I built another one of my LM3886 boards up yesterday but left pin 7 unsoldered, bent up so I could do some measurements. The amp didn't even function if it was left disconnected, and when I measured the current I got a constant 2.4mA regardless of whether the amp was idle, playing music, or muted. A simple experiment but I think you're correct @Sangram. For now I've connected it directly to the point on my board where the signal ground and speaker ground meet, at the speaker ground side.
Bruno Putzey's article on the subject is informative and entertaining reading:
https://www.diyaudio.com/archive/bl...d1460406090-bruno-putzeys-micropre-g-word.pdf
https://www.diyaudio.com/archive/bl...d1460406090-bruno-putzeys-micropre-g-word.pdf
I didnt get all the notifications for replies, so thanks all for the replies.
I went ahead and did a 0.1v board, where the signal ground is on its own ground plane on the right side of the board as shown in the image.
The board is produced and I'm awaiting shipping, but would certainly welcome any comments (and criticism 😀 ), as this is my first ever board I'm making (other than the PSU board I sent off at the same time).
I went ahead and did a 0.1v board, where the signal ground is on its own ground plane on the right side of the board as shown in the image.
The board is produced and I'm awaiting shipping, but would certainly welcome any comments (and criticism 😀 ), as this is my first ever board I'm making (other than the PSU board I sent off at the same time).
Last edited:
Also unsure if the images are loading again.
Thank you @jbau for the link, that would be good bedtime reading for amp grounding.
Thank you @jbau for the link, that would be good bedtime reading for amp grounding.
Yes, I know how this man feels. I've seen "signal ground" and wondered ... hey, how do I deal with this.
There is not "only one ground" as someone said ... and isn't it fair to say there isn't when one is labeled differently. When you read schematics there is chassis, there is actual "earth" and there is ground and ground may not be the same as earth. As someone points out ... when you link them all together they will be pulled to whatever is the strongest sink ... unless there is some strong voltage leak pulling it up somewhere. A lot of signal can be induced as interference/RF, wire induction noise from parallel, the magnetic fields of transformers ... so there is the whole shielding deal. In a good design this will flow to ground and not loop around the circuit and flow back into the signal path.
So you know you can just put a 10ohm resistor as an experiment and see if you like that making the path to ground locally at the PC or other source shorter than the path to ground over in your other audio device circuit - your amp. Why not. People do it in some designs. I've got an LM3886 and I didn't do it - and it has no hot chassis its all just plastic. The transformer isn't earthed (it should be but it isn't) and the thing is silent, clean and great with the DAC which probably would be an earth connection but it's running from a USB cable so that is uncertain.
There is "virtual ground" too ... and that can definitely not be zero volts.
So for it to explode into all these what ifs and different grounds is a sign someone is thinking, and yes possibly too much, but they might want to read an audio expert explain the way to deal with it. I'm not that guy but I follow one guy that seems to be and here - you can read a huge long explanation
Earthing (Grounding) Your Hi-Fi - Tricks and Techniques
There is not "only one ground" as someone said ... and isn't it fair to say there isn't when one is labeled differently. When you read schematics there is chassis, there is actual "earth" and there is ground and ground may not be the same as earth. As someone points out ... when you link them all together they will be pulled to whatever is the strongest sink ... unless there is some strong voltage leak pulling it up somewhere. A lot of signal can be induced as interference/RF, wire induction noise from parallel, the magnetic fields of transformers ... so there is the whole shielding deal. In a good design this will flow to ground and not loop around the circuit and flow back into the signal path.
So you know you can just put a 10ohm resistor as an experiment and see if you like that making the path to ground locally at the PC or other source shorter than the path to ground over in your other audio device circuit - your amp. Why not. People do it in some designs. I've got an LM3886 and I didn't do it - and it has no hot chassis its all just plastic. The transformer isn't earthed (it should be but it isn't) and the thing is silent, clean and great with the DAC which probably would be an earth connection but it's running from a USB cable so that is uncertain.
There is "virtual ground" too ... and that can definitely not be zero volts.
So for it to explode into all these what ifs and different grounds is a sign someone is thinking, and yes possibly too much, but they might want to read an audio expert explain the way to deal with it. I'm not that guy but I follow one guy that seems to be and here - you can read a huge long explanation
Earthing (Grounding) Your Hi-Fi - Tricks and Techniques