DAC 2496 (AK4393) DAC KIT With CS8416+AK4393+5532

Since that AK4393/6 dacs are analog balanced signal out dacs, I suggest you to trial avoiding analog/opamp section with tapping analog L/R channels directly from pins. You have to add dc blocking cap if you will tap unbalanced signal from + pins and analog ground.

Blocking cap type and value can be determined during tests. I think 2.7 uF or greater would fit. You have to tap analog ground somewhere close to dac chip.
 

Attachments

  • ak4396-directout-unbalanced.jpg
    ak4396-directout-unbalanced.jpg
    96.4 KB · Views: 3,047
Last edited:
Just remember that I was comparing a modded board to the stock standard board. After mods the AK4396 might come out the winner.

I refer to my previous post ;

..and all this without the components being burned-in & the standard NE5532 o/p chip in situ ?

:D


So far, I can tell you this of the Calyx dac vs my AKM4396/opa627
..the difference is not Night & day ! :)

And later, there'll be a pair of Sowter 9545E in place of the opamp ;)
 
AK4395 Vs AK4396

Was the CS43122/ 4397 implemented the same way with EQ processing, digital level control etc ?
The AK4393/5/6 are all interchangable
The AK4395 has a better filter

Allan

Ok, so I've pulled up the data sheets for the AK4395 / 4396:

The 4396 goes one step further in adding DSD (direct stream digital) functionality (see - Direct Stream Digital - Wikipedia, the free encyclopedia)

Although the AK4395 chip only goes up to 192khz sampling frequency, it appears to have better specs concerning the 24bit filter compared to the 4396 ;

__IC_____________Ripple__________Attenuation
AK4395_______+/- 0.0002db_________110 db
AK4396_______+/- 0.005 db__________75db

I accept specification differences may not always translate thru to real-life experience, but in this instance it could explain the 4395's increased SQ and Allan's conclusion.

I'm ordering an unpopulated pcb an a separate 4395 ic to check this out for myself. If it can give more spacial cues & depth as previously described, then I'm all for it :)
-Andy-
 
__IC_____________Ripple__________Attenuation
AK4395_______+/- 0.0002db_________110 db
AK4396_______+/- 0.005 db__________75db

I accept specification differences may not always translate thru to real-life experience, but in this instance it could explain the 4395's increased SQ and Allan's conclusion.

Agree, I have readen a conversation with an AKM technic during the time of AK4396 just released, he was talking about that its Sigma-Delta modulator was different from earlier models. He sad they spend a lot of research and real listening tests before mass production of AK4396.

Quote:
Richard Kulavik of AKM Semiconductors explained it this way: "This DAC is a large departure from other delta-sigma DACs designed by us and others like BurrBrown, Analog Devices and Cirrus Logic. The AK4396 is an entirely new modulator, pioneered and patented by AKM. It achieves something unique. In the past, many of the old Phillips and BurrBrown parts were R-2R* based products. These older products were looked upon as some of the best. One of the reasons was high frequency noise. In older R-2R parts, HF noise was not present. In all delta-sigma parts prior to the AK4396, everyone has fought HF noise caused from the delta-sigma modulator with the insertion of large filters and other parts to attempt to solve a problem created by the delta-sigma design. The AK4396 today effectively does not suffer any modulator-induced HF noise and is over 60dB better than the nearest Cirrus and BB devices. All of this HF noise can cause many audible artifacts downstream. That is the 'miracle' we believe is making the difference today. This part gives you the performance and linearity of a delta-sigma device with the noise performance of an R-2R part, something that was never previously available." (In the professional markets, AuzenTech's X-Meridian 7.1 and Sondigo's Inferno sound cards for example each run four AK4396s "for DVD-A/SACD level fidelity for professional audio applications.")

AK4395's better filter specs may be an answer for this HF noise, mentioned above. BTW, I haven't any experience with AK4395, but I'm looking forward for it.

I'm ordering an unpopulated pcb an a separate 4395 ic to check this out for myself. If it can give more spacial cues & depth as previously described, then I'm all for it :)
-Andy-

Are you ordering separate AK4395 from same seller?
 
Last edited:
Allan - Does the 4395 really have the better filter compared to the 93 / 96 ?

Erin - have you had similar experience of this ? I'm willing to accept that a new/higher chip may not necessarily lead to better SQ.

-Andy-

Hi Andy sorry for the late response.

I was looking at the spec before and wondering myself of, how much difference between the 95 and 96 would be.
Besides the digital filter, most other spec the 96 is less than 3db better, but implementation could negate most of these anyway.
Looks like the 96 is a later update with DSD capability.
So hard to say the diffeence between them.

I have a 4393 pcb and i'll change the dac to a 4395
From the 4396 datasheet all the 93/95/96 can do 3.0 to 5.25volt digital.
 
5V

Hi Andy sorry for the late response.

I was looking at the spec before and wondering myself of, how much difference between the 95 and 96 would be.
Besides the digital filter, most other spec the 96 is less than 3db better, but implementation could negate most of these anyway.
Looks like the 96 is a later update with DSD capability.
So hard to say the diffeence between them.

I have a 4393 pcb and i'll change the dac to a 4395
From the 4396 datasheet all the 93/95/96 can do 3.0 to 5.25volt digital.

The AK4395 won't reliably run on 3.3V. That is why it is seldom used even though it sounds rhe best.
 
It finally arrived!

Today in the mailbox I've found a small packet with an R-Core transformer and an assembled nice blue AK4396 DAC...

I must wait this night to test it but I've managed to check my previous work and found some errors...

Attached are the updated PCB paths and possible BOM for updates

The BOM includes Erin suggestions (Erin can you check it?)
 

Attachments

  • Kit 800.jpg
    Kit 800.jpg
    62.8 KB · Views: 2,263
  • PCB - Percorsi.jpg
    PCB - Percorsi.jpg
    469.5 KB · Views: 2,526
  • BOM.pdf
    8.9 KB · Views: 590
Clave, good work!

There are one or two variations. However I don't see any need to make corrections as its a rough guide. Hopefully others will experiment to get a sound they like.
It's highly likely building the kit using my suggested parts will yield a very smooth sound which may be too smooth for some peoples tastes.
Let me know what you think of the sound when you get around to it.
 
Another one to consider..?

Sorry to put the cat amongst the pidgeons here, but I've received a new dac board, again which is pre-populated.

It uses the AD1955, has spdif elec/toslink inputs and also a usb input.
I quickly wired it up last night & have left it to burn in. I gotta admit... it sounds damm nice !
This little baby was sourced from the same place in china where I bought my 4396 dac kit from. Having said that, it did cost more (£51).

Hell, it's cheap to get sounds this good at this price point.
As for the AKM4396, I'd willingly pay £400+ for that sound quality... :D