INA134 unbalancing....quick question...

[redrabbit]

Hello.....

In this circuit, where do the 2 grounds connect to ?

I have the XLR's "pin 1" connected directly to chassis.
The rest of my circuit grounds go to PSU star. That star ground is then connected to chassis through a "loop breaker" (resistor + capacitor).

I'm primarily concerned with the R1-R2 ground.

Parts:
OPA2134 (A1, A2)
INA134 (A3)
R1, R2 = 1.5 Meg.

[pinkmouse]

The input ground of the circuit should run only to pin 1 of the XLR for best isolation.

[AndrewT]

Read Rane and Jensen papers.

[redrabbit]

Originally posted by AndrewT
Read Rane and Jensen papers.

Andrew, are you tired of seeing these balancing questions here ?

I read the papers, in fact the illustration is from a Jensen white paper.
But of the articles I did read (including Jung/Analog's), I was unable to find any mention regarding where to place those grounds in the schematic, specifically.
It is possible I missed it in the texts, or perhaps it's common sense implementation.....but is not yet common to me....therefore, I asked the question.

Very often I ask a question where I suspect I know the answer, but need confirmation....and I ask it in a way that may be helpful to others who use the "search" function.

Sorry for the mini rant.
I'll connect the R1, R2 ground to "pin 1/chassis" ground, and the other ground to PSU/star ground. I assume R1, R2 are usually placed right at the input XLR pins. I placed them on the board.
R1, R2 are matched.
http://i5.photobucket.com/albums/y17...t/DSCN2946.jpg



=RR=

[AndrewT]

Sorry,
it wasn't intended to sound like that.

Rane & Jensen, particularly, have addressed this signal grounding issue and why it must be kept separate from the chassis.

[redrabbit]

No problem.



=RR=

[Speedskater]

The R1/R2 ground should go to the audio ground.
The XLR pin 1 ground is not part of the audio circuit.

[jcx]

R1/2 provide DC bias to the op amp inputs

the balanced signal inputs could be a floating xfmr secondary, or Cap coupled with no DC path

without a DC current path the op amp input bias current would pull the floating balanced inputs beyond the op amp input common mode operating range

The DC path has to return to amplifier power supply/signal ground

[janneman]

Originally posted by jcx
R1/2 provide DC bias to the op amp inputs

the balanced signal inputs could be a floating xfmr secondary, or Cap coupled with no DC path

without a DC current path the op amp input bias current would pull the floating balanced inputs beyond the op amp input common mode operating range

The DC path has to return to amplifier power supply/signal ground

Yes, correct. The input signal consists of the difference between the two signal inputs, it has no reference to any ground. Of course, it should be *near* ground to stay within the opamps' linear region. The R1,R2 should go to supply ground. Because the output is SE, the output RCA ground pin also connect to supply ground.

Jan Didden

[redrabbit]

So I have it wrong.....the R1, R2 ground is signal ground, not "pin 1" ground/shield.
Pin 1 goes to chassis, with NOTHING else attached to pin 1.


If I trusted all schematics ( ) , there would have been a "chassis/earth" symbol there , instead of the above signal ground symbol (triangle), if was actually meant to be chassis ground.

Sorry ....I'm but I'm still learning. Thanks for helping me through this. I DO want to learn "why", and not just "how".

I wish it was easy enough to say "Pin 1 to chassis, that's it, forget about it, don't touch it."
...but I see data sheets that components are sometimes (optionally) connected to Pin 1, such as in the THAT 1646 driver datasheet:

Fig. 8
http://www.thatcorp.com/datashts/1600data.pdf


=RR=