PCB layout for audio - Page 2

banana · 22nd August 2007, 04:24 PM

Quote:

Originally posted by aparatusonitus

Could you elaborate a bit and explain how one can address this problem?

Suggestions from Opamp app note work well. Rub off ground plane right below output pin, as well as inverting input pin.

aparatusonitus · 22nd August 2007, 04:42 PM

Thank you for the tip banana, I like it 'cos it's so easy to implement

Conrad Hoffman · 22nd August 2007, 10:05 PM

Usually, if a circuit is extremely complex, just using a ground plane won't help. It may make the thing even harder to lay out, or you end up with a spidery little bunch of ground plane around all the traces- essentially a randomly wired ground system, rather than a planned one. What is typically done is to go to a 4-layer board. That way you can have your power and ground planes, plus two layers for horizontal and vertical traces. IMO, not really optimal for audio if done in the traditional digital fashion. Good analog design has its own set of requirements unlike just about anything else. The usual way to take care of parasitic capacitance, say around a high value feedback resistor, is to simply void the ground plane underneath the part. Leave a rectangular hole just a bit larger than the part.

AndrewT · 22nd August 2007, 10:14 PM

Hi,
reading between the lines of this and previous threads on ground planes, I have come to the conclusion that it takes more skill to successfully implement a ground plane than to work without one.

Or put it another way.
If you have to ask, then you don't have the skill to find and eliminate the problems likely to arise as a result of using a badly designed ground plane.

suzyj · 22nd August 2007, 10:18 PM

A ground plane is really only useful if you're doing controlled impedance work, and that's not gonna happen at audio frequencies.

That said, copper pours are very useful, as they're the lowest impedance way to rout a given net. Same goes with double sided boards - they give you much more opportunity to do a decent low impedance connection, that doesn't snake all over the board. They also allow you to put components on both sides of the board, making the board more compact, again with commensurate gains in connection impedance.

jaycee · 22nd August 2007, 10:28 PM

All this makes me wonder if I screwed up my latest pcb by adding a ground plane ! It is single sided, but the ground plane was the easiest way to take care of routing the audio ground:

http://www.darkmatter.myby.co.uk/preamp-brd.pdf

and the schematic for those interested

http://www.darkmatter.myby.co.uk/preamp-sch.pdf

The pot is a 100K linear track motorised ALPS potentiometer I salvaged from a mid-80's Pioneer stereo system that the neighbours had thrown away. I plan to use NE5532 for the opamps.

Did I use the ground plane in a Bad Way? I'm still learning how to lay out PCB's well!

panson_hk · 23rd August 2007, 01:10 AM

Hi jaycee

I have done similar in my recent layouts. Try to place all traces on top and a solid ground bottom.

Apart from stray cap, can a ground plane provide protection against interference?

Shall we add a ground (return path) under a high current trace to cancel the generated magnetic field?

panson

sam9 · 23rd August 2007, 02:38 AM

It is probably well to distinguish "power ground" (the one that cap caps between it and the rails) from "signal return ground" (the one that also can act as a shield to the signal input). In the case of the first, I try to keep it as short as possible taking up little space on the PCB, the leading it away 90 deg from the PCB as quickly as possible and keeping it away from the input signal, etc. I do similarly with the power rails as much as possible.

The quiet ground may have to wander around the PCB. If you limit it so it doesn't get into the output section where there are higher currents, making it copper fill might be OK. (I didn't find it made a noticiible difference but maybe I'm just an insensitive lout.)

banana · 23rd August 2007, 03:34 AM

Quote:

Originally posted by panson_hk
Apart from stray cap, can a ground plane provide protection against interference?

Shall we add a ground (return path) under a high current trace to cancel the generated magnetic field?

In practice, it's only a partial shield, and the stray field in audio frequency is only very short range.

It does help reduce crosstalk between adjacent trace after all.

h_a · 23rd August 2007, 03:06 PM

What about hatched groundplanes?

Some designers prefer to use these, are these a handy tradeoff in the sense that they do considerably more good than harm? At least stray capacitance is strongly reduced.

What do you think?

All the best, Hannes

22nd August 2007, 04:42 PM	#12
aparatusonitus diyAudio Member Join Date: Jan 2004 Location: Split, Croatia	Thank you for the tip banana, I like it 'cos it's so easy to implement __________________ Non é mai abbastanza...

23rd August 2007, 03:06 PM	#20
h_a diyAudio Member Join Date: Sep 2006 Location: Graz, Austria	What about hatched groundplanes? Some designers prefer to use these, are these a handy tradeoff in the sense that they do considerably more good than harm? At least stray capacitance is strongly reduced. What do you think? All the best, Hannes __________________ fresh matched IRFP240/IRFP9240 fets \|\| AlephJ/JX-kits F5 transistor kits \|\| Burning Amp BA-1/2 transistor kits

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
JL Audio Mainboard Layout?	XxGoKoUxX	Car Audio	9	20th February 2008 03:33 AM
audio buffer layout	digi01	Chip Amps	71	7th October 2006 05:34 AM
could John Audio re-desigh the Trace elliot TDA 1514A layout	RIC	Chip Amps	1	27th December 2003 04:39 PM
PCB layout of audio controller with PGA2310	louischia	Pass Labs	20	24th April 2003 11:11 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

22nd August 2007, 10:05 PM	#13
Conrad Hoffman diyAudio Member Join Date: Mar 2007 Location: Canandaigua, NY USA	Usually, if a circuit is extremely complex, just using a ground plane won't help. It may make the thing even harder to lay out, or you end up with a spidery little bunch of ground plane around all the traces- essentially a randomly wired ground system, rather than a planned one. What is typically done is to go to a 4-layer board. That way you can have your power and ground planes, plus two layers for horizontal and vertical traces. IMO, not really optimal for audio if done in the traditional digital fashion. Good analog design has its own set of requirements unlike just about anything else. The usual way to take care of parasitic capacitance, say around a high value feedback resistor, is to simply void the ground plane underneath the part. Leave a rectangular hole just a bit larger than the part.

22nd August 2007, 10:14 PM	#14
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	Hi, reading between the lines of this and previous threads on ground planes, I have come to the conclusion that it takes more skill to successfully implement a ground plane than to work without one. Or put it another way. If you have to ask, then you don't have the skill to find and eliminate the problems likely to arise as a result of using a badly designed ground plane.

22nd August 2007, 10:18 PM	#15
suzyj diyAudio Member Join Date: Jan 2006 Location: Sydney	A ground plane is really only useful if you're doing controlled impedance work, and that's not gonna happen at audio frequencies. That said, copper pours are very useful, as they're the lowest impedance way to rout a given net. Same goes with double sided boards - they give you much more opportunity to do a decent low impedance connection, that doesn't snake all over the board. They also allow you to put components on both sides of the board, making the board more compact, again with commensurate gains in connection impedance.

22nd August 2007, 10:28 PM	#16
jaycee diyAudio Member Join Date: Apr 2004 Location: Norwich, UK	All this makes me wonder if I screwed up my latest pcb by adding a ground plane ! It is single sided, but the ground plane was the easiest way to take care of routing the audio ground: http://www.darkmatter.myby.co.uk/preamp-brd.pdf and the schematic for those interested http://www.darkmatter.myby.co.uk/preamp-sch.pdf The pot is a 100K linear track motorised ALPS potentiometer I salvaged from a mid-80's Pioneer stereo system that the neighbours had thrown away. I plan to use NE5532 for the opamps. Did I use the ground plane in a Bad Way? I'm still learning how to lay out PCB's well!

23rd August 2007, 01:10 AM	#17
panson_hk diyAudio Member Join Date: Aug 2005 Blog Entries: 10	Hi jaycee I have done similar in my recent layouts. Try to place all traces on top and a solid ground bottom. Apart from stray cap, can a ground plane provide protection against interference? Shall we add a ground (return path) under a high current trace to cancel the generated magnetic field? panson

23rd August 2007, 02:38 AM	#18
sam9 diyAudio Member Join Date: Jun 2002 Location: Left Coast	It is probably well to distinguish "power ground" (the one that cap caps between it and the rails) from "signal return ground" (the one that also can act as a shield to the signal input). In the case of the first, I try to keep it as short as possible taking up little space on the PCB, the leading it away 90 deg from the PCB as quickly as possible and keeping it away from the input signal, etc. I do similarly with the power rails as much as possible. The quiet ground may have to wander around the PCB. If you limit it so it doesn't get into the output section where there are higher currents, making it copper fill might be OK. (I didn't find it made a noticiible difference but maybe I'm just an insensitive lout.)