Low Pass Filtering in Op Amps...

[Ricky]

Trying to use a differential op amp with low pass filtering in the inverting configuration has always seemed limited. If the cap is across the feedback resistor, at the highest frequencies it can only provide as much attenuation, as the op amp provides gain. When the feedback resistor is shorted by the capacitor, the circuit becomes a voltage follower. Not so effective as a low pass filter.

I found these designs in a TI app note. Page 39, figures 9 and 10.

https://www.ti.com/lit/an/slyt119/slyt119.pdf

The problem is, they use differential outputs. However, I don\'t think differential outputs are required. In this design, each output can be from a different op amp, where one pair of inputs is connected to the other pair of inputs, but reversed. This will make one output inverting and the other not.

The circuit in mind will have a gain of about 4, so the cross coupling of DC offsets should not be significant. The inputs and outputs are AC coupled and the only DC concern is keeping as close to center of the power rails so the AC is not too far off center, causing clipping near one rail long before the other. I would need to provide a DC operating point by adding DC load resistors after the input blocking caps.

Anyone see problems with this approach?

--

Rick C.

- Get 1,000 miles of free Supercharging
- Tesla referral code - https://ts.la/richard11209

 

[bitrex]

On 3/4/2023 4:46 PM, Ricky wrote:

Trying to use a differential op amp with low pass filtering in the inverting configuration has always seemed limited. If the cap is across the feedback resistor, at the highest frequencies it can only provide as much attenuation, as the op amp provides gain. When the feedback resistor is shorted by the capacitor, the circuit becomes a voltage follower. Not so effective as a low pass filter.

I found these designs in a TI app note. Page 39, figures 9 and 10.

https://www.ti.com/lit/an/slyt119/slyt119.pdf

The problem is, they use differential outputs. However, I don\'t think differential outputs are required. In this design, each output can be from a different op amp, where one pair of inputs is connected to the other pair of inputs, but reversed. This will make one output inverting and the other not.

The circuit in mind will have a gain of about 4, so the cross coupling of DC offsets should not be significant. The inputs and outputs are AC coupled and the only DC concern is keeping as close to center of the power rails so the AC is not too far off center, causing clipping near one rail long before the other. I would need to provide a DC operating point by adding DC load resistors after the input blocking caps.

Anyone see problems with this approach?

Here\'s something like a differential in, single-ended out Sallen-Key if
that\'s helpful, long out of patent protection:

<https://data.epo.org/publication-server/document?iDocId=112950&iFormat=2>

 

[Ricky]

On Saturday, March 4, 2023 at 11:41:49 PM UTC-5, bitrex wrote:

On 3/4/2023 4:46 PM, Ricky wrote:
Trying to use a differential op amp with low pass filtering in the inverting configuration has always seemed limited. If the cap is across the feedback resistor, at the highest frequencies it can only provide as much attenuation, as the op amp provides gain. When the feedback resistor is shorted by the capacitor, the circuit becomes a voltage follower. Not so effective as a low pass filter.

I found these designs in a TI app note. Page 39, figures 9 and 10.

https://www.ti.com/lit/an/slyt119/slyt119.pdf

The problem is, they use differential outputs. However, I don\'t think differential outputs are required. In this design, each output can be from a different op amp, where one pair of inputs is connected to the other pair of inputs, but reversed. This will make one output inverting and the other not.

The circuit in mind will have a gain of about 4, so the cross coupling of DC offsets should not be significant. The inputs and outputs are AC coupled and the only DC concern is keeping as close to center of the power rails so the AC is not too far off center, causing clipping near one rail long before the other. I would need to provide a DC operating point by adding DC load resistors after the input blocking caps.

Anyone see problems with this approach?

Here\'s something like a differential in, single-ended out Sallen-Key if
that\'s helpful, long out of patent protection:

https://data.epo.org/publication-server/document?iDocId=112950&iFormat=2

Thanks. I need this, but I also need single in and differential out. That can always be handled by a differential in and out by grounding one input. I don\'t actually need two poles. I\'d be happy with a more simple circuit with one pole and two outputs.

The figure 9 circuit in the document I provide a link for, seems very simple and does what I want as far as I can tell. I can\'t think why this would not work using two op amps, each with a single output. They share the same inputs, but one has the two inputs swapped wrt to the other. That will give the inverting output.

--

Rick C.

+ Get 1,000 miles of free Supercharging
+ Tesla referral code - https://ts.la/richard11209