[londere]

After some mods made to my ak4499 NOR I noticed that the voltage generated by digital board was
+/- 7V and not 9V any suggestions on what I can check? On Dual Gad I had +/-12V without mod, in advance thanks

 

[Merlin-PT]

@londere

I don't have the ak4499 model, so this can the wrong, but based on what I see in the randomly googled photo I attached, I think this ak4499 model opamps voltage is +-7V.

The ak4497 that I have has the same resistors positions in that MP1542 circuit, so I assume the two resistors that setup the MP1542 voltage to be in the same position, but if you want to be sure, you need to read the MP1542 datasheet and trace in your PCB to see if the resistors I marked red are the R1 and R2 in the datasheet circuit.

I marked in RED the 2 resistors that (I think) sets the voltage for the MP1542 boost convertor:
R1 = SMD resistor marked 224 = 220K
R2 = SMD resistor marked 473 = 47K
Vout = 1.25 × (1 + R1/R2) = 1.25x(1+220/47) = 7.10V

 

[londere]

@londere

I don't have the ak4499 model, so this can the wrong, but based on what I see in this randomly googled photo, I think this ak4499 model opamps voltage is +-7V.: https://twitter.com/mustdidreamer/status/1325093638796271616/photo/1

The ak4497 that I have has the same resistors positions in that MP1542 circuit, so I assume the two resistors that setup the MP1542 voltage to be in the same position, but if you want to be sure, you need to read the MP1542 datasheet and trace in your PCB to see if the resistors I marked red are the R1 and R2 in the datasheet circuit.

I marked in RED the 2 resistors that (I think) sets the voltage for the MP1542 boost convertor:
R1 = SMD resistor marked 224 = 220K
R2 = SMD resistor marked 473 = 47K
Vout = 1.25 × (1 + R1/R2) = 1.25x(1+220/47) = 7.10V

thanks for your help @Merlin-PT as you mentioned these are the correct values for 7.10V, On GAD Dual AK4497 they use a 10Kohms and 92C SMD Resistor (searching on the web are stated to be 88.7Kohms <=1%) and the datasheet stated on page 7 formula R1=R2(Vout-Vfs)*(1/Vfs) = 10K(12-1.25)*(1/1.25)=86K but I see on page 4 the worst characterization on efficiency in this config.

So the question is: is worth mod the dc converter to up 2V? In my case, I had the op1612 as gain, LPF op275 as roll muses02(not verified), muses 8920 (not verified), and ne5555

 

[Merlin-PT]

Regarding your question, read here:
https://tangentsoft.net/audio/opamp-wv.html

In your case, you can't increase the opamps supply voltage, just to make a trial and listen to it, only because we don't know what's the rated voltage of all the capacitors they use in the opamps power supply lines. You risk damaging those capacitors if you exceed their rated voltage.

In my case, before I raised the opamps power supply voltage, I had to replaced all the capacitors in the opamps power supply lines, to make sure they supported 12V (I used caps rated at >=16V).
With the opamps I had at the time I didn't noticed any difference in sound, but I wanted to buy and roll some opamps I had read would sound better at +-12V and then had to fix the voltage because my microscopic LDO done without PCB doesn't allow changes.

BTW, for future use I bought a generic prototype PCB to solder that LDO and it's now very easy to solder it:
https://aliexpress.com/item/33053018423.html

 

[londere]

Update: After change the R1 to 330K I got the 9V stated by a zishan to be precise 9.15V and -9.02V (by the way I don't know why is this difference). After the mod the heat over the PCB have been raised significantly. Even after played 1 song!!! I think this is the reason of zishan to limit the voltage to +/-7V I added a little sheet of cooper (sell by Coolaboratory) as you can see in the pictures.

About sound I notice a wider punch on my headphones (Onkyo HF-300 32ohms) at 30 volume value using as gain amp OPA1612.

 

[maxifuny]

Well it worked for my ES9038 player. I don't know, I asked multiple times at NiceHKC to provide 2 files as a combined package, and they just don't understand. This is all I got.

When I install 0.5c on my player there is no sound, with 0.5d sound is back.

Push up. Take a look at the menu on the right. If there is one, just switch to normal play mode - all

*Provided the correct firmware is available.

 

[raabje]

Yes I tried that too, in the volume and EQ section, the option one, folder, all, etc.

When a track plays and is finished, the next track just stops. If I press fast forward (= hold the button a bit longer), then the track plays. Or sometimes white noise is heard. If I press next song, the next song plays. But when it is finished, it stops again. On my CS version it all works normal. I have made a short video of the problem:

https://streamable.com/my6frf

I have send the DSD with ES back. Would love to hear from others their experience with the DSDs ES version with firmware 0.5d. If I was just unlucky, I will order a new DSD with ES after my refund.

 

[Merlin-PT]

@londere
Nice, mine from 9V to 12V also heats more, maybe I'll revert it back to 9V since I never bought those opamps I wanted to try at 12V.

From your photos I can't see how you are using the "Coollaboratory Liquid MetalPad – Notebook". You only using the copper sheet or also the thermal compound to stick the copper sheet to the PCB? Do you also isolate the batt. from the heat? I can't see how it was done.

If it fits the space, one idea is to use a bigger copper sheet, so half of it would make contact with the PCB back, then it would fold around the batt and make contact with the metal box, to try to remove the heat from the PCB at the dac back, into the outside metal box case. The battery should be thermal isolated from the copper, but I don't know if it will fit.

 

[Setmagic]

Mod inspired by @maxifuny
And resistors Vishery Metal for audio 2,2Ohm

 

[maxifuny]

Mod inspired by @maxifuny

a little more attentive to the denominations of the elements

 

[londere]

From your photos I can't see how you are using the "Coollaboratory Liquid MetalPad – Notebook". You only using the copper sheet or also the thermal compound to stick the copper sheet to the PCB? Do you also isolate the batt. from the heat? I can't see how it was done.

@Merlin-PT My bad, I meant to use the cooper sheets from the kit I use the thinnest one. Really I did not isolate the battery only used a binding sheet (in the assumption: it's made of plastic). For the compound maybe add some computer thermal paste such as cooler master mastermaker or artic silver 5 or noctua h2, (do not add liquid metal sheet it can destroy your PCB traces)

a binding sheetsIf it fits the space, one idea is to use a bigger copper sheet, so half of it would make contact with the PCB back, then it would fold around the batt and make contact with the metal box, to try to remove the heat from the PCB at the dac back, into the outside metal box case. The battery should be thermal isolated from the copper, but I don't know if it will fit.

since I don't have any copper sheet I can't do it the only contact I do was to the side (where is the rail that the PCB slides) also thinking, with one sheet of binding and the cooper it barely fit also thinking to change to an acrylic case but no money at the time... Or maybe drill holes to the back of the case.....

Any idea how to isolate the battery?

 

[Merlin-PT]

@londere
I would avoid the thermal compound only because it would make a mess, or if needed use only a very small bit at the pcb back position at the dac location, to easily extract the heat from the pcb at dac back location.

Google or search Aliexpress for "thermal insulation or insulator" materials.
There are some thin thermal insulators used for example in bike exhaust pipes or car motor pipes and some other different material types for other uses.
I think that anything (like the binding sheet you use) is better than the direct metal contact.
Anyway it was just an idea, I never tried it.


I think the acrylic box could be worst to dissipate heat.

 

[Setmagic]

@maxifuny Thank You for guidelines
AD826 can be replaced by AD8620 in terms of LPF usage?
In 22uF I have 50V ones it's still good?
Only 470 6.3 don't match in terms of spec.
I will replace 220 16V - same size or one side 470 16V and other 220 16V?

 

[maxifuny]

Anyway it was just an idea, I never tried it.

All this will not help. There will be overheating. The battery will die quickly.

 

[maxifuny]

AD826 can be replaced by AD8620 in terms of LPF usage?
In 22uF I have 50V ones it's still good?
Only 470 6.3 don't match in terms of spec.
I will replace 220 16V - same size or one side 470 16V and other 220 16V?

1) yes, but ADA4898-2 will be better
2) ок
3) yes
4) where exactly?