I got a preamp that i designed. Its a single chassis dual mono, mixture of ground plane and star ground, completely seperate left and right channels except for the hum breaker nodes joined at chasiss.
I found a weird symptom recently. If i take the signal part of the rca jack (the tip) and touch the ground part of the preamp, that is one tip to the L ground and the other to the R ground, ill get a faint music irregard of the position of the volume pot. This doesnt happen if i touch just one tip to ground but if its to both channels theres sound. Which makes me think its a loop but in normal conditions the preamp is dead quiet. Infact its hum free enough to be a headphone amp for iem's.
Whats the mechanism behind why this might be happening?
I found a weird symptom recently. If i take the signal part of the rca jack (the tip) and touch the ground part of the preamp, that is one tip to the L ground and the other to the R ground, ill get a faint music irregard of the position of the volume pot. This doesnt happen if i touch just one tip to ground but if its to both channels theres sound. Which makes me think its a loop but in normal conditions the preamp is dead quiet. Infact its hum free enough to be a headphone amp for iem's.
Whats the mechanism behind why this might be happening?
Hard to say without a detailed examination but the mechanism has to be that the two 'live' signal tips generate a voltage between parts of the preamp that can respond to that signal.
As you need the two tips contacting the two input grounds you could check to see if the 'problem' disappears with a mono signal i.e. if both tips have the same voltage at the same time then they should not be able to generate any difference signal between them.
It could also I suppose be something related to mains grounds and any ground lift resistors. For example does it do this with a battery powered source like an MP3 player.
Hi Mooly,
Your suspicion is correct. If i turn the signal into mono there is no symptom. Also if i pull out one of the two output jack the symptom disppears.
And also the sound is always louder on one channel.
The input is fed from my phone which is floated.
I took out the hum breaker circuit (back to bsck diode//cap//resistor) and the symptom was the same.
What is going on here?? 🙁
Your suspicion is correct. If i turn the signal into mono there is no symptom. Also if i pull out one of the two output jack the symptom disppears.
And also the sound is always louder on one channel.
The input is fed from my phone which is floated.
I took out the hum breaker circuit (back to bsck diode//cap//resistor) and the symptom was the same.
What is going on here?? 🙁
Here is how the signal ground looks like for my board. Im marking them in the other pic so its easier to identify. The output gtound is on the left side and the input on the right. The signal trace is on the bottom layer. Could it just be capacitive coupling since the grounds and traces are so tightly together?
Attachments
What im most concerned about is if this symptom reveals some kind of a design flaw and thinking back i dont think it does and if anything what im doing is just a method to introduce artifacts.
I forgot to mention that the output ground meets at a common point on a headphone jack and im listening to this symptom on a headphone
The input is also floated in the form of a phone. So what im doing when i touch the input jacks to the ground nodes im creating a huge floated loop with the sensitive nodes being modulated by grounds that have signal voltages running through.
I think i shouldve been more surprised if something weird didnt happen with what did.
Unplugging one output jack kills the loop so the artifact goes away too.
Please let me know if my thinking is right
I forgot to mention that the output ground meets at a common point on a headphone jack and im listening to this symptom on a headphone
The input is also floated in the form of a phone. So what im doing when i touch the input jacks to the ground nodes im creating a huge floated loop with the sensitive nodes being modulated by grounds that have signal voltages running through.
I think i shouldve been more surprised if something weird didnt happen with what did.
Unplugging one output jack kills the loop so the artifact goes away too.
Please let me know if my thinking is right