My new NOS DAC project-TDA151X based

sumanhomroy

Active Member
Joined
Dec 19, 2013
Messages
50
Points
28
Location
Kolkata
Hi All,

I have started building one of my dream, that is NOS DAC (Non- oversampling) with the help of my friend who is a master technician. I always had a dream of having a NOS DAC, however because of the very high price of NOS DAC and also the fact that it is not available in India that easily, I actually thought this dream of mine will never be realized.

That is when my friend gave me hope that ....Yes...it is very much possible to build NOS DAC locally and it will meet the expectation called "Musical", "Analog"...and the journey started.

Will post step by step update of the DAC project....hope your wishes and encouragement will be there with us in the journey that we have taken.
 
Best wishes to your project :), isn't it TDA154X based instead of 'TDA151X'? Two very popular DAC chip made by Philips 2 decades ago are TDA1541 & TDA1543 which almost have cult following status. They also cheaper to implement or design around compared to other alternative R-2R.

Recently I have ordered a generic but finished one based upon TDA1541AS2 from Hong Kong, but not sure whether it is NOS or not, eagerly waiting to taste the holy chip.
 
Last edited:
Thanks for your wishes friend...we are trying the following options-

1. TDA1541A single
2. TDA1541A Double/ Parallel
3 TDA1541A- R1 single
4. TDA1541A- R1 Double/ Parallel
5. TDA1543- Quad/ Parallel

Input stage-

1. Coax
2. Optical
3. USB
4. BNC

Output-

1. RCA Tube out
2. RCA OPAMP out.

Power stage(Filter)-

Design will follow. Planned for separate power filtration for input receiver (CS8412/CS8414), Digital conversion stage, Tube circuit/ OPAMP Circuit.

Option- With Digital filter/ Without Digital filter.

We are putting a lot of effort to make it as close or better than the available non oversampling DAC available in the market.

Will keep you posted on the progress...
 
IMHO, TDA1541 leaves all other DAC's in the dust. I have a 1543 Phillips CD player and a Rotel 955 with a bad display and bad carriage with a 1541 and these 2 are my most favorite CD players, and the Rotel beats the Phillips handily.
I also have a cheap 4 X 1543 standalone DAC, its good, but my GF Tubedac 11 beats it. Of course I want to turn that rotel into a tube follower standalone dac some time in the next 100 yrs.
Cool.
Srinath.
 
Yes, right TDA1541/A/R1/S1/S2 are the most discussed among the community called Audiophiles...also there are many DIY projects of this DAC's. We have also heard many things about this chips...its time for the show...lets see which one takes our heart. TDA1543 is cheap and made for cheap Cd players, however do not undermine the capability of this chip as IMHO this one sound much better than a lot of modern oversampling Dac's. I am not counting frequencies or making any opinion in my mind till the time I actually hear it.

I really thank you for taking interest in my project and sharing your experience and views...keep posting...it will be a big encouragement.
 
What is a NOS DAC.....lets hear it from the father...

An Interview with Kusunoki San, initiator of non-oversampling DACs theory

[Italian version]
[Ryohei Kusunoki]Profiles of Ryohei Kusunoki

(NB: The following is an excerpt from Ryohei Kusunoki's article "Building a House Dedicated to an Audio Room" appeared in the March 1999 issue of MJ. The author notes that he spent more than 100 hours to write this painstaking article).

Ryohei Kusunoki was born in Osaka., Japan January 1956.

Young Ryohei was fascinated with his elder brother's graceful moves when he plays a disc, but his attention was focused on astronomy, rather than electronics.

He came back to the audio world during his college years, when the Japanese Audio Industry was at the zenith of its prosperity. His major interest was physical science, but his private life was dedicated to music. He enjoyed listening to various kinds of music and tried some live recordings with his new portable tape recorder.

Next, he builds a small preamp based on the circuit diagram published on MJ. He tried to improve the amp's sound with higher-quality wires and parts and made a phenomenal success.

After graduation, he joins an automobile manufacturer in Osaka as an engine designer. Busy, with this totally different life, his attention to music and audio, is suspended temorarily until he marries. The youny newly weds move into a small apartment.
Unsatisfied with the acoustics of his new appartment, Kusunoki purchases a Technics SH-8000 sound analyzer and begins to investigate the reasons.
The SH-8000 and an equalizer unit flattened the frequency response to +/- 0.5dB, but, still dissatisfied with the results he tried to find new clues that would improve sonic performane by going to the concerts held at The Osaka Symphony Hall.
This led him to investigations in sound stage and newly rising audio-visual trend. In his pursuit to reproduce a "sound stage" just like that of concert halls, he tries almost anything available at the time.
As a result, Kusunoki finds out that the key to his quest is in fact sound "reflection". At that time, the Japanese "economic bubble" had burst and with it brought unexpected falls in the property prices. Kusunoki now decides to build his own house dedicated to audio.

He now commits his research to reading through heaps of architectural magazines and finds a very good design in the end. The design objective for Kusunoki's audio room was to implement a full-volume listening auditorium at night time without disturbing his family. As a result, his audio room had a similar construction to that of recording studios: i.e. double walls with double laminated windows and a door within a totally floated inner room.
Looking back his life as an audiophile, Kusunoki's interests in audio was not centered at equipment or music.
He notes:
The essence of my audio life is comprised of questions, such as "Why something sounds good?" "Why the sound has improved?" "Why we feel that certain sounds are good?" and so on. From reading books and magazines, analyzing something, thinking about most things, be troubled with everything, and building stuff with substance, audio can exist in various perspectives.
It is also quite enjoyable to talk about these with my friends.

TNT-AUDIO >
Which kind of music do you prefer?

KUSUNOKI SAN >
I used to listen to Japanese folk songs, and my tastes have changed to classical music, jazz music and then to pop music. Avant-gardes such as Geinou Yamashirogumi are my favorites, too. My latest preferred player is Chie Ayado, Japan's female jazz singer. Her powerful voice is really fascinating.

TNT-AUDIO >
How did you begin designing audio systems?

KUSUNOKI SAN >
I started to build my own system during my college days. Like most students, I didn't have money for fancy equipment. So I decided to build my own system to maximize the limited funds.

TNT-AUDIO >
What do you think of tubes vs solid state? And of digital vs analog? What do you prefer?

KUSUNOKI SAN >
Both, tube amps and solid-state amps have their own distinctive sound. I'm using a tube amp for TV at my home, because the amp helps me to listen quite relaxed. I also use a solid-state amp for pure audio reproduction. If I have to choose one between these two, I would pick the tube amp. I like the rich and crisp sound from tube amps.

TNT-AUDIO >
You are generally recognized as the initiator of the zero-oversampling theory. How did you make up that kind of idea?

KUSUNOKI SAN >
Non-oversampling was a popular technique on early CD players, usually came equipped with a high-order Chebychev filter. My method substitutes the Chebychev filter with human ear's intrinsic "filter function," and this is the method's distinctive aim.
I believe that audio components should not conclude by themselves; instead, they must be designed with how the user actually listens in mind. I have been strongly interested in this particular aspect, so the idea of my non-oversampling design was a sort of inevitable consequence.

TNT-AUDIO >
We are not going to ask you about your philosophy just because you made it perfectly clear in the MJ article, which is available on the Web in English. Have you anything to add or change?

KUSUNOKI SAN >
I have made some additions to the MJ articles in my interview published on the volume 6 of Audio Amigo magazine. (NB: Here's an excerpt I've added to Audio Amigo magazine.)

I have been paying attention to the digital filters these days. I have described my DAC design as a "non-oversampling" in the MJ articles, and the appellation got out of control thereafter. Among those DAC components, the digital filters should be more important -- that's what I think at this moment.
The choice of Philips's TDA-1543 to the non-oversampling DACs published on MJ was a pure coincidence, but anyhow the DAC chip had one unique quality: a very high output. This helped to use discrete passive I/V conversion circuit rather than IC chips. The discrete I/V circuit sounds much better than the ICs.
The non-oversampling DACs have distinctive tonal quality, but I couldn't figure out the reasons in the early stages. I found the answer after listening to a DAC using eight DAC ICs to bring about 8-times oversampling without digital filter. The DAC's sound clearly indicated that oversampling was not the culprit of sound degrading, but the real offender was the digital filter.
Digital filters cut off signals beyond 20kHz with a very steep curve, but needs around 2msec of time to calculate the enormous data. I think this is the reason of "diffusion of sound coherence", the characteristic tonal quality of the oversampling DAC.
There is a slight possibility that a digital filter-less DAC's intrinsic quantizing noise, existing beyond the audible range, can badly influence the sound. In my experiments, however, the noise is effectively eliminated with a first-order low-pass filter.
The original Compact Disc format was based on the assumption that a "human can hear up to 20kHz" in essence. So why bother oversampling and cutting off the "inaudible sounds" generated by oversampling? I hope my readers to be skeptical on this methodological inconsistency.
So, what is the sampling frequency in essence? Sampling the sound with 44.1kHz means that the CD can "differentiate the sound up to 25 microseconds." Raising the sampling frequency to 96kHz, for example, should not be considered as an extended frequency range up to 48kHz; it should be regarded as an "enhanced precision - over time domain," instead.
TNT-AUDIO >
There are studies showing the human ear sensitivity is extended to frequencies higher than 20kHz, at least in dynamic situations. This seems to contradict your theory. Our ears, anyway, tell us that you cannot be far from being right. What do you think about this?

KUSUNOKI SAN >
My theory is based on the assumption that our audible range is limited to 20kHz, as I have explained in the Audio Amigo interview. Therefore, if we can hear the sound beyond 20kHz and be influenced by it, this would be inconsistent to my theory.

TNT-AUDIO >
Listening to zero oversampling offers a lot of different and opposed experiences at the same time. On one side the sound appears much richer in harmonics than other CD players. Do you agree with this?

KUSUNOKI SAN >
I think there are no differences among them regarding the harmonics. However, I think my DAC can reproduce the transition of harmonics quite accurately. With my DAC for example, you can hear the subtle details of the artist's performance.

TNT-AUDIO >
One possible explanation is that the ear can perceive ultrasounds, but cannot recognize the pitch, so the brain figures out they are harmonics. Could you comment on this point?

KUSUNOKI SAN >
The assumption doesn't sound wrong, however, very high-frequency sounds are easily attenuated in the air. Therefore, we can hear only those ultrasonic notes existing very close to our ears, and this assumption cannot be regarded as a valid contention.

TNT-AUDIO >
On the other side there is an audible loss in high frequencies, a few dBs at around 20KHz. According to you, is the drop in high frequency response an advantage or a shortcoming?
If you consider this a limiting factor, have you ever tried to solve it?

KUSUNOKI SAN >
Certainly there is such loss when you measure the frequency response. However, the loss can be detected only by those people who are very sensitive to high-frequencies, and most listeners cannot differentiate the attenuation of sound. The loss is not favorable, but I think it is not that important.

TNT-AUDIO >
There are also at least a few design (some industrial, some DIY) filters that in some way introduce a drop in high frequencies minimizing phase rotation. Do you consider this a reasonable way to eliminate digital sources edginess?

KUSUNOKI SAN >
With my experience in building DACs, there are no direct relationship among the edgy, harsh sound of the digital source and its frequency response and phase characteristics. The edginess of sound is generated at different places - the DAC's mechanical construction is particularly important.

TNT-AUDIO >
Which is the story of your designs? We know at least two schematics, one with a quadruple TDA1543 converter, and one with a single TDA1543 with a re-clocking system based on a 50MHz local clock. In which order were they published, and is there anything else we do not know?

KUSUNOKI SAN >
There are three DACs that I have published so far. They are:

The first one used four TDA1543s with which I have entered MJ's DAC Contest. This one was intended specifically for the contest with painstaking work and utilized unusual technical features. So I did not fully explain the details in the article. The DAC is also known as the "Bucket DAC," because I did use a bucket as the power supply unit's casing. The article appeared on November 1996 issue of MJ.
This is the simplified version of the "Bucket DAC." Its basic configuration utilizing the TDA1543 and passive I/V conversion remains the same, but I've limited the power supply to 5 volts, so the output level went down to 0.9Vrms. This one is known as the "Simple DAC." The article appeared on March 1997 issue of MJ.
This one is essentially the "Simple DAC" fitted with an independent, non-PLL clock. I also made some tonal fine adjustments. Regarding the layout of the circuit board, I think the Simple DAC is better because the third one was a sort of evolution model of the Simple DAC. This one is known as the "Improved Simple DAC." The article appeared on December 1997 issue of MJ.
TNT-AUDIO >
Could you explain, how does your re-clocking system work?

KUSUNOKI SAN >
The operation of the system is quite simple: to rectify the clock signal generated by the DAI-IC, such as a CS8412, with another free-running clock. The problem here is the reason why we feel the expansion of sound stage? Other friends and myself verified the existence of this phenomenon. My presumption is that the clock signal generated by the existing DAIs contain jitter that correlate to the source signal, and the re-clocking breaks up this correlation.

TNT-AUDIO >
You take a great care in designing the layout of your systems. How much do you think the result depends on the care in these details?

KUSUNOKI SAN >
This is a very important aspect in digital circuitry, and its magnitude far exceeds that of the analogue circuits. The DACs built without careful attention to the layout tend to reproduce cold, irritating "digital" sound.

TNT-AUDIO >
Your designs make use of a high value I/V conversion resistor, so that the output value can directly drive the pre-amplifier. Isn't there a dynamic problem, in this way?

KUSUNOKI SAN >
If the "dynamic problem" means distortion at high output levels, the DAC used here (TDA1543) is the one only chip without any problem. Unfortunately, other DAC chips can cause a serious problem.

TNT-AUDIO >
What do you think of new digital technologies, SACD and DVD?

KUSUNOKI SAN >
The new technologies are welcome and I'm quite interested in them. But if the reasons to alter the original CD format are to expand frequency range and dynamic range, I must say, they are barking up the wrong tree. This is why the new technologies can provide just a tiny improvement to the sound quality compared to using better quality capacitors and resistors. I think the multi-channel system is the only sole hope remaining.

TNT-AUDIO >
Are you interested or active in other audio areas, apart DACs?

KUSUNOKI SAN >
I haven't published anything other than DACs, but I'm interested in all areas of audio. I make cables and assemble speaker systems quite often. I believe making any possible component gives much more fun and satisfaction than simply buying audio equipment over the counter.
In my opinion, Audio would have to be one of the Worlds best hobbies... even for the professionals.
TNT-AUDIO >
Are you working on any new design?

KUSUNOKI SAN >
There are so many things I'm planning to build, thered need to be ten Kusunokis though, even, this wouldnt be not enough. If I am planning to publish one of these projects and somebody asks me that "when can we expect the next tocome out?" This question would be the most discouraging to me. When I'm finished with something that leads to new findings... I will positively publish it.
 
Best wishes to your project :), isn't it TDA154X based instead of 'TDA151X'? Two very popular DAC chip made by Philips 2 decades ago are TDA1541 & TDA1543 which almost have cult following status. They also cheaper to implement or design around compared to other alternative R-2R.

Recently I have ordered a generic but finished one based upon TDA1541AS2 from Hong Kong, but not sure whether it is NOS or not, eagerly waiting to taste the holy chip.
I read some articles on NOS and R2R DACs. Its quite intriguing. Eagerly waiting for your TDA1541 dac review.

Could you please elaborate which R2R implementation to go for? Is there any other (or better) alternatives available?
 
Could you please elaborate which R2R implementation to go for? Is there any other (or better) alternatives available?

Many brand out there, you have to choose as per budget and reputation. Like Vinyl, R-2R seems to be making strong come back than ever. R-2R's basic working principal reproduce natural or original sound timbre more faithfully, but their specs not so attractive like DS Chips, just like playback of Vinyl where specs (like Signal Dynamics, SNR etc....) are not too attractive but the analog timbre is hard to match.

PCM1704 is very famous R-2R chip which being R-2R also achieved nice spec. Analog devices also have some very renowned R-2R chip.

Metrum is a strong brand which use some industrial R-2R chip (in their DAC) about which they kept secret till now. I have almost never heard any negative review about this brand. They also appreciated by hifi lovers for much longer years than schiit audio, so the brand is time tested.

However I have choose TDA1541 due to it's great reputation and lower priced implementation offerings. Reputation and Value intersects here at best possible point according to my budget and understanding. Also before spend too much or any final jump on PCM1704 based DAC I wanted to taste R-2R with a cheaper alternative.
 
Last edited:
Dear Saikat...sorry for the belated reply...Merry Christmas to all our friends in hifivision. R-2R is very different concept than what we are trying to implement here...we need the sound as natural as recorded in the cd...which is 16Bit 44.1 KHz. A Dac which will produce pure natural sound without oversampling and filteration..R-2R on the other hand is 24Bit and 382Khz which cant be natural. The reason we selected TDA1541A because it can produce 16bit and 44.1Khz and it arguably the best musical DAC chip of the famous TDA line up...we are trying quite a few implementation option which I have mentioned in my thread earlier...I have listen to many oversampling Dac's , however NOS Dac's can produce close to Vinyl sound as per my opinion, and the reason we will not put any filteration in between because i want my ears to decide what is bad or good in a recording...rather than cutting off some frequencies...I said earlier..we will not be counting bits and frequencies...we want pure musical sound...hope you understnad
 
PCB's getting prepared-

1. DAC board (TDA1541A), option- single/ parallel chip, with digital filer/ without digital filer
2. Digital receiver board- CS8412/14
3. Input board- 5 input options- for me I have selected Coax, Optical, BNC and USB.

Will keep posting the progress...
 
Last edited:
Dear Saikat...sorry for the belated reply...Merry Christmas to all our friends in hifivision. R-2R is very different concept than what we are trying to implement here...we need the sound as natural as recorded in the cd...which is 16Bit 44.1 KHz. A Dac which will produce pure natural sound without oversampling and filteration..R-2R on the other hand is 24Bit and 382Khz which cant be natural. The reason we selected TDA1541A because it can produce 16bit and 44.1Khz and it arguably the best musical DAC chip of the famous TDA line up...we are trying quite a few implementation option which I have mentioned in my thread earlier...I have listen to many oversampling Dac's , however NOS Dac's can produce close to Vinyl sound as per my opinion, and the reason we will not put any filteration in between because i want my ears to decide what is bad or good in a recording...rather than cutting off some frequencies...I said earlier..we will not be counting bits and frequencies...we want pure musical sound...hope you understnad

I think you are having a misconception here, the basic design (core) of a DAC which we use in Audio generally comes in 2 packages (or types): R-2R (or ladder) and DS (Delta Sigma or Sigma Delta for ADC or DAC). A new very popular DAC which use FPGA dac that is also a sigma delta implementation but programmable as per best knowledge and requirements known by the designer.

Now whether you want to use any chip or disign with NOS or not (choice of filter) that is completely different matter, in fact it is matter of your own ear, taste and logic.
 
Last edited:
The following should explain-

Functional description
The TDA1541A accepts input sample formats in time multiplexed mode or simultaneous mode up to 16-bit word length. The most significant bit (MSB) must always be first. The flexible input data format allows easy interfacing with signal processing chips such as interpolation filters, error correction circuits, pulse code modulation adaptors and audio signal processors (ASP).
The high maximum input bit-rate and fast setting facilitates application in 8 oversampling systems (44.1 kHz to 352.8 kHz or 48 kHz to 384 kHz) with the associated simple analog filtering function (low order, linear phase filter).

True 16-bit performance is achieved by each channel using three 2-bit active dividers, operating on the dynamic element matching principle, in combination with a 10-bit passive current divider, based on emitter scaling. All digital inputs are TTL compatible.

Thanks for sharing your knowledge, sorry for the delayed reply as I was away of town.
 
Your explanation have not touched the issue you yourself raised on R-2R or Ladder chip "R-2R on the other hand is 24Bit and 382Khz". Seems to be you have just quoted the explanation from here Philips TDA1541A S2 double crown d/a converter, which is just functional description.

You are using a ladder chip (TDA1541A) but at the same time you are denying it by making statement like "R-2R is very different concept"

Whatever it is, if you can understand it the way you want then be it, good luck! But one should try to make the basic concept clear especially before make any statement on public forum.

I will not derail your enthusiasm anymore.
 
Last edited:
If I can built something which could sound as good as a USD2000 DAC, I will be happy, at least my friends who could not afford $$$$ will find it helpful...and I will be happy to reach my destiny...surely will welcome all for audition...its my first project...and nothing can derail me...until I finish it.
 
Dear Suman and Saikat,

I think we are having a bit conflict of interest. We all want a happy ending. Don't we?

I'd like to thank both of you for enriching this thread with your knowledge.

Now I don't want anyone of you to drop out or stop contributing more and more to this thread due to a small misunderstanding.

I really learned a lot from all of you. I am really thankful. Now gentelmen, let's shake our hands and let's enrich this forum even more.

Thanks.
 
Right...agree..in fact I did get a lot of insight from Saikat...thanks to him. Will soon upload more update regarding the DAC project...hope o complete the project by 1st week of Jan.
 
That's awesome.

However while looking for different R-2R implementations, I just stumbled upon this:

dam1021 Product Range - *NEW* - Audio Products - Products

I hope you experts could provide more insight on this.

It's very promising and quite future proof ladder design, have lots customization option. Very very popular among DIYers - Reference DAC Module - Discrete R-2R Sign Magnitude 24 bit 384 Khz - diyAudio, a dream project if you can dedicate yourself.

But just only the ladder module costing you $200 to $300 (add shipping cost and duties xtra), and so many other components and tasks remained there. So I guess it is close to $1k project (assessing the time and labour also) which is quite premium for DIY effort.
 
Thanks for sharing this...we will surely look into the same once this project is over, the PCB's are complete and the other components like DAC chip, receiver chip, cap's, transformers should be in place by this week...lets see how it goes.
 
Purchase the Audiolab 6000A Integrated Amplifier at a special offer price.
Back
Top