Hi,
I have a rather "uncommon" ground problem about common grounding of two audio circuits.
Basically what I want to to do is to create an external circuit which amplifies both the line-out and line-in of a PC sound card. The circuit is powered from it's own battery and is floating when unconnected. It's a data acquisition system for some sensors which work in the audio frequency band
Here's the catch:
My sound card has a bias voltage on the sleeve of the line-out jack. I took the sound card apart and figured out that they saved the output capacitor and instead put in a buffered voltage divider so that the jack's sleeve is at the center voltage (signal swing from 0 to 3.3V and the sleeve is at 1.65V with respect to earth ground).
The sleeve of the line-in jack however is directly connected to earth ground (measured with a multimeter).
Now I'm designing my interface board with it's own power supply, usually I'd connect the sleeve of both jack's to the negative terminal of the battery. However in my scenario this would happen:
- The LINE-IN jack's sleeve would pull the ground of my circuit to earth potential.
- The LINE-OUT jack's sleeve which is at 1.65V with respect to earth potential would be shorted to earth -> Problem!
Basically my question is what to do with the LINE-OUT jack's biased sleeve
I've attached a drawing of the situation.
PS: It's an external USB sound card with the VIA VT1620A chipset, but this shouldn't matter. They connected USB GND and USB shield together, that's why the ground on the sound card is at earth potential.
I have a rather "uncommon" ground problem about common grounding of two audio circuits.
Basically what I want to to do is to create an external circuit which amplifies both the line-out and line-in of a PC sound card. The circuit is powered from it's own battery and is floating when unconnected. It's a data acquisition system for some sensors which work in the audio frequency band
Here's the catch:
My sound card has a bias voltage on the sleeve of the line-out jack. I took the sound card apart and figured out that they saved the output capacitor and instead put in a buffered voltage divider so that the jack's sleeve is at the center voltage (signal swing from 0 to 3.3V and the sleeve is at 1.65V with respect to earth ground).
The sleeve of the line-in jack however is directly connected to earth ground (measured with a multimeter).
Now I'm designing my interface board with it's own power supply, usually I'd connect the sleeve of both jack's to the negative terminal of the battery. However in my scenario this would happen:
- The LINE-IN jack's sleeve would pull the ground of my circuit to earth potential.
- The LINE-OUT jack's sleeve which is at 1.65V with respect to earth potential would be shorted to earth -> Problem!
Basically my question is what to do with the LINE-OUT jack's biased sleeve
I've attached a drawing of the situation.
PS: It's an external USB sound card with the VIA VT1620A chipset, but this shouldn't matter. They connected USB GND and USB shield together, that's why the ground on the sound card is at earth potential.
Attachments
Have you measured this "bias" voltage when a load is present. Try a 10k and see if it drops.
Also... a DVM may show unpredictable voltages if there is any high frequency hf present. A scope is a better guide, or make a simple low pass filter (say 100k and 1uf) and measure the voltage across the cap for a true DC reading.
(I find it hard to believe that a true DC voltage exists on the output tbh but I suppose you never know )
Also... a DVM may show unpredictable voltages if there is any high frequency hf present. A scope is a better guide, or make a simple low pass filter (say 100k and 1uf) and measure the voltage across the cap for a true DC reading.
(I find it hard to believe that a true DC voltage exists on the output tbh
but I suppose you never know )Have you measured this "bias" voltage when a load is present. Try a 10k and see if it drops.
Also... a DVM may show unpredictable voltages if there is any high frequency hf present. A scope is a better guide, or make a simple low pass filter (say 100k and 1uf) and measure the voltage across the cap for a true DC reading.
(I find it hard to believe that a true DC voltage exists on the output tbh
I'll give it a try and hook up a load and check the voltage on the scope.
The DC voltage is only there because of the missing output capacitor, usually you have 0V with respect to the power ground on the sleeve and the signal swings positive and negative. In my case it's the same, the headphones which are usually connected can't reference the 1.65V to any other potential, so from the point of view of the headphones this 1.65V is the zero line and 3.3V is max. positive swing and 0V is max. negative swing.
I wonder what would happen if I directly connect the line-out to the line-in, theoretically it would blow up my sound card!
Thanks for your answer! The differential amplifier is a solution indeed, in my case the amplifier is a MC34119 (http://www.promelec.ru/pdf/mc34119.pdf), a simple headphone amplifier.
As far as I understand from the datasheet I could connect the signal to pin V_in and the virtual ground to pin FC1 and the GND of the amplifier goes to my "real" power supply ground on the board.
Another (rather dumb?) question: If I want to add a simple RC low pass filter before the input to the amplifier, can I connect the filter capacitor of the RC filter to the virtual ground of the input?
Thank you very much
Attachments
Thats exactly one of the main reasons I find it hard to believe there is DC present by design.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.