Register to reply

Low-pass filter with unideal op-amp

by xortan
Tags: filter, lowpass, opamp, unideal
Share this thread:
NascentOxygen
#19
Dec16-12, 11:10 PM
HW Helper
Thanks
P: 5,356
At 100Hz the gain is < 10⁴, placing it more than 20dB down.

It seems (based on my brief reading) that the BIAS MODE is largely directed towards reducing the differential offset towards the ideal. This hardly seems important to most filter apps, so I think you'd be choosing the bias that maximizes gain around 1MHz. Low bias will stretch GB to 1.7MHz
NascentOxygen
#20
Dec17-12, 05:39 AM
HW Helper
Thanks
P: 5,356
Here's some semilog graph paper showing how the logarithmic cycle is divided into 10 equal steps. http://www.science-projects.com/SemiLog0.GIF

Revising my estimates: gain is about 3x10⁵ at 1Hz, and 7x10⁴ at 100Hz.

The GB product changes to some extent with supply voltage VDD and BIAS. The graph you provide is for VDD =5V. So that's the supply voltage you will be using? (See Fig 27 of the datasheet I referred to.)
xortan
#21
Dec17-12, 10:51 AM
P: 59
Yeah that is the supply voltage I am using, I understand how to read log scale, I was looking at a different graph then the one I post (sorry). Even if I change the bias mode I still have no idea how to look at the graph provided in the datasheet and modify the ideal response I plotted in matlab. That's all I want to know.
yungman
#22
Dec17-12, 11:48 AM
P: 3,898
Excuse me in making a comment that is not related to the question here.

As I practice design engineer, this is not a LPF I would design.
1) If I want a 1st order LPF with gain of 1 only, I use a simple RC, not with opamp.
2) You are having pole frequency of something like 10KHz from the graph, at frequency where the opamp unity GBWP of below 1MHz, you are only -30dB down. You are going to see the effect of the opamp.
3)If you really want to do this, I question the opamp choice. Look at the bode plot, the second pole of the opamp kicks in before unity gain, you have extra phase shift that likely cause a peak close to cross over. The situation might be made worst because you are working in the gain of << 1 because you only start at gain of 1!!!! At least find an opamp that has higher GBWP that absolutely have a single pole cross over.

I know this is a practice circuit, but you still want to think about real world design. Look at a simple second order active filter:

http://en.wikipedia.org/wiki/Sallen%...93Key_topology

In the diagram, C2 is at the input of the opamp, this will roll off the high frequency before even going into the opamp. This will mitigates the peaking effect for the opamp and produce better result.

I am not saying this won't work, I did not do simulation, but there is potential problem just by looking at the circuit.
NascentOxygen
#23
Dec19-12, 05:38 AM
HW Helper
Thanks
P: 5,356
Quote Quote by xortan View Post
Even if I change the bias mode I still have no idea how to look at the graph provided in the datasheet and modify the ideal response I plotted in matlab. That's all I want to know.
Are you giving matlab the circuit, or the transfer function? (I don't know matlab.)

If you want to simulate the circuit, in place of the ideal differential amplifier use that ideal differential amplifier followed by a low-pass filter (R and C) followed by an ideal non-inverting buffer amplifier (perhaps having an output impedance). Then around that as your amplifier, you arrange your filter's passive components.
carlgrace
#24
Dec19-12, 02:48 PM
P: 555
Quote Quote by NascentOxygen View Post
Are you giving matlab the circuit, or the transfer function? (I don't know matlab.)

If you want to simulate the circuit, in place of the ideal differential amplifier use that ideal differential amplifier followed by a low-pass filter (R and C) followed by an ideal non-inverting buffer amplifier (perhaps having an output impedance). Then around that as your amplifier, you arrange your filter's passive components.
Matlab deals with transfer functions. It has no schematic capture.

A better tool for exploring circuits is LTspice, and it's free! http://www.linear.com/designtools/software/#LTspice


Register to reply

Related Discussions
High pass filter and low pass filter question? Engineering, Comp Sci, & Technology Homework 6
Transforming a low pass filter to a band pass and low pass filter Engineering, Comp Sci, & Technology Homework 0
Transforming a low pass filer to a high pass filter Engineering, Comp Sci, & Technology Homework 0
High Pass filter & Low Pass Filters features Engineering, Comp Sci, & Technology Homework 1