Okay that link worked. Maybe you can ask a moderator to delete those last two or three posts without a working link.
Already have a board of that version here. They made things a little more compact, including the display and button which are now on the board rather than on a smaller board at the end of a ribbon cable. It is about the same as the old board in most other respects.
Bottom line, that one should be okay.
Instructions should be just like the ones for the old version which is at the bottom of the ad at: AK4137 Raspberry Pi IIS декодер DAC доска SRC аудио Поддержка I2S 32bit 384 K DSD256 DSD I2S IIS преобразования с ЖК дисплей Дисплей-in Усилитель from Бытовая электроника on Aliexpress.com | Alibaba Group
It expects a USB board with a connector having two rows of pins in order for it to read all the configuration information for input sample rate, DSD_ON, and Mute. We could probably figure out how to patch in a USB board that has those signals, but in a different pin layout format. However, some USB boards don't have all the necessary signals in which case you would have to configure the input for manually for each type of file being played.
Last edited:
I've found a local supplier of parts. Looks like they have some opamps (there is a NE5532 on the previous page too), caps are Jamicon, resistors ill take 1% tolerance
Got it! Judging by the traces from the 3.3v rail, those are these 2
So, i just simply dont connect Vref to anything? Ill just leave those connections open in case of an I/V output stage in the future.
So the diagram is correct i presume, i didnt expect that.
Now to just wait for some things to arrive...
Absolute maximum voltage for non-5v-tolerant pins is 4.7v relative to ground (except DVDD which is 1.8v maximum).
Recommended voltage for those pins is DVCC voltage +-0.3v
A possible problem with LM317 could be the adjustment pot, something which has a known risk of going noisy after some time.
Last edited:
Another thing about LM317 which is explained in the data sheet is that if using a large output cap with it, a reverse protection diode should be connected across the LM317 to protect it in case the input cap discharges first when power is turned off. Also, sometimes a diode is used to protect the reference pin from a similar type of problem. If the LM317 fails, then it might not protect the dac chip next time power is applied. Recommend to read the data sheet carefully.
Also, I tend to prefer 2-stage regulators for critical audio circuits. Among other things, it can more fully attenuate input ripple that way. It is possible to make ripple very small using a high-gain-error-amp LDO instead, but as stated before that doesn't usually sound the best for analog audio circuitry, IME.
Even when using LDOs for digital loads, I still prefer to use 2-stage regulation typically with the first stage outputting around 8v.
That's also plenty of input voltage to drive an LT304x acting as a low noise 3.3v reference for AVCC.
Also, I tend to prefer 2-stage regulators for critical audio circuits. Among other things, it can more fully attenuate input ripple that way. It is possible to make ripple very small using a high-gain-error-amp LDO instead, but as stated before that doesn't usually sound the best for analog audio circuitry, IME.
Even when using LDOs for digital loads, I still prefer to use 2-stage regulation typically with the first stage outputting around 8v.
That's also plenty of input voltage to drive an LT304x acting as a low noise 3.3v reference for AVCC.
Since you mentioned pots, the dac board's volume controls dont pass the signal trough the pot, right? I think on es9038 analog and digital signals are separate. I know my O2 amp does run the signal trough the pot (but thats a whole different animal).
Correct. The volume pot has DC on it and is read by an ADC inside the MCU. The MCU then writes digital data to the volume control registers in the dac chip.
Also, a single gang pot used for analog-audio volume control would only work for one of the two stereo channels anyway. Just knowing that is enough to exclude audio going through the pot.
Last edited:
OK, just received Khadas tone board. Hooked it up to my office system via USB. It is clear and pretty transparent but I am getting a similar LME49720 character out of it. Hard to describe. However if you've auditioned the difference between the LME49720/10 and the OPA1612/1 I think you will understand. Not a lot of warmth as some would say a tad bit lean but dynamic. For $99 with USB, not a bad deal ..great actually and certainly not a bad listen. Plug and play baby.
Further listening is called for. It is a nice addition to a pair of LSIM 703s I use in the office. In the next couple weeks, I'll haul it to the main rig.
Further listening is called for. It is a nice addition to a pair of LSIM 703s I use in the office. In the next couple weeks, I'll haul it to the main rig.
Last edited:
Hmmm... what that sort of suggests to me is that your other ESS dac still hasn't been modded as much as it could be.
Hello Mark, im sorry but i got curious and i have another question...i hope you forgive me.
So hoping the AVCC diagram in my previous post is correct, building on the I/V stage, i've taken a look at MrSlim's schematic for balanced output and worked from there (i have a single ended amp inputs so i would use regular rca output), i took your original schematic from the .rar
I have made this diagram for wiring, please tell me if this is how it was intended to be, I hope im not making a big blunder.
Pic
I have a couple of questions though, if i could ask
1. Maybe i misunderstood something, but when i see people build these, there is 3 physical opamps DIP packages. But the lme49720 is a dual opamp, so the opamps marked OP1 and OP2 on my picture are the same physical package (so i would imagine only 2 opamps)?
Or are we using a single output from 3 lme49720?
2.Also does it matter which Vref goes to which I/V output?
3. The points where i take the DAC output pins, i keep those SMD inductors in place, right?
4. Instead of using new RCA jacks, can i somehow hijack the rca outputs on the board?
Thank you
So hoping the AVCC diagram in my previous post is correct, building on the I/V stage, i've taken a look at MrSlim's schematic for balanced output and worked from there (i have a single ended amp inputs so i would use regular rca output), i took your original schematic from the .rar
I have made this diagram for wiring, please tell me if this is how it was intended to be, I hope im not making a big blunder.
Pic
I have a couple of questions though, if i could ask
1. Maybe i misunderstood something, but when i see people build these, there is 3 physical opamps DIP packages. But the lme49720 is a dual opamp, so the opamps marked OP1 and OP2 on my picture are the same physical package (so i would imagine only 2 opamps)?
Or are we using a single output from 3 lme49720?
2.Also does it matter which Vref goes to which I/V output?
3. The points where i take the DAC output pins, i keep those SMD inductors in place, right?
4. Instead of using new RCA jacks, can i somehow hijack the rca outputs on the board?
Thank you
Three opamps are needed for each channel's recommended output stage. Two I/V converters and one differential summing filter per channel. That takes a total of three dual opamps, or three single opamps per channel. For an opamp AVCC supply, one dual opamp or two single opamps are needed, that's one opamp for AVCC_L and one for AVCC_R.
One Vref goes to the left channel I/V opamps, and likewise for the right channel. AVCC and Vref should be for the same channel.
There are SMD resistors attached to each of the + and- dac outputs for each of the two stereo channels. Those SMD resistors should be removed when using an alternate output stage, as we recommend doing. There are some pics posted in the thread showing that the SMD resistors have been removed and wire leads going to the new output stage have been soldered onto the solder pads where the resistors used to be. Maybe we could help you find the the pics, if you would like to see them.
The output RCA connectors can be used for a new output stage. Simply remove the capacitors connecting them to the existing output stage. Follow the traces to find the caps
Last edited:
Oh yeah, of course! So in the I/V stage one opamp handles one channel, and then both go to the third opamp for summing. Why didnt i remember this..
Sorry for asking redundant questions, i found the post where you describe this. Somehow they always slip trough. Seems like i also have to cut the +15V trace, that sounds like an important bit
I see you have removed all the components, but i'd like to just bypass them if possible (in case i need to revert). I cant really see clearly which traces go to the capacitors on the 1.07 green board, so i'm assuming the layout is the same as the blue board in your post. I marked out the components (pic) to remove in red, but maybe the answer was the SMD components just ahead of what i crossed out.
Additionally the ones crossed out in yellow i imagined would be good to remove to isolate the output from the "rear" since they seem to connect to the caps? Then just put the signal/gnd in the holes where the caps were and it leads to the jacks.
So other than that original wiring picture from the last post seems about right?
Thanks!
Last edited: