2times
+5v 83mA
-5V 83mA
-15v 83mA
GND

Pcb B :
Powersupply for philips cd vam 1254 :
Here we use a separate toroid of 2x9vac again regulators LM1085 3A
A voltage doubler is used to obtain the -20v for the gas light display bias voltage
The voltages needed are :
+9v 1A
+5v 1A
-20V 10mA

to get the 2,5 volt necessary for the display of the cd player , you need to wind approx 16 windings on the 2x9vac transformers , they are connected to the 20volt negative bias supply .

Pcb C : high voltage schema
The high voltage for the tube outputstage is rectified here , a triac is used as delay switch
The high voltage is mos regulated at 220 volts dc
The low voltage for the filaments and for the relais and the triac firing are all obtained here and through a 4060 counter activated .

Pcb D : actual dac board

This high quality double sided pcb with separate gnd planes is for the actual dac and outputstage , with the optional CS8412 input clock

first tests and listening impressions :

sound is very good with a shy high frequency output however , voices are beautifull with excellent soundstage , wider and deeper then ususal , easy listening , no fatigue .

Compares really to analogue , its very close but something is missing ( I know the quest for high end is really tough )

What is wrong or what is missing :

Please note that it is barely mentionend on the i-net, not very documented or maybe i got so excited by this new approach that i didn't read the fineprint .Because the nature of the non oversampling dac is that there is no manipulation within time , with the signal , it is converted as it happpens and it no error correction algoritms are used either .( to me it is what single end is to transistor amps ).

Stephan , decides to put the dac on the test bench to see how it performs .

Because he suspects something is flawed and its always nice to see that the specs are not too bad .
1Khz sinewave is good but you clearly see that the sinewave is made of steps , but 4 Khz sinewave is quite flawed with fewer steps making the sinewave and 20Khz is a catastrophy .2 steps represent the whole sinewave , in comparison my pioneer PDS06 has a perfect sinewave with no steps even at 20KHz .however this is not the real problem as it is the nature of the non oversampling non filtered design .

the problem measured shows how bad it is : ( maths prove this too )

study of this phenomenon , shows that the clock at 44,1Khz has subcarriers ( harmonics )
that modulate the signal :actually put in words : the whole audio band is duplicated around the carrier 44KHz above and beneath and this continues with every harmonic = 88.2Khz ect
so the problem is the entire audioband 0-20Khz is reproduced thus from 44 + 20 = 64Khz and 44-20 = 24 Khz this particulary hurts audio reproduction as they intermodulate so we get if we have a 20Khz sinewave in the dac we get a duplication at 24Khz but worse we have the same energy as an intermodulation at 4Khz !!!!!!!!
Very easy to test we take a ref CD with testtones and listen to them and indeed at 16Khz you can clearly hear a lower frequency through and at 20 , barely audible this frequency , we crank up the volume of the pre amp and the 4 Khz tone appears very nicely .

Again the solution is to filter : 3 rd order is the minimum at max 15Khz as the filter starts slowly before really working , but with SE tube amps it is not a big problem , theire limit of bandwith naturally slopes it and theire low power protects tweeters from becoming endangered species .

So basically this is a disaster : not even something to publish .

However listening tests suggest the problem is not so bad , as the sound of this dac is practically identical with what my pioneer ( 20bits 8xoversamplig BB dac and legato link filtering ) produces , actually the difference is small , which is a big compliment to the pioneer cd player, sometimes we favor the non oversampling sometimes we favor the pioneer , MY WIFE immediately favored the non oversampling , her opinion , its less aggressive ..

Solutions :

the problem is huge ( I slightly exaggerate ) and a notch will not cure this , we are looking at solutions but they are not evident , the solution is either brickwall filtering or to return to oversampling, not corrected oversampling = no error correction in it , it will take another 3 months to mnf and test .
A gentle filter slope will not cure it either as the problem occurs just one octave above the audioband, and if you don't use tube power SE amps , you may be in quite some trouble .( lots of ultra sonic energy )

Will keep you posted

Copywright 04-2002 BG design Aquablue