Line Input and Channel Gain module for summing Amp

abbey road d enfer said:

No. The "common return" is to the reference point of the PSU, which is the point the regulators (or the last smoothing cap if there's no regs) are referenced to. The xfmr's center-tap is a point that is best left floating at the end of an otherwise unconnected antenna (more in the original sense of "appendage" than in the radio-electrical sense).

Click to expand...

That makes sense now. At the time I designed the PSU, I was looking at "ground" of the PSU (including the center tap) as one contiguous 0V node. I see now why that's totally incorrect. I'm also rapidly beginning to despise the word "ground" as a terribly misused term in circuit design. The concept of a ubiquitous 0V, 0 impedance reference which is convenient in theoretical design, simply doesn't exist in the real world. I wish I had understood this sooner.

abbey road d enfer said:

You need to connect all theses "grounds" to each other with resistors (it offers many possibilities) and look at the consequences.
Remember that some of these points are inducing current, some are just receiving and some are both simulatneously.
ND1 is a receiver (forget about bias current), ND2 is a "transmitter" (assuming PSRR is large enough) and ND3 is both (it transmits the current flowing through the fader AND receives part of the parasitic voltage that is imprinted to it. ND7 is just a "transmitter", and ND5 is typically both; that shows the importance of not making the NFB network values too low because it increases the parasitic current.
When you establish connections between the various "grounds, you will see that the hierarchical arrangement is the one that creates the less cross-talk between the various paths. And indeed ND3 should be as close as possible to ND2, same for ND6 & 7.

Click to expand...

When you say connect these points with resistors, do you literally mean resistors of significant value, or do you mean jumpers / traces? (Which I suppose are technically resistors themselves)

bjoneson said:

When you say connect these points with resistors, do you literally mean resistors of significant value, or do you mean jumpers / traces? (Which I suppose are technically resistors themselves)

Click to expand...

I meant virtually for the sake of analysis.

abbey road d enfer said:

bjoneson said:

When you say connect these points with resistors, do you literally mean resistors of significant value, or do you mean jumpers / traces? (Which I suppose are technically resistors themselves)

Click to expand...

I meant virtually for the sake of analysis.

Click to expand...

I had a feeling. Some day before too long, hopefully I won't ask ridiculous questions. 

I really appreciate the time and patience you and several others have given me.

-Bob

This was another great article that has helped me "unlearn" much of the myth of a "perfect ground"

http://www.analog.com/static/imported-files/rarely_asked_questions/moreInfo_raq_groundingClean.html

Guys, thanks again for all of the feedback. I've learned so much over the last several weeks. I've been reading, and reading (and reading some more). Really trying to wrap my head around how to control return currents, trade offs and different approaches.

The first thing I focused on with this PCB was to try to get as many of the signal and supply traces as possible on to one layer. It took a number of iterations, but I was finally able to get it all laid out with just a very small handful of short jumpers / vias. This now allows for maximum flexibility for routing return currents on the bottom layer of the board.

As I read and tried to wrap my head around this, I think in my case a solid ground plane makes sense. With the relatively small currents I'll be working with, and some forethought to component placement, I think the risk of crosstalk via common impedance is relatively small in this application.

I tried to investigate as many different practical small signal implementations I could find, and it seems a ground plane is relatively common. Once you start driving significantly low impedance such as in power / headphone amps, this risk becomes more substantial with the additional current, and it seems more common to see "return traces" / star schemes.

I still plan to connect each module to a system level star ground point. But it was difficult for me to justify the complexity of a "star" type scheme on the PCB, when that module would still be sharing a common impedance back to the system level star ground.

I suppose the alternative would be to plane each of the 4 channels independently on the PCB, and provide 1-2 pins on the card edge for each of those planes that would connect back to a system star ground. Given I'll be running a total of 24 channels, a start ground with 24 leads starts to sound really impractical to me.

I'm new to this, and if you guys see anything that looks totally out of whack, or have any comments on the grounding / current return scheme, I'd really appreciate it.

Thanks again!

-Bob