tiny SMD soldering
I managed to solder tiny 0402 (1.0x0.5mm) smd X7R caps right on the pins of the AK4399 chips, as suggested by paolo. I found this a good idea, because the position of the 100n film caps on my board was pretty far from the pins, and the film caps were also too large being that close to the dac. So I removed them and placed the smd caps instead.
I'm very happy with the result, the sound has become more refined and 3d picture has improved as well.
The hand soldering is not easy, but I have a few tips. First apply some extra solder to the pins on which an smd cap must be attached. Then apply plenty flux and delicately position the smd cap on the two pins, as close to the chip as possible, leaving still some part of the pad free. When positioned the cap well (I used a needle & patience), take the iron and quickly heat one pad, allowing the solder to melt and attach to the cap as well, followed by the other pad. Don't touch the smd cap itself, or it will stick to the iron. Use of a magnifying glass on stand is a must.
Nino
Hi guys,
I managed to solder tiny 0402 (1.0x0.5mm) smd X7R caps right on the pins of the AK4399 chips, as suggested by paolo. I found this a good idea, because the position of the 100n film caps on my board was pretty far from the pins, and the film caps were also too large being that close to the dac. So I removed them and placed the smd caps instead.
I'm very happy with the result, the sound has become more refined and 3d picture has improved as well.
The hand soldering is not easy, but I have a few tips. First apply some extra solder to the pins on which an smd cap must be attached. Then apply plenty flux and delicately position the smd cap on the two pins, as close to the chip as possible, leaving still some part of the pad free. When positioned the cap well (I used a needle & patience), take the iron and quickly heat one pad, allowing the solder to melt and attach to the cap as well, followed by the other pad. Don't touch the smd cap itself, or it will stick to the iron. Use of a magnifying glass on stand is a must.
Nino
Attachments
Last edited:
make sure that the red led on the usb board is light up when connected to your computer
if the red led not glowing either the card is wrongly seated or the card is broken
if the red led not glowing either the card is wrongly seated or the card is broken
Separate power supply
Hi Nino, very well job.
At first, I also wanted to go that way too. But then I realized that it’s needed to bring the power source to the power consumer as close as possible . Now in my new DAC (AK4495) I have 4 additional voltage regulators that are connected directly to the appropriate capacitor pins (length of Connection wires ca. 10mm). I use REG101 chips (two REG101-5 and one REG101-3.3) for VREFL,VREFR and AVSS. I have cuted the PCBs for these regulators from the evaluation board s(DEM-SOT23LDO). The modified PCB has the size ca. 10mm x 20mm. For the I/V stage (VDDL and VDDR) I use a single ultralow noise voltage regulator TPS7A4901. I have cuted the PCB from the appropriate evaluation board too. The modified PCB for TPS is not as small as for REG, because the ground foil is used as additional power dissipatior.
Before this modification the sound was very good, but now it is FANTASTIC!
Hi Nino, very well job.
At first, I also wanted to go that way too. But then I realized that it’s needed to bring the power source to the power consumer as close as possible . Now in my new DAC (AK4495) I have 4 additional voltage regulators that are connected directly to the appropriate capacitor pins (length of Connection wires ca. 10mm). I use REG101 chips (two REG101-5 and one REG101-3.3) for VREFL,VREFR and AVSS. I have cuted the PCBs for these regulators from the evaluation board s(DEM-SOT23LDO). The modified PCB has the size ca. 10mm x 20mm. For the I/V stage (VDDL and VDDR) I use a single ultralow noise voltage regulator TPS7A4901. I have cuted the PCB from the appropriate evaluation board too. The modified PCB for TPS is not as small as for REG, because the ground foil is used as additional power dissipatior.
Before this modification the sound was very good, but now it is FANTASTIC!
Last edited:
No, if you mean the original PCB from Weiliang.
It is possible to modify this board, to add support of DSD, but in this case it’s necessary to switch the DAC to the software control mode using, for example, an Arduino MC. Additionally you need an UCB receiver (XMOS) that supports this format. Using the USB receiver from Weiliang is not a good idea, because it converts the DSD stream to the SPDIF.
I have no plans to add support of DSD in my AK4495 PCB, but for that I have already purchased a small board with AK4490 and the appropriate XMOS PCB from diyinhk.com. The DAC AK4490 is at least as good as the DAC AK4495.
Last edited:
Hi pca01,
Regardless of nearby regulators, which is well done of course, I think it's worth the effort to place the 100n smd capacitors, because their main purpose is to decouple HF noise, generated by the dac itself, to ground.
By doing this, my dac's sound has become less sharp edged, so better highs, refinement. I'm pretty sure nearby placed low noise regs improve sound further, but I think the small caps are most effective for HF decoupling.
Could you please post some pics of your AK4495 board including the regs/evaluation boards you used?
Regardless of nearby regulators, which is well done of course, I think it's worth the effort to place the 100n smd capacitors, because their main purpose is to decouple HF noise, generated by the dac itself, to ground.
By doing this, my dac's sound has become less sharp edged, so better highs, refinement. I'm pretty sure nearby placed low noise regs improve sound further, but I think the small caps are most effective for HF decoupling.
Could you please post some pics of your AK4495 board including the regs/evaluation boards you used?
Hi, Nino
OK, some pictures.
On the third photo is an old (large) version of 5V regulator. The last photo displays the current (draft) state. I would like a little bit reduce the size of PCBs and replays the wires.
I've already installed a 4-pins connetor to the DAC PCB.
OK, some pictures.
On the third photo is an old (large) version of 5V regulator. The last photo displays the current (draft) state. I would like a little bit reduce the size of PCBs and replays the wires.
I've already installed a 4-pins connetor to the DAC PCB
.Attachments
The problem with Weiliang XMOS is not only that it converts DSD stream in SPDIF, but mainly that it does it SO BADLY!!! In facts, by using it, DAC's SQ is much lower than using the external SPDIF connector. And it doesn't accept DSD input, anyway.
But i replaced it using a JLsounds I2SoverUSB board, connecting it to Weiliang SPDIF internal connector (the one used by their XMOS), and the USB sound quality improved VERY much: SQ is now just a bit lower than connecting it in I2S to the DAC (i tested that too), and with much less problems. It plays well DSD files, too, using DoP.
So Weiliang XMOS board is a real crap, but you don't strictly need to go I2S to have a very good sounding USB input, even if I2S is the real best.
You can find a description and some pics of my mods back in the thread.
Hi Luca72c,
I have purchased this board
XMOS DSD DXD 384kHz high-quality USB to I2S/DSD PCB - DIYINHK
This board suports the DSD streams 2.8224MHz (64fs) and 5.6448MHz (128fs), the
AK4490 additionally supports the 11.2896MHz (256fs).
It is possible to use the volume control for both PCM and DSD, but in the last case it's possible to bypass this control.
I have purchased this board
XMOS DSD DXD 384kHz high-quality USB to I2S/DSD PCB - DIYINHK
This board suports the DSD streams 2.8224MHz (64fs) and 5.6448MHz (128fs), the
AK4490 additionally supports the 11.2896MHz (256fs).
It is possible to use the volume control for both PCM and DSD, but in the last case it's possible to bypass this control.
Last edited:
I own that board too. That's the board i used to test direct I2S connection to the DAC and results were very good. But i couldn't use it connected to AK4113, because it doesn't have a SPDIF output.
Direct I2S connection using Weiliang controller is a bit unstable, with scratchy sounds on certain files, and it would be complicated to keep AK4113 switching capabilities. So i choosed to loose a tiny (but really tiny!) bit of SQ to keep the ability to switch signals and not be forced to replace the controller (something i absolutely am not able to do).
Hi folks!
I just recieved this dual AK4399 DAC from diyhifishop.
The main filter caps aren't Philips as expected, but Nichicon Gold Tune. Don't asked this way, but no problem.
Every website, I have seen by seeking this product shown a slightly different fixtures.
I am not amused by the BC caps in general, so I'll go to try some changes around the opamps and DAC chips.
My first mod was to change the original SB560 schottkys to BYV27 ultrafast/soft recovery. After this action the sound quality has grown up at least one class.
I just started to read this topic from the beginning, but if anybody has a good recipe for upgrade, my ears are open!
I just recieved this dual AK4399 DAC from diyhifishop.
The main filter caps aren't Philips as expected, but Nichicon Gold Tune. Don't asked this way, but no problem.
Every website, I have seen by seeking this product shown a slightly different fixtures.
I am not amused by the BC caps in general, so I'll go to try some changes around the opamps and DAC chips.
My first mod was to change the original SB560 schottkys to BYV27 ultrafast/soft recovery. After this action the sound quality has grown up at least one class.
I just started to read this topic from the beginning, but if anybody has a good recipe for upgrade, my ears are open!
Attachments
Last edited:
pca01: Thanks for sharing pictures!
megaymir: Welcome to the thread. In what way did the sound of the DAC improve when diodes were swapped to BYV27? They are fast recovery, as well as Schottkys.
If you're looking for a good mod for this dac, I would recommend to do the TVDD mod first.
Nino
megaymir: Welcome to the thread. In what way did the sound of the DAC improve when diodes were swapped to BYV27? They are fast recovery, as well as Schottkys.
If you're looking for a good mod for this dac, I would recommend to do the TVDD mod first.
Nino
Hola hi привет ciao γεια σας.
Starting a project based mainly on your development, monsieur NinoSimona.
The diff is that I'm using two AK4397 and trying to parallel them. Not a trivial task, taking that they are not mono-enabled. So - either cross-wiring (not that awful idea, considering common ground on Weiliangs v2 board), or i2s splitter for 2 chips.
Some more info later.
Starting a project based mainly on your development, monsieur NinoSimona.
The diff is that I'm using two AK4397 and trying to parallel them. Not a trivial task, taking that they are not mono-enabled. So - either cross-wiring (not that awful idea, considering common ground on Weiliangs v2 board), or i2s splitter for 2 chips.
Some more info later.
Hi, Nino
OK, some pictures.
On the third photo is an old (large) version of 5V regulator. The last photo displays the current (draft) state. I would like a little bit reduce the size of PCBs and replays the wires.
I've already installed a 4-pins connetor to the DAC PCB
Can you post the schematic?
Hi cu6apum,
Are you using original controller? What changes did you make on v2 pcb, besides DAC chips? Did you assemble the DAC yourself? Why are you using AK4397? This problem (no lock) could have many possible causes.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.