Guide me through assembling DAC board?

[Markw4]

Is there any advantage to one over the other?

The idea is to use a clean, low-noise voltage source as the reference. You existing power supply might do okay if you don't use it for anything else. The LTC6656 in the schematic is an ultra-low noise voltage reference, but maybe overkill for this case.

And should ignore the AVCC-R portion of the thread's circuit as well as the final resister/capacitor portion (red circle)?

If you only have one 3.3v input on the dac board for AVCC, then you would only use a single AVCC buffer opamp.

1) How do I identify which of the two 3.3V inputs on my DAC board power the opamps?

You have to follow the traces to see where they go. Sketch up a schematic of what you find along each power input path. Where they traces lead will give their purpose.

2) I could assemble this circuit on a breadboard, but it would look nicer and take up less space if I could put it on a PCB. Any idea on how I could go about doing this?

There are various ways to fabricate a circuit. Maybe worth doing a google search to find some of the ways. Adhesive-backed copper foil can be used to make a ground plane to attach all ground wires to. There are websites that show how. There are also some examples in the ES9038Q2M board thread where you found the schematics I linked.

When you think you have some understanding, then come maybe come back and talk some more before diving in.

3) Any idea what the "-DAC+" inputs in board's picture are for?

The are probably outputs. The I probably stands for current which is represented by the letter I in electronics (such as in ohm's law I = E/R). Again, follow the traces and see where they go to find out for sure. Sketch up a schematic of what you find.

*When I use the quote feature, I get an error message saying the post is too long and contains something like 7,000,000 characters.

Sounds like you are doing something wrong. Attached below is an example of how quotes work. You have to select to quote a whole post, then edit it to look like the example file. The first quote at the top refers back to the post that was quoted, so it contains more info between the brackets.

 

[mark3141]

Finally was able to make some progress on this after being busy, blowing fuses, and other mishaps. I ended up gutting an old CD player and putting this inside so now I should be a little safer with both a fuse and power button.



I've assembled the AVCC power supply, but I'm not sure how to properly ground this. Which ground do I use for these components? Should all components between the 3.3V power supply path and op amp as well as the op amp's 3.3V output supply reference the original 3.3V power supply for ground, and should the components in the between the +-12V power supply reference the 0V line from that power supply for ground?



I've run each of those as well as wiring the 3.3V reference and +-12V reference together, and all have some noise on the oscope. I'm not really sure which is better or what to make of the noise just yet, but some does seem to have regularity ('periodicity') to it. It's also quite possible I'm just seeing noise from the oscope and probably and need to read up more on that.


But I'm pretty sure there is only one 3.3V AVCC input here - the other 3.3V input on my board powers the USB receiver. I was also able to figure out the "-DAC+" part - they allow you to bypass the onboard IV stage and run your own so it's just the DAC input. Took me quite awhile to trace it all out, though. I really should have bough a good magnifying glass a long time ago. (What you thought was an "I" was actually a "1" - they're numbered to correspond to each channel.)

 

[Markw4]

I've assembled the AVCC power supply, but I'm not sure how to properly ground this. Which ground do I use for these components? Should all components between the 3.3V power supply path and op amp as well as the op amp's 3.3V output supply reference the original 3.3V power supply for ground, and should the components in the between the +-12V power supply reference the 0V line from that power supply for ground?

The DAC board is designed so that all the power supplies connect to board's ground plane. Beyond that, the board itself should be grounded to the chassis (if there is one) and to the AC power ground (if there is one).

It is standard practice for the AC power ground to be connected to the chassis at the point where the AC power enters the case. The DAC board ground plane can be grounded to the case (chassis) ground at a point of your choice. Sometimes chassis RCA connectors are where it is done. Other designs connect the ground plane to the chassis though one of the DAC board mounting screws.

The idea is to have a ground somewhere, but to locate it where hum and noise is not made worse by the choice of location.

In general grounding is a big subject, can't say very much here. You might try googling for the term "grounding and shielding" and see what you find.

 

[spikestabber]

The DAC board is designed so that all the power supplies connect to board's ground plane. Beyond that, the board itself should be grounded to the chassis (if there is one) and to the AC power ground (if there is one).

It is standard practice for the AC power ground to be connected to the chassis at the point where the AC power enters the case. The DAC board ground plane can be grounded to the case (chassis) ground at a point of your choice. Sometimes chassis RCA connectors are where it is done. Other designs connect the ground plane to the chassis though one of the DAC board mounting screws.

The idea is to have a ground somewhere, but to locate it where hum and noise is not made worse by the choice of location.

In general grounding is a big subject, can't say very much here. You might try googling for the term "grounding and shielding" and see what you find.

Grounding the dam1021 to the chassis is a good way to introduce a ground loop, many have had this issue in the main thread including myself. Standard practice is to not ground it directly to the chassis but to use insulating spacers, your power supply common should be grounded directly to the chassis via a screw mount and the common audio path upstream of that should not be grounded directly the chassis either when the ground itself is in the audio path as this will make a loop once again.

XLR pin1 can and should be grounded directly to the chassis but do not wire the ground path for the coaxes back from the dam1021 to the XLR ground, only the dam1021 itself should supply grounding for the coax shielding if you use that, some use twisted pair and even then I would never recommend tying an RCA ground directly to the chassis, let the dam1021 handle that.

In my dual mono build with relay headphone switch, this star ground setup results in absolutely no audible hum, tested well on both my nc400 and modulus-286 amplifiers in their standard grounded configs.

I would wager much of this is why the AC supply on the dam1021 should be avoided and instead use a much higher quality linear regulator. Then again I could never recommend the dam1021 for anyone starting out, go straight to the 1941 if you hate modding.
Even then, so much was improved that you should just do that anyway.

EDIT: Silly me realized this wasn't dam1021 specific, but it still has some good grounding info especially when using bipolar supplies.

 

[mark3141]

The DAC board is designed so that all the power supplies connect to board's ground plane. Beyond that, the board itself should be grounded to the chassis (if there is one) and to the AC power ground (if there is one).

It is standard practice for the AC power ground to be connected to the chassis at the point where the AC power enters the case. The DAC board ground plane can be grounded to the case (chassis) ground at a point of your choice. Sometimes chassis RCA connectors are where it is done. Other designs connect the ground plane to the chassis though one of the DAC board mounting screws.

The idea is to have a ground somewhere, but to locate it where hum and noise is not made worse by the choice of location.

In general grounding is a big subject, can't say very much here. You might try googling for the term "grounding and shielding" and see what you find.

Got it - that makes sense as to why I was seeing all this. I was measuring directly after the opamp output without connecting to the PCB. Looks like my board's ground terminates at the RCA connectors so this should all be moot, especially since I've never heard any ground noise.

Looking forward to the weekend and testing this out with the updates to the power supply.

And spikestabber, glad you updated your post to say you were talking about something. I read through it having no idea what you were talking about aside from general tips

 

[mark3141]

Not surprisingly, I've run into a new issue - I think with the DAC chip itself. I'm only getting output on channels 5-8; channels 1-4 are dead. I've tested this on both a Windows and Linux machine, verified I have continuity between each channel's input and output paths before and after the DAC chip, I've swapped the final output amps for channels 1/ 2 and channels 3/4 with those for channels 5/6 and channels 7/8 respectivelby, and I've swapped the USB to I2C board's channels 1, 2, 3, 4 output with channels 5, 6, 7, 8 respectively. In all cases, I see the same issue where there's no output on 1-4, only 5-8. Any thoughts on what I could have done and if I can fix it?



But Mark, I do have to thank you for the suggestion on the AVCC power supply. I was only expecting a subtle improvement, but the difference is readily apparently.

 

[Markw4]

You should check that you have AVCC supply voltage for both channels, left and right.

Check all power supply voltages.

Turn of all the power, let all the filter caps discharge in the power supply for a few minutes, then power it back on and make sure the problem persists.

Let us know what you find.

 

[mark3141]

Checked that my power supplies are outputting the right voltage and verified continuity with all AVCC supply voltages on the DAC chip itself and the +-12V supply voltages to the offending channels' final opamp. I powered everything down, waited awhile, and the issue persists.

For what it's worth, there's a single 3.3V AVCC supply on my board that feeds both AVCC L and R. On the DAC itself, AVCC L powers the odd numbered channels, and AVCC R powers the even numbered channels. Since 1-4 have the issue and 5-8 do not, and the results of troubleshooting so far, I do not think this is a power supply issue, at least not between AVCC L and AVCC R. However, it is weird that consecutively numbered and physically adjacent channels have issues and consecutively numbered and physically adjacent channels do not. That's why I'm still hoping the issue is something dumb I'm overlooking on my part.

If useful, there is another 3.3V supply on this board, but this powers the USB input.

 

[Markw4]

Not clear if there is an MCU on the board, or if USB is part of the board? If there is a volume control? Feeding the USB from Windows? How is the board controlled in use, does it have any controls, ability to switch inputs, etc.?

Some pictures might help. The big picture, and the close up details.

There could problems in any of those above mentioned places. I would start checking digital signals with a scope and take a look at the dac chip I2C registers. If that is more than you know how to do, its still possible there is something further under your control that could be checked.

I would agree the problem seems very suspicious unless maybe something happened along the way you haven't told us about that might account for it. To have a dac working fine one day and then half the channels not working the next day is very odd.