Unity Gain buffer and AC coupling

Thread Starter

Gibson486

Joined Jul 20, 2012
327
I have a unity gain buffer that is AC coupled. on the input and the output. I am sort of confused. Do I need to add a DC bias anywhere? If there is a cap on the input and the output, what is the bias? I ask because the opamp I am using (OPA564) says that I need to make sure that the Vin+ and Vin- need to have a differential below .5V at all times if used as a buffer.
 

R!f@@

Joined Apr 2, 2009
9,734
Unity gain is a buffer. Used to match Z.
OPAMP buffers do not need bias as I recall.

How about showing the diagram. Will help to explain better
 

Papabravo

Joined Feb 24, 2006
13,926
The only way you could be in trouble, is with an AC signal so fast that the output cannot follow to maintain the differential requirement.
 

MikeML

Joined Oct 2, 2009
5,444
Stand on the non-inverting input. Can you find a DC path less than 1megΩ to a voltage that is between the opamp's V+ and V- pins?

Stand on the inverting input. Can you find a DC path less than 1megΩ to a voltage that is between the opamp's V+ and V- pins?

If the answer to either question is No, then you have a floating input... (bad)
 

Papabravo

Joined Feb 24, 2006
13,926
For the inverting input a buffer will have a low impedance path to V- through the output.
I hadn't considered the possibility of the non-inverting input floating to a value outside the supply rails. It would be a bad bad thing.
 

MikeML

Joined Oct 2, 2009
5,444
For the inverting input a buffer will have a low impedance path to V- through the output.
I hadn't considered the possibility of the non-inverting input floating to a value outside the supply rails. It would be a bad bad thing.
Those are two little things to add to your checklist anytime when deploying an opamp...
 

ErnieM

Joined Apr 24, 2011
8,046
One should add the reason for the DC path is to give the input bias current somewhere to go. Without this path it WILL charge the cap to the power rails with consequently bad effects on the output.
 

Thread Starter

Gibson486

Joined Jul 20, 2012
327
Awww man....I feel dumb now....I had no idea you had to add that bias....it seems so simple and obvious... for simplicity, I added a pic of what I did. It is what I described, an op amp follower with a cap at the input and output.
 

Attachments

GopherT

Joined Nov 23, 2012
8,012
Awww man....I feel dumb now....I had no idea you had to add that bias....it seems so simple and obvious... for simplicity, I added a pic of what I did. It is what I described, an op amp follower with a cap at the input and output.
Not so dumb. I just did it myself (so, therefore, it cannot be so dumb). I had a Sallen Key low pass filter that I capacitively coupled it to previous stage. When I realized my dynamic range was shrinking on each power up - about 3 to 5 seconds, I know what my issue was - I just had to trace my circuit to find my floater.


I had capacitor on input to left of R1. Notice that r1 and r2 are not electrically connected to anything.

image.jpg
 
Top