How should I improve my circuit for squaring waveform input?

E

elonjia

Newbie level 6
Joined
May 6, 2024
Messages
11
Helped
1
Reputation
2
Reaction score
1
Trophy points
3
Activity points
85


the circuit is just like a bandpass amplifier above, and when I input a square waveform, the output is just like the yellow curve; how should I fix this?
 

Sort by date Sort by votes
By nature of bandpass filter, if the input waveform is a square wave, the output waveform will be different. Not clear what you want to achieve.
 

Upvote 0 Downvote
FvM said:
By nature of bandpass filter, if the input waveform is a square wave, the output waveform will be different. Not clear what you want to achieve.
Click to expand...
So it's good in terms of the waveform? I want to output to be the square waveform either, should I increase the bandwidth?
 

Upvote 0 Downvote
The whole point of a bandpass filter is to attenuate low and high frequencies. There's no way you can expect fast edges AND attenuated high frequencies. Fast edges=high frequency.
 

Upvote 0 Downvote
barry said:
The whole point of a bandpass filter is to attenuate low and high frequencies. There's no way you can expect fast edges AND attenuated high frequencies. Fast edges=high frequency.
Click to expand...
Thanks, could I move this high f to a higher freq so it has a better output performance? Also, is there any recommendation for the high-frequency circuit structure?
 

    FvM

    Points: 2
    Helpful Answer Positive Rating
Upvote 0 Downvote
Obviously, the lower (high-pass) cut-frequency is above fundamental input frequency, thus no chance to reproduce the waveform. I'd try factor 10 lower cut-off frequency (factor 10 larger coupling capacitors) as first step. Upper cut-off frequency doesn't seem to be critical.
 

Upvote 0 Downvote
what “noise” are you trying to filter? Maybe you just need a low pass filter, not a band pass. with zero requirements it’s impossible to give an intelligent answer. what’s your rise time requirement? what’s your pass and tprequirement?
 

Upvote 0 Downvote
crutschow said:
I suggest a low-pass filter to remove high frequency noise (if needed), followed by a high-speed analog comparator (for example MCP6566) to generate a good square-wave.
Click to expand...
A comparator with 80nS rise time isn’t going to give a very good square wave at 1MHz.

Again, with no specs, it’s pretty difficult to say WHAT OP needs.
 

Upvote 1 Downvote

Upvote 0 Downvote
Thanks for your reply, I wonder if I can use a high pass filter so the square waveform can be passed better?
BTW, do you also have any ideas for the question below? Thanks!
https://www.edaboard.com/threads/the-input-reference-noise-voltage-is-larger-than-vdd.412442/
 

Upvote 0 Downvote
A high pass filter will pass a square wave better than a low pass, but it won't do anything to reduce high frequency noise. What, exactly, are you trying to accomplish? It's enormously unclear.
 
Upvote 0 Downvote
The input is a 500 kHz square wave with at least 50 MHz bandwidth
Your Op Amp lacks the bandwidth to pass this signal even with a partial differentiator arrangement and does not meet your expectations and degrades the signal. So it is not clear why you are doing this.

Even a logic gate would work better to buffer a square wave.

We have no idea what you want to do but this is not likely to be improved unless you wanted a sine wave RC filter at 500 KHz or any odd harmonic..
 
Last edited:

Upvote 0 Downvote
You must log in or register to reply here.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…