INA217 output coupling cap?

[therecordingart]

http://www.ti.com/lit/ds/symlink/ina217.pdf

I'm pretty much copying the design from the data sheet, but the output isn't going to have the DC servo. I want to run the output into a TL071 non-inverting input.

How big of a coupling cap should I use? Should there be any series resistance?
The output z of the opamp  isn't  listed so I don't know how to calculate it. Any help is appreciated.

 

[joaquins]

Not very big, depends on the resistor you use at the input of the opamp to bias it. It makes an RC HPF. As you could make that resistor as big as 10M for the TL071 the cap value isn't a problem. Maybe pick a nice cap which fits there, any good film cap, see what value is smallish and easy to get, and then calculate the resistor for a LPF at 2Hz, so you don't get phase problems in band from this filter, or you could use this as your HPF for your preamp, changing resistors or caps.

JS

 

[therecordingart]

Not very big, depends on the resistor you use at the input of the opamp to bias it. It makes an RC HPF. As you could make that resistor as big as 10M for the TL071 the cap value isn't a problem. Maybe pick a nice cap which fits there, any good film cap, see what value is smallish and easy to get, and then calculate the resistor for a LPF at 2Hz, so you don't get phase problems in band from this filter, or you could use this as your HPF for your preamp, changing resistors or caps.

JS

Thank you. Here is a crude drawing of what I'm thinking:

http://imgur.com/BAStbJq

The math on just the RC filter puts the corner freq at 1.59hz, but I'm not sure what impact the INA217 or TL071 has on that.

 

[joaquins]

That should be good, you still need a fixed reference at the INA but I think it's the pin 5 which is already grounded, check with the data sheet.

JS

 

[PRR]

> The output z of the opamp  isn't  listed

Unless there is some other need, the output of a feedback amp is "zero", or realistically 0.1 Ohm bass 10 Ohms top of audio band.(*)

> The math on just the RC filter puts the corner freq at 1.59hz, but I'm not sure what impact the INA217 or TL071 has on that.

1 Meg plus the 0.1 Ohm of the '217 shifts your -3dB point to 0.999,999,9 of the 1 Meg alone, which is still 1.59 for *any* practical purpose.

We can assume the '072's input is over 100Meg for low-frequencies (we often ignore it in 1  Gig condenser mike head-amps). Worst-case (100 meg) your -3dB shifts up 1%, still same-as for *any* practical purpose.

(*)Take TL072. It is about 300 Ohms on the inside. This is reduced by the amount of NFB applied. Speaking roughly, '072 is a 10MHz GBW chip and the audio band is not far past 10KHz. Gain on most all-purpose opamps falls directly as frequency. Working unity-gain, since 10MHz is 1,000X 10KHz, the opamp has 1,000X of excess gain, output Z is reduced 1,000X, and 300r/1,000 is 0.3 Ohms. Working at gain of 10, excess gain is 100X, 300r/100 is 3 Ohms. Yes you are working the '217 at much higher gains, but it is made for that, has even more excess gain when the pot is turned small. I would expect "<10r" to be pretty near right, and also much-much smaller than any reasonable load.