ebay:Weiliang Dual X2 AK4399 DAC with LCD

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Valéry in same way I agree with you. Infact it seems don't get any advantage from High quality USB cable.
I think an advantage is possible by using in these weiliang xmos an high quality cristal oscillator. My friend use a battery powered xmos usb to coax with three high performance cristall oscillators and he is very exicted by result, more than the internal weiliang xmos. Any way I will experiment the Dxio pro3 xmos interface to see if nothing will change or not.

Inviato dal mio SM-N9005 utilizzando Tapatalk

Hi Keres,
I believe that quality of USB cable does not make any impact on sound quality. PC asynchronously sends data via this cable into buffer of XMOS module. Any cable will serve well. Quality of power supply, oscillators and SPDIF cable can make big impact on jitter and consequently on sound quality. I can experiment only with battery power supply because it is very hard to change oscillators on my module.
 
Last edited:
Replacing the unknown capacitors in the feedback loop gained a large amount of clarity and depth, were nearly the last caps changed on the entire board, may have just been the weakest link possibly, but for the $2.80 that the surplus wimps cost, it was well worth it.
I can hear the ice cubes clinking in the glasses now on some of my live recordings made in clubs.

At least you know the designer played around with the arrangement instead of just duplicating the data sheet with the resistors on the dac chip.

Would your experience indicate that modifying the caps directly after the dac ought to be the first step?
 
Esgigt, those 10uf that are near the dac would be worth experimenting with trying something other than the silmics, not sure if would be the first thing to try though.

Those capacitors are so very close to the dac chip that it's easy to cook the dac from too much heat through the traces, as I have found :(

My more recent attempt involved snipping out the existing caps, and just tacking on the replacements, without heating things too much. I used some large 100uf panasonic pureism that were surplus just to try it out, and have left them in since they sound so much better than what was in there. I would imagine that something around 47uf would also be fine.

My dac is starting to look like a donut with sprinkles, with all the capacitors hanging here and there.
 
Has anyone info about how treat analog outputs of DAC in mono mode with an officially confirmed technique?

AK4399 datasheet shows an output section (page 37) for stereo mode which differs from actual values on this kit. For example, how they did decide to use 11K resistors on each legs of analog outputs whereas datasheet shows 1K5 resistors.
The technique used is pretty official I think... There's no difference to stereo mode. All that matters is the resistance of the two 11k parallel (5k5). The values for the resistors and capacitors are chosen according to the desired gain, cut-off frequency and filter characteristic.
btw: the resistors are 2k4 in the figure, not 1k5.

I don't know what your plans are, but I wouldn't mess with the filter values. I proposed an alternative filter, which I think is is better than the one used, see my post in this thread.


One thought about the local decoupling of both DAC ICs: I think the used 47u is just fine, although other brands/types may positively influence sq. Keep in mind the main purpose of these caps, relative big currents still have to be delivered by the regulated psu, some centimeters away from the load.
 
The technique used is pretty official I think... There's no difference to stereo mode. All that matters is the resistance of the two 11k parallel (5k5). The values for the resistors and capacitors are chosen according to the desired gain, cut-off frequency and filter characteristic.
btw: the resistors are 2k4 in the figure, not 1k5.

Official ak4399 datasheet shows 1.5K resistors. I just wonder that how did they choose such high value (11K of each output). Legs are carrying same signal in mono mode but they aren't connected internally. I look at datasheet for confirming such implementation but there is no any info about how to treat analog outputs in mono mode.

I also assume that the filter circuit's actual role is the balance-unbalance conversion. Since AKM dacs uses an unique sigma-delta implementation which avoiding hi-freq disturbance at analog outputs, analog stage's actual role becomes more critical in these conditions. If you can find a more simple and efficient way for doing balanced-unbalanced conversion, you probably will get higher quality sound. This is why transformers doing very well on AKM dacs.
 

Attachments

  • ak4399-output.jpg
    ak4399-output.jpg
    13.8 KB · Views: 474
Last edited:
Strangely you have another datasheet than me, but anyway: On page 8 of the datasheet max load capacitance and resistance is defined. As long as the load on the analog output doesn't exceed 1k5, you can choose any value you want. As I explained, this value is dependent on the filter properties you want. In your example cut-off frequency is 99.2 kHz, gain is unity.

It's true that the SCF outputs are unique and don't require extensive filtering. But in all AKM examples at least 2nd order filtering is used with cut-off at about 125 kHz. If filtering (according to AKM) wasn't needed and just conversion to single ended, you could leave the capacitors and use the op-amp as a differential buffer amp, which is not the case in the example circuits.

I think 11k resistors on each output is used (instead of one 5k5) to avoid connecting both + outputs directly. Because the DAC starts up in stereo mode by default, it's not guaranteed both + outputs have the same potential, which could lead to short circuit. This way it's safe.
 
I think 11k resistors on each output is used (instead of one 5k5) to avoid connecting both + outputs directly. Because the DAC starts up in stereo mode by default, it's not guaranteed both + outputs have the same potential, which could lead to short circuit. This way it's safe.

I'm seeking such explanations in order to get an idea about designer's opinion. I sent a message to AKM from website but didn't get any answer yet.
 
I would be repeating myself how amazing those HDAM modules are so I'll be more than happy if someone else could buy it and test it otherwise I will look like an idiot who can hear weird stuff :D

+ 1.000.000.000
Simply.... hdam are the best in this dac, very nice dinamic, analog, and coerent sound. What to say .... Thanks ... Kukynas

Inviato dal mio SM-N9005 utilizzando Tapatalk
 
direct-out mod

Hello.How can i do direct out mod with this DAC? Thanks in advance.

Hello, I made a sketch in order to modify analog output for direct-out mod. This is just one of the methods to isolating dac outputs from analog circuit.

You can do that also with removing each 11K resistor's opamp side legs and then tap the positive signal output from floating legs of 11K's (legs are connected again). In this case, you don't have to remove many parts from analog stage. 10uf film capacitors are mandatory if you want to prevent DC going to preamp or amplifier.
 

Attachments

  • 4399mono-directout.jpg
    4399mono-directout.jpg
    60.8 KB · Views: 436
Hi Terranigma, what are the crosses in your photo stand for? Removing these parts from the PCB?

You should somehow isolate dac outputs from analog circuitry. I look at schematic, I saw that those ones connected to the signal path and then they have to be removed in order to tap signal from pcb.

I think easier way is removing upper legs of 11K resistors and then tap signal from disconnected legs. Each corresponding leg must be tied again in float position. In this case, you don't have to remove any part from pcb.
 

Attachments

  • 4399mono-directout2.jpg
    4399mono-directout2.jpg
    40.9 KB · Views: 412
Last edited:
Hi Kastor L,

Thanks for your explanation and the link.

If your AK4399 DAC hasn't got software control, the DAC works in parallel mode and the digital filter is the sharp roll-off one. Only in serial mode (controlled by a micro controller) the minimum phase (or short delay) filter can be set.

Hi Nino Simona,

I'm interested in this 2x AK4399 now, since I feel like an improvement from my 1x AK4399.

I'm just wondering if I buy it, how do I change the filter setting?

I need to buy a micro-controller right?
 
Last edited:
Hi Kastor L
You don't need to buy a micro controller, it comes with the 2xAK4399 DAC. It's located on the display PCB.

The DAC has 3 buttons on a separate PCB, but only 2 functions: One button switches between the 2 filters, the other two switch 3 digital inputs (spdif/toslink/usb).

If you want to improve the +5V analog psu of this DAC, better buy first version with AK4113 receiver.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.