PCB layout for phonostage

[MJL21193]

Originally posted by robmil
MJL - Is connecting a 10R resistor between PSU Gnd and Audio Gnd common practice as I've never seen it done before. Do you use a 5 watt resistor for this? If the resistor value rises (heat etc) what happens to your RIAA accuracy?

Rob.

Hi Rob,
I don't know what is common, but I'll try nearly anything to eliminate hum. I'm just using a 1/4 watt resistor as the current would be very tiny, so no worries about it heating up. Shoot, my regulators don't even get warm.
I have gone through several changes and board revisions (getting quite a stack of used boards) and I'm nearly there. I have a ground plane on the bottom layer that is connected to the power supply common. This handles all of the decoupling. I then have the signal ground separated from this by the small value resistor. This picks up the shunts from the filters as well as the audio ground.
I just etched and drilled board #5, ready to stuff and test. Hopefully it's a winner this time.

[robmil]

That sounds interesting. I'll have to try it.

Thank you

Rob

[h_a]

Ah, nothing like a morning with my harmless little post torn apart in true DIYAudio-style

Whatever you gentlemen prefer, however at least reading the full post before replying would be nice, eh?

ACD Where do you find these voltage drops and large currents in a phono stage? Maybe you should concider a redesign to match voltage and current to the actual job...........

as in my very same post:

Anyway the current demands in a phono stage are very small in general.

My post was intended to give simple advice to a novice and not to show-off expertise (which I anyway don't have).

Nevertheless, I can't resist to post a snipet of "A Practical Guide to High-Speed Printed-Circuit-Board Layout" By John Ardizzoni [analog.com]

A ground plane acts as a common reference voltage, provides shielding, enables heat dissipation, and reduces stray inductance (but it also increases parasitic capacitance). While there are many advantages to using a ground plane, care must be taken when implementing it, because there are limitations to what it can and cannot do.

...

Nevertheless, there are exceptions, and sometimes less ground plane is better. High-speed op amps will perform better if the ground plane is removed from under the input and output pads. The stray capacitance introduced by the ground plane at the input, added to the op amp’s input capacitance, lowers the phase margin and can cause instability. As seen in the parasitics discussion, 1 pF of capacitance at an op amp’s input can cause significant peaking. Capacitive loading at the output—including strays—creates a pole in the feedback loop. This can reduce phase margin and could cause the circuit to become unstable.

Pick your poison

Of course I agree that ground planes (and power planes) are necessary for RF-and digital applications (fast rising edges), however in audio I prefer a clean star-ground and the possibility to use the 2nd side for routing other signals.

Now go and tell me there's a simple answer to grounding.

Have fun, Hannes

EDIT: found another Maxim-appnote with general advice (well, it's about cell phones...doesn't P-A work for a cell-phone company? )

1. Establish a Continuous Ground Plane for Digital Circuits
2. Keep Ground Currents Separate
3. Use the Star Grounding Technique for Analog Circuits
4. Maximize the Effectiveness of Bypass Capacitors
5. Flood All Unused PCB Area with Ground

http://www.maxim-ic.com/appnotes.cfm/an_pk/4079

[MJL21193]

Originally posted by robmil
That sounds interesting. I'll have to try it.

I highly recommend it. I just tested the latest board and I think I have a winner. I used a ground plain on both top and bottom, as described above. Sitting out in the open (no chassis yet) I tested it with RMAA for noise level and it is below -75db. I confirmed this with Adobe Audition, recording the output without a record playing.
My turntable is not on a very solid stand and it has a computer close by with a rather noisy fan going so this figure might be lower. Compare with the previous board version, I have made a serious drop in noise. The noise level of the previous version tested at -43db.