DAC AD1862: Almost THT, I2S input, NOS, R-2R

@ClaudeG ,

Hard to tell w/o having trying it in your setup. Here the sound is more different than similar : it is less than 24 bits and it is NOS, so in theory less resolved if you have 24 bits materials and more "polluted" cause the alliasing. But has also ears after 50 are notching more and more I would be curious to benchmark both.

What you hear at 25 is not what you hear at 60 for illustration... Is digital image can spread back till 1 K hz ? I can't remember.

Here with Miro1360 shematic, you haven't feedback 6 db filtering, so it is another story more, op amps have more so so behavior because of that and lack of postfiltering and peaky RF audioband of some op amps (fastest) can go back like a boomerang in the audioband, making even some false feeling of ligth or brigther upper mids and trebles.... so hard to tell and compare apple and orange.

But at least you have enough material in this thread to try and have differentoptions, compare and enjoy. If you ask me, for the price of it and the fun, I would try. For the QP I don't know as if you stack several designs it migth cost more than some good modern SD off shelves if not high end. Hard to tell what your equilibrium is in that ask. Good R2R seems even more priced, so why not if R2R sounds is liked by you.

If you don't have high res or don't stream in high res web music portails, I will give it a try and focus on the AD1862.

hope that helps, but anyway the quest beginns with the find of the dac chips...

A SU-10 will cost you less than a basic AD1862 conf. Or very near with casing. Courtesy from EUVL : https://fr.aliexpress.com/item/1005007250532607.html?gatewayAdapt=glo2fra
 
Last edited:
Hi diyiggy,

Thanks for your kind and detailled response

Indeed I forgot to mention: CD format only - coming from low jitter CD Player through a coax cable - so no high resolution has to be considered in this case.

The thing is I am very happy whith my DAC but I like tweaking as every DIYer. I could update my DAC's output stage - the most logical step and perhaps the best improvement path - but it would require me designing a new board etc.. So...While being at it anyway, I thought why not starting from scratch with better parts and knowledge and go the route offered in this thread? But then it really HAS to sound much better than my own DAC as otherwise I would be very disapointed. And air gets thin up there... as said, the friends that came with their modern DACs at mine drew also the conclusion mine was still the killer. Having said that, the Eversolo Z8 is though a nice DAC, starting to get serious sound.

TBH, the Ares 12th was tried in all configurations (phase, OS /NOS, 2 filters for OS), with CD material only, and regardless the setting it was beaten on all aspects (mainly on treble extension, lower bass extension, micro and micro dynamic, precision, speed and stability on complex orchesters) but on one where it shone vs my current DAC. The Ares in OS Slow mode (less so in NOS mode although I prefered that one slightly overall), and also for example the Eversolo Z8, do a better job in terms of soundstage (depth and precision/location of instruments, non sterile "air" between instruments). Trademark of R2Rs one might think, but then some ESS DACs do that aswell (SU-9 wasn't bad on that one). I know that changing my outpustage would slightly improve on that but also that it is probably mainly addressed with a new DAC chip.

So I either stick to what I have, or upgrade my output stage to bring it up to date... or go for a brand new design as the ones offered here. But the new design has to be a majour improvement in terms of musicality. I have some doubts, especialy after the Ares 12th I just tried (and defo 25y ago vs the Audio Note 3.1 signature I tried) that "any" R2R can really bring me some sonic improvement. As I can't try them easily, I took the Ares as benchmark for whatever it is worth in hope someone here had done somes comparisons with it as it seems quite a common commercial DAC.

The question stands: is the Ares 12th on a par with the builds discussed here or are these builds a significant step better? Anyone?
Please 🙂


Many thanks in advance

Claude
 
@ClaudeG
"hand-matched 0.01% high-precision resistors"

For N-bit DAC you need accuracy better than 1/(2ˇN) x 100.
For 8-bit DAC: 1 / 2ˇ8 x 100 = 1/256 x 100 = 0.4%
For 16-bit DAC: 1 / 2ˇ16 x 100 = 1/65536 x 100 = 0.0015%

... Discrete R-2R DAC with 0.01% resistors is precise in reality only as 13-bit DAC ... is that enough? Well, it can be, but there are other factors to consider ...

From my testing, the most important thing in R-2R are switches. This makes the sound different 🙂
TDA1540 is 14-bit DAC and it sounds well.

Focus on the switches used in that DAC - is that something advanced and immune to digital-analog noise?
You can buy it and test it ... test it on low audio volume - here you can hear how good switches are 🤣
But from what I see, it has 74xx595 registers which also function as switches ... is that good enough? From my testing no, but only your own testing and comparing will say more 🙂
 
13 bits.... and if you add pcb stray capacitance, inductance, dust and moisture.... it is less....🤣
It is a little like a washing machine sold 6000 rpm that drop 4000 RPM 3 months after !

But it is rectified with cpld, isn't it ?

ClaugeG doesn't need a new DAC, as it comes from the worse front end that is cd players. He needs a buffer, an upsampler and reclocker like the gadget from Chord or others if wanting to stay CD player to go further. Some DAC when well made do that as well....

99,9% of the guys here are on USB with digital library. Any benchmark will not tell you...
 
Last edited:
Do you remember the resistance and capacitance values you used inthe feed back loop, please ?
Standard values not so high Voltage output, with rc parallel to Rfb
Its input resistance is monstrously high but the AoL is high... did you model its impedance bandwith ?
it is alone because of the combine jfet input, out of the circuit... in transimpedance circuit it is low
Did you noticed a subjective lack of bass or at least tigther/more damped ?
on the contrary, very very good LF
 
  • Like
Reactions: diyiggy
For AD797 in any case, 47pF from pin 8 to Out is must. Also Decoupling as stated in datasheet is impotrtant. Off set null is option but if i remember well, it is not an issue, anyway it should be existing as option.
Sound is very good and unique. Nor as usual OP amp sound.
.
I was maybe more satisfied with AD8066, but this amp is in class A and cant stand alone without heatsink noted PS values. So I decresed to +-8V. After that was more than good... Consider this AD8066 seriously.
So 47p between 6 and 8 pin? Is it together from the diagram from the posted note by diyiggy?
 
I fully understand the argument that R2R with resistors and Co aren't the best solution. But that's what I had on hand... and that Audio Note 3.1 signature - with its Analog Device chip this time - didn't shone either 25y ago, so cold water... eventhough I understand perfectly that here we are talking about a completely different unit, hence my question.

Anyway, just to "translate" what has been said: are we saying what I tested isn't representative of what you guys achieve with AD1862 chip DACs? More specificaly that your AD1862 DACs sound significantly better than the Ares 12th? I would love that! But please confirm (if you know it).

Regarding CD Player, mine is quite special and low jitter - sounds equal to better than when my DAC is fed per USB, but that's not the point (and yes good point my DACs need a dedicated input buffer etc.).

What I need to know is "has anyone benchmarked these DIY DACs vs some known commercial entity?"

If not, no worries, I understand the fun of building units and if you follow me you know I am part of all that since ages.

Thanks for your replies

Claude
 
Last edited:
I saw the shunt regulator schematic shared by Vunce in post 7343, that gives an output of +/- 12V. Is it possible to tune that design to get the value +/- 5V output?
 
  • Like
Reactions: fergas
What I need to know is "has anyone benchmarked these DIY DACs vs some known commercial entity?"

The conundrum is, as always, there is not any benchmarking metric that directly correlates to "smiles per minute" and "musical enjoyment".

Are measurements important? Of course!

...but no matter how much I'd like it to be false, the higher-numerically performing DAC on the ASR list sound further and further from musical, to me. I've got 2 DAC in the blue ranking (above 111 SINAD) and although they excel at the audiophile stuff, and are gloriously quiet, overall they leave me wanting, and I rarely listen to them for very long.

I have no idea how one of these AD1862 DAC would measure, and don't honestly care much, as they are the first DAC I have owned in a while that I have actually looked forward to listening, the last was a Meridian from a very long time ago that I probably remember more fondly than is realistic, but compared to any I have now, I like to listen to the AD1862.

Will this be your experience if you try it? No idea. And who knows... I may have tin ears and not any clue to what I'm talking about, some in this thread will agree with that. But If you are curious as to what these DAC can bring, I highly recommend trying them.
 
Like in the datasheet. (once i was ommit to solder this C on the pcb, where i already put the part, and spend a lot of time until i measured right result - after soldering this little C...
Decoupling - also as in the datasheet.
Cheers
I use such an IV for DAC Tda1541. It's not NOS. As you can see, there is no capacitor here. However, don't ask me about the details because I'm just playing with building modules invented by smarter than me 🙂
 

Attachments

  • IMG_2280.jpeg
    IMG_2280.jpeg
    632.9 KB · Views: 82
For sure if you want to smell honey in your mouth you have to put the fingers into the honney pot.

Méridian had well regarded devices whatever what were inside, they know their stuff and idem with Linn...

We can discuss tyres and drive feelings around à circle of guys whom drive the same car , but if you don't buy the car, you will have no idea how your Hummer is! (translated Rabitt hole notion into US litterature motion)... whatever you can find guys to say they can not benchmark the black of their tyres with yours...
 
Last edited:
  • Like
Reactions: Sorenm and 6L6
Put a stone mark on your agenda, I am rarely agree with Brijac, but on that about Soekris vs a well integrated R2R dac chip, I agree ! :hphones:

I am not sure the Ares is a pure R2R, all that modern discrete DAC are cpld assisted, so more hybrids.

But about the PCM63 : it is purely a question of taste and marriage with the rest of your hifi chain. I subjectly think PCM 1704 and PCM63 are not as good as ad1862. The PCM56/58 were good. PCM1702 is on the clear/ligth side, you may like it or not, anyway needs to be coocked as all the DAC chips, they don't sound the same according what is around them on the pcb.
 
Last edited:
  • Like
Reactions: ClaudeG and 6L6
Hi Jim 🙂

And thanks for your kind feedback!

Not sure though my message was clear... I am as you not interested in measures at all. Have been there, had lots of "perfect per ASR" DACs to try at mine and friends during Covid and recently and... clearly heard how different they sounded despite being all well "inaudibly perfect" as top SINAD ranking.

I am only looking for listening impressions, of course well bearing in mind that those impressions / findings might indeed very likely be very different from mine. I hope I have learned after a few dacades to read between lines and get a rough idea - For me it is Go or no Go. Have enough projects on hand, no hurry, but that one intrigues me LOL.

So doesn't seem easy, argh! Thing is I am somewhere tempted (although I don't need any additional DAC), but really VERY suspicious on the other hand. My bad no one has one nearby, at least I guess I was lucky with my neighbour having amazingly an Ares down here 🙂

Many thanks again... and don't forget to land at the Bourget one day ;-)

Claude
 
  • Like
Reactions: andyjevans and 6L6