Right, learning is really the main interest here. I have already designed a stereo DAC from these chips and wanted to challenge myself with a different project.
But your question is essentially the same as mine. Is the paralleling using 100 ohm resistors a good way of paralleling the hot and cold signals? I am asking due to the questionable errors and lack of info in the datasheets, as well as information regarding this topic in general is hard to come by.
But your question is essentially the same as mine. Is the paralleling using 100 ohm resistors a good way of paralleling the hot and cold signals? I am asking due to the questionable errors and lack of info in the datasheets, as well as information regarding this topic in general is hard to come by.
If such information is not available the normal design procedure is to test and find out yourself. The problem which both Altor and me are addressing is that the possible gains of dual mono setup (better DR/SNR) are very difficult to realize in practise and also very difficult to verify.
That's fine. I appreciate the inputs, and I mostly asked if anyone had been sitting with this problem already. I fully intend to bring this project to life. I shall design a test board then!
Also a big question - if it is really required.
I tried parallel mode with different DAC chips, and only with AD1853 and PCM179x there were some improvement in SQ.
No one AKM has no gain..
The old 4499 (which was 4-channel) in any mode, including 2x2, was worse than a single 4497б and the new 4499 sounds great without mono.
And when they turn on the ES903xPro in mono (i.e. 8 channels in parallel), I personally just laugh.
All IMHO, of course 🙂
That's why I asked - what is your goal, what do you want to achieve by using the mono mode?
Alex.
No need for that if you already have a AK4499 stereo board. Should be easy (and non-destructive) to modify it to work in mono mode.
The stereo board is set up in pin control mode to output stereo and the output stage is quite different. Though the I/V conversion is the same so can probably test from there, good point.
Hey Exetic,
It sort of sounds like you want to try something that Altor already tried without much to show for it, so he doesn't understand why you would want to do it. Maybe the answer is something like, "Altor, I want to find out things the same way you do, by trying things to see what works for myself."
The thing is if you want to be like that then you need to try multiple design variations and compare them. Building only one dual mono dac will teach almost nothing since there are so many variables not explored and compared.
Anyway, usually the reason for dual mono is better separation between channels, including for noise effects of one channel affecting the sound of the other channel in some possibly subtle way. Such effects can be hard to measure in a very useful/meaningful way if they are signal correlated. That said, things that are hard to measure sometimes can still be evaluated to some extent by a trained set of ears. In that case the rest of the system needs to be up to par so that any differences in sound area easily audible. Good speakers in a good room with good amplifiers can help a lot but it can also cost a lot of money, maybe even as much as an AP 555x.
In addition, if you are up for trying things eventually you have to be willing to try going to 6 or 8 layer boards so you can try having multiple power planes with sufficient interlayer capacitance, etc. It costs time and money but that's what some of the pros do.
Okay, now you have some opinions from a third person. They are all just opinions of course.
Mark
It sort of sounds like you want to try something that Altor already tried without much to show for it, so he doesn't understand why you would want to do it. Maybe the answer is something like, "Altor, I want to find out things the same way you do, by trying things to see what works for myself."
The thing is if you want to be like that then you need to try multiple design variations and compare them. Building only one dual mono dac will teach almost nothing since there are so many variables not explored and compared.
Anyway, usually the reason for dual mono is better separation between channels, including for noise effects of one channel affecting the sound of the other channel in some possibly subtle way. Such effects can be hard to measure in a very useful/meaningful way if they are signal correlated. That said, things that are hard to measure sometimes can still be evaluated to some extent by a trained set of ears. In that case the rest of the system needs to be up to par so that any differences in sound area easily audible. Good speakers in a good room with good amplifiers can help a lot but it can also cost a lot of money, maybe even as much as an AP 555x.
In addition, if you are up for trying things eventually you have to be willing to try going to 6 or 8 layer boards so you can try having multiple power planes with sufficient interlayer capacitance, etc. It costs time and money but that's what some of the pros do.
Okay, now you have some opinions from a third person. They are all just opinions of course.
Mark
The same is said about using single op-amps instead of dual ones. However, without the mono mode of the DACs and with dual op-amps, the crosstalk attenuation at high frequencies is >110-120 dB, is this not enough?
Alex.
P.S. Remind me, what is the crossover attenuation of vinyl at 10 kHz? 6 dB or 10 dB?
Attachments
I don't know. What does the correlated noise in the spectral line skirts look like?
Anyway, I'm not sure your question explains why you bothered to measure some differences in the first place? Why shouldn't Exetic try such things like you did? Are you essentially telling Exetic, "Do as I say, not as I do."
Why does it matter? My dac already gives better imaging than my much more costly vinyl setup.
That's the thing easiest to measure. What about correlated substrate-coupled cross-channel noise? Again, as humans we tend to think of whatever is easy to measure as being all that exists, all that can exist. The cognitive bias in that case is called WYSIATI.
Which also brings up the issue of what is the difference between correlated-noise and distortion? They can both sound ugly/rough; they can both have a high crest factor too. If its too hard to measure one of them in a meaningful way then maybe its easier to build a dual mono dac and listen?
Last edited:
Dual Mono hardware mode makes an easier power supply arrangement. And PCB is probably more easy to design. For instance?
Why would that not show up in multitone measurement?
WYSIATI applies best to sighted listening evaluation. When blinded it becomes WEICHIA (Without Eyes I Cannot Hear It Anymore).
This is analog dilter Fo in datasheet proposed output circuit.
red is after Cfb, yellow/blue are after 1st stage of OP IV on ballanced output connector.
Green is after bal/se last OP circuit. So the noted Fo are more or less tha same as in the PDF.
But the circuit has some issues and not so good.
.
red is after Cfb, yellow/blue are after 1st stage of OP IV on ballanced output connector.
Green is after bal/se last OP circuit. So the noted Fo are more or less tha same as in the PDF.
But the circuit has some issues and not so good.
.
Yes? 🙂
.
(I am almost sure that the sonic results will noticeable better in mono mode. As in almost every newer, dac chip I tried...)
It maybe correct for the Power Amplifiers, but for the DACs - I'm not agree.
It more the question of how many PCB layers you have 🙂
We talked about channel separation (crosstalk).
Alex.