Op-Amp Integrator Montage (Biased Signal)

Thread Starter

rfpd

Joined Jul 6, 2016
101
upload_2017-8-2_16-9-11.png

I only found this schematic, I didn't find any 'real' scenario since the negative part would be cut off. What should I add do this so that I can integrate a biased square wave?
 

Thread Starter

rfpd

Joined Jul 6, 2016
101
I've seen them, doesn't mention how to do it to a biased signal. And what's the advantage in using this instead of a simple low pass filter.
 

OBW0549

Joined Mar 2, 2015
3,566
I only found this schematic, I didn't find any 'real' scenario since the negative part would be cut off.
It won't be cut off if you power the op amp from plus and minus supplies rather than merely + and ground.

What should I add do this so that I can integrate a biased square wave?
A negative supply, is all.

By the way, what's the difference between an integrator and a low pass filter?
With a low pass filter, the magnitude of the filter's response levels off below a certain frequency (the "cutoff" frequency) and remains constant from there all the way down to DC. An integrator has no cutoff frequency and its response continues to increase with decreasing frequency. In a perfect integrator, this continues down to DC, where the response becomes infinite; however, in practice every op amp has a maximum DC gain, and this limits the gain of any practical integrator as the frequency approaches zero.
 

Thread Starter

rfpd

Joined Jul 6, 2016
101
It won't be cut off if you power the op amp from plus and minus supplies rather than merely + and ground.


A negative supply, is all.


With a low pass filter, the magnitude of the filter's response levels off below a certain frequency (the "cutoff" frequency) and remains constant from there all the way down to DC. An integrator has no cutoff frequency and its response continues to increase with decreasing frequency. In a perfect integrator, this continues down to DC, where the response becomes infinite; however, in practice every op amp has a maximum DC gain, and this limits the gain of any practical integrator as the frequency approaches zero.
I would need a dual supply to do that, it's not pratical since I would use later on a 9V transformer.

So a low pass filter would have differente results for 100 Hz and 300 Hz (just an example), and the integrator would have the same results, but both with amplitude differences (depending on the frequencies).
 

OBW0549

Joined Mar 2, 2015
3,566
I would need a dual supply to do that, it's not pratical since I would use later on a 9V transformer.
In that case you will have to "lift up" the entire operating range of the integrator by connecting the (+) input of the op amp to some positive reference voltage (for example, Vsupply/2 from a resistive voltage divider) and considering that reference voltage as your "virtual ground."

So a low pass filter would have differente results for 100 Hz and 300 Hz (just an example), and the integrator would have the same results, but both with amplitude differences (depending on the frequencies).
I have no idea what you mean by the above. The difference between a low pass filter and an integrator is exactly as I described in post #5 above.
 

bertus

Joined Apr 5, 2008
22,270
Hello,

Assuming the input signal is a squarewave of about the same amplitude as the powersupply, you could use the following circuit:

opamp_integrator_with_offset.png

Bertus
 

Thread Starter

rfpd

Joined Jul 6, 2016
101
In that case you will have to "lift up" the entire operating range of the integrator by connecting the (+) input of the op amp to some positive reference voltage (for example, Vsupply/2 from a resistive voltage divider) and considering that reference voltage as your "virtual ground."


I have no idea what you mean by the above. The difference between a low pass filter and an integrator is exactly as I described in post #5 above.
For example, using a three stage low pass filter for a 10 hz square wave, I get this:

upload_2017-8-2_19-59-8.png

For 40 Hz, I get this:

upload_2017-8-2_19-59-52.png

I'm asking if for 10 Hz I would get a good sine wave like for 40 Hz using an integrator.

I'll look into the virtual ground, thanks!
 

OBW0549

Joined Mar 2, 2015
3,566
For example, using a three stage low pass filter for a 10 hz square wave, I get this:

For 40 Hz, I get this:

I'm asking if for 10 Hz I would get a good sine wave like for 40 Hz using an integrator.
That depends entirely on the filter's cutoff frequency, which you haven't specified; if it's less than or equal to the frequency of your square wave, you can get a reasonable approximation of a sine wave on the filter's output. Also, the more filter sections you cascade, the better job the filter will do of attenuating the square wave's harmonics and thereby leaving you with a clean(er) sine wave.
 

Thread Starter

rfpd

Joined Jul 6, 2016
101
Hello,

Assuming the input signal is a squarewave of about the same amplitude as the powersupply, you could use the following circuit:

View attachment 132125

Bertus
I have a square wave that has 9V or 0V.

upload_2017-8-2_21-9-55.png

It seems to be cropping the wave.Uses 1uf capacitor and an 8k resistor.

Tried lots of values and it seems to always crop it.
 

bertus

Joined Apr 5, 2008
22,270
Hello,

What opamp are you using?
What offset voltage did you create?
By playing with the voltage divider on the + input, you should get a symetric output signal, assuming the input signal has a 50/50 dutycycle.

Bertus
 

Thread Starter

rfpd

Joined Jul 6, 2016
101
That depends entirely on the filter's cutoff frequency, which you haven't specified; if it's less than or equal to the frequency of your square wave, you can get a reasonable approximation of a sine wave on the filter's output. Also, the more filter sections you cascade, the better job the filter will do of attenuating the square wave's harmonics and thereby leaving you with a clean(er) sine wave.
I'm using three filter sections, the volume is low and it sounds good, but when I amplify it, the lower frequencies get fuzzy. So I want to try something that has better filtering across the frequencies (same for all, not having some frequencies fuzzier then others).
 

Thread Starter

rfpd

Joined Jul 6, 2016
101
Hello,

What opamp are you using?
What offset voltage did you create?
By playing with the voltage divider on the + input, you should get a symetric output signal, assuming the input signal has a 50/50 dutycycle.

Bertus
upload_2017-8-2_21-20-9.png

In this simulation it's the 741, and at the input I get a square wave 50% 10V, 50% 0 V, with variable frequency.

EDIT: Oh of corse, must be it, didn't remember I was inputing 9V. I'll check it out.
 
Top