yes but think about the bang for the buck, especially if you use a newer dac like the wolfson wm8741 or sabre. i mean, your planned 200$ dac is quite cheap to start with, so a little more wont break the bank and still be very affordable. i think it may appeal to quite a number of people given there already is the buffalo and the AMB y2 uses the newer dac chip but with op-amp, which are very popular. a new generation well engineered dicrete stage diy dac could be a different, cleverer, logical step.
Last edited:
also it would be very accessible for people to witness with their hands and ears the (claimed) greatness of your DISCRETE IS BETTER philosophy
They do:
The large recessed heat sinks, allows for the amplifier to run with a high bias, thereby securing that most power in normal listening conditions are delivered in Class A.
No wonder, neither your DAC ever gets finished, nor you have schematics to show as you seem to have nothing better to do than filing lawsuits against other people. :bs:
By the way, for a shunt that delivers power to a more or less static load, not more than 1W dissipation in the regulating device should be necessary.
edit:
For my M25, Pioneer states 30W in ClassA, maybe you could teach them...
Last edited:
The Wolfson 8741 is not really a high end DAC chip, and the ESS chip is so secret, that you´ll have to sign a NDA before you get the datasheet, which I find an interesting way of marketing.
So that leaves us with AKM, CS, BB, AD and a few other more or less interesting alternatives. To me the CS seemes by far to be the best U out chip, and the BB takes the price amongst I out chips.
If you like the discrete design, feel free to build it. You can even have the PCB @ a very nice price.
Btw. if anyone thinks that small companies do have the resources to do DAC chips far superior to i.e. AKM, CS, AD or BB, I think they ought to reconsider. I.e. BB has a very large marketshare, if anyone should threaten them, why on earth should they go on sleeping? That is not the way things work.
The BB1794A is a top notch chip and so is CS4398 and AD1955 and maybe also AK4399. The ESS is still not well known, and it seems as if the performance advantages are acomplished by paralleling several DACs. This can be done with all the others as well, but this seems quite pointles as the DAC chip still, by far is the best measuring component in the whole signal path without parralleling.
@1
No They are way to small for anything but a few watts of class A.
Remember that an amp stating a higher AB output than the class A output runs way hotter than a clean class A amp.
Densen amps are proven not to be class A amps, not even partly.
It is also proven beyound any reasonable doubt+-, that a Densen amp rated @120 Watts in 4 Ohms only produces 1/10 of that.
@2
Tell that to mr. Sillesen, as it was him filing the lawsuit, unfortunately he came out completely loosing, I´ve got the engineers report from the lawsuit, and it is just about the most funny reading on hifi ever.
just say so, if you want a copy.
@3
No!
The pioneer M25 is a genuine class A amp. The data submitted by Pioneer are real and honest.
thanks. i ll consider. but the price (absolute not relative), my level of confidence in making a working unit and the lack of feedback makes it unattractive for me at the moment. but the more your dac four-and-a-half will gain in popularity, the more i ll be interested.
btw, what is the price of the pcb and do you have some kind of list of all the parts needed? just wondering if collecting parts is easy and what the total cost looks like.
A few question to the DAC
Long thread so excuse me if I have missed something....
With what other audio components has the listening tests been conducted (source, amp's, and speakers)?
With which other DACs have you compared it directly? I already know Buffalo32 is not one of these😉
What are the accepted digital audio formats, bit-depths and sampling rates?
Is attenuation in the digital domain possible/implemented?
Is the choice between SPDIF and Toslink input selectable or is it one or the other?
Thanks,
Nic
Long thread so excuse me if I have missed something....
With what other audio components has the listening tests been conducted (source, amp's, and speakers)?
With which other DACs have you compared it directly? I already know Buffalo32 is not one of these😉
What are the accepted digital audio formats, bit-depths and sampling rates?
Is attenuation in the digital domain possible/implemented?
Is the choice between SPDIF and Toslink input selectable or is it one or the other?
Thanks,
Nic
Hey!
A lot of other equipment has been used for listening tests, but some of it is Denon and Sony high end transports both standard and electrically modified ones, amongst others my own Dali Detal preamplifier and Gravity class A power amplifier, Denon high end gear, NAD, ICEpower, Doxa and our own "maybe some day comming" class D amp and more. Speakers from Dali, Gamut and others. Cables has been amongst others, Van den Hul, Furukawa and standard cables from i.e. Canare, Belden, NKT, Mercodan.
Other DACs used for comparisons are amongst others DACs from Denon, Benchmark, Lavry, Orpheus, DAD, Alchemy, Accuphase integrated CD players from Denon. Rega, Sony Accuphase, Wadia and others.
SPDIF RCA accepts 24 bit/96KHz, theoretically it should accept 24/192, but we did not test that, as we do not have any source material.
The output of the ASRC is fixed @ 97,3 KHz, which means anything coming out of the AD1896 has that sample rate, because the CS4398 performs at its best below 100KHz as most other DACs.
Digital attenuation sucks, so the attenuator is disabled, and I think it takes a µP to enable it, so this is not a possibility.
Also we chose not to include any USB inputs, as you then typically will pollute the groundplane with noise and large voltage potentials from your computer, and recieve a high jitter input signal anyway. So basically this unit is not intended for computer audio, or connection to equipment powered by SMPS, at least not direct connection. Toslink might be the right solution for that.
There is an input selector on the rear panel, where you can choose your input electrically, and if only the PCB is bought, a jumper does the shifting job.
We are thinking of doing a computer based audiosystem ourselves, but we have to consider if its worth the effort. We think that maybe some Logitech device or others can be modified to deliver a clean and low jitter signal, without any ground plane noise or large voltage potentials. We do not yet know if that is possible, until then I think you ought to use any SMPS powered source with caution, and eventually use pulse transformers as isolation devices. If not possible then connect first, and turn on afterwards, if you fail to do so, and the inner pin on the RCA connector catches contact first, the reciever CS8416 will probably send out smoke, and probably take other components with it. So computer audio calls for caution, and that goes for all DACs without input transformers.
I recommend Toslink for that.
Last edited:
Thanks KvK. I hope you did not switch between all this gear when you were optimizing individual components🙂
Another question: will the DAC take TTL level S/PDIF safely?
Another question: will the DAC take TTL level S/PDIF safely?
The WM8741 may be never than CS4398. But that does not automatically mean that it is better. If it was, I guess we would have choosen WM8741 😉 .
That's the great difference between our DAC and the typical mainstream ones.
We decided that the cheaper version should be based on an op-amp, to make it much more easy to build. Honestly, the current DAC is not that suited for DIY beginners!
But, we do not intend to just build a application note DAC like any other. We will spend time listening to the DAC, tweaking and optimizing. The idea is, that at $199 it must not be possible to get anything better, and preferable not even at $399.
As KvK said, there will be 2 options for the analog stage, in the same PCB, where one of them will be 100% free from negative feedback!
If you wanna know the difference, just build the DAC, that this thread is all about 😉
The low price version is intended as both a low price and low complexity version.
Why on earth would you even consider using TTL levels at the SPDIF??
The SPDIF is terminated by a 75 ohm resistor. Feeding a 5 volt TTL signal will result in a current of 66.7mA. That's crazy!
The result is a power dissipation of 0.33watt. Assuming 50% duty cycle, the dissipation will be 0.16watt.
But then... NO, you can not use 5 volt TTL at the input. The CS8416 runs at 3.3 volt, and does not accept input higher than Vcc.
So not signal higher than 3.3 volt. This still gives you 44mA of signal current, and a dissipation in the input resistor of 75mW assuming 50% duty cycle!!
But be aware, that at normal TTL output will NOT like the idea of 44mA!
Will the DAC drive headphones directly?
I'm hoping that your answer will be different from: "why would anybody want to use cans", or "seems rather useless without some kind of attenuator".
I admit my ignorance and will prepare myself for another lesson....
Cheers,
Nic
I'm hoping that your answer will be different from: "why would anybody want to use cans", or "seems rather useless without some kind of attenuator".
I admit my ignorance and will prepare myself for another lesson....
Cheers,
Nic
Easy guys!
No there is no headphone amplifier on board the DAC.
The output voltage is chosen as the exact standard which is 2,0V@0dBFS, if that is sufficient for your headphones I do not know.
Be aware that most other DACs have slightly higher output for what ever reason typical 2,8V. This is a cheap trick, that makes them sound just a bit louder for easier to catch customer attention in the hifi shop. The fact is, that already 2,0V are twice as much as normal power amps need to get well into clipping.
Another matter is impedance matching, the output stage on the DAC board is very strong, and can probably drive a pair of headphones, but it just was not intended to do so. So if it sounds nice driving headphones, nobody knows until tried out. But you ought to be carefull fidling with the output since, there is no fuse or other protection circuits, as i.e. in op-amps. This is a high current stage, intended to drive cables and inputs in preamplifiers @10K Ohms and up.
I sugest you build or buy a dedicated headphone amp for your headgear, unfortunately also these often are sloppy op-amp based designs taken directly from an application note.
I will not recommend using the DAC to drive a pair of headphones directly. The DAC has a fixed output impedance around 110ohm and most headphones are around 32ohm (I guess 😕). This is almost a short circuit of the output.
This means that you will only get around 0.45 of output, which may not be enough.
Also this will lead to current draw of approx 15mA, which will pull the output out of class-A operation.
The output will "only" run pure class-A into impedances as low as 100ohm. But that should be enough for most pre-amplifier inputs 😉
I have had a few people asking about adding a headphone output to the DAC. But we decided to not go for it, since we generally do not like the idea of mixing things to much in high end audio.
The headphone output will require a volume-control and an extra output stage. And that does not fit into our design philosophy.
We are also considering a pre-amp, based on the knowledge we got from designing the DAC. And here it would be a more straigt forward design to implement the phono output. I will keep this in mind 🙂
This means that you will only get around 0.45 of output, which may not be enough.
Also this will lead to current draw of approx 15mA, which will pull the output out of class-A operation.
The output will "only" run pure class-A into impedances as low as 100ohm. But that should be enough for most pre-amplifier inputs 😉
I have had a few people asking about adding a headphone output to the DAC. But we decided to not go for it, since we generally do not like the idea of mixing things to much in high end audio.
The headphone output will require a volume-control and an extra output stage. And that does not fit into our design philosophy.
We are also considering a pre-amp, based on the knowledge we got from designing the DAC. And here it would be a more straigt forward design to implement the phono output. I will keep this in mind 🙂
- Status
- Not open for further replies.