Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Regarding switch capacitor based low pass filter

Status
Not open for further replies.

Pawanshekarr

Newbie level 6
Joined
Jun 19, 2017
Messages
13
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
100
Hey everyone...
Iam currently doing a project related to switched capacitor based low pass filter which has an attenuation of more than 72dB at 500KHz. I have questions regarding the effect of non linearities related to finite op-amp bandwidth and finite op-amp gain and also regarding bilinear transformation. It would be really helpful if can get some tips regarding this topic or if any of you guys have come across a document related to this, Please do share it with me, It would be really helpful.
Thanks in advance.
 

Hi,

Why "switched capacitor" instead of pure analog?

Usually the Opamp is at the output of a SC filter. Therefore it just has to amplify the reduced bandwidth.
But for sure you have to choose a suitable Opamp for this.

Klaus
 
For answering your question it would be best (if not necessary) if you would show us the circuit under discussion.
This is because the influence of all non-idealities (active and passive) depends on circuit topology.
 
Hi,

Why "switched capacitor" instead of pure analog?

Usually the Opamp is at the output of a SC filter. Therefore it just has to amplify the reduced bandwidth.
But for sure you have to choose a suitable Op-amp for this.

Klaus
Hi , Iam using switched capacitor because the integrating frequency is a function of capacitor ratios and can be finely tuned. It is also possible to achieve higher attenuation by using switched capacitor. One question in my mind is that what if my input signal is in the same range of the op-amp's f3dB pole or the dominant pole, wont it undergo an attenuation according to frequency response of the op-amp used?
 

 

Attachments

  • Screen Shot 2017-06-20 at 13.58.51.png
    Screen Shot 2017-06-20 at 13.58.51.png
    38.6 KB · Views: 95

Here are the main requirements for an S/C filter - as far as the amplifier requirements are concerned:
* Very high input impedance (small parasitic capacitor discharging),
* Very high gain (mainly to maintain the virtual ground at the inverting terminal)
* Sufficient large-signal bandwidth (slew rate) for exact switching operation .

In addition you should not overlook (a) parasitic resistive behaviour of the capacitors and (b) realistic (non-ideal) resistive properties of the switches.
More than that, deviations from the ideal filter properties are to be expected due to the bilinear transformation.
 
Switched capacitor lowpass filter ICs have been available for at least 20 years. Instead of making one why don't you buy one? Many years ago I used National Semi ICs in a very low distortion generator and distortion measuring circuit. They worked perfectly.
 

Regarding switch capacitor output response

hello everyone, I have a query regarding my output of a switched capacitor low pass filter response .(Chebyshev 6th order The specifications are attenuation greater than -72dB at a stopband frequency at 500kHz, cutoff frequency of 50kHz . I have hereby attached my output at cut off frequency i.e 50kHz and also the output at 500kHz. I want to know the reason for infinite attenuation as my output. Any kind of help is appreciated.please helpcutoff.png500k.png200k.png Thanks in advance
 
Last edited by a moderator:

Hi,

I want to know the reason for infinite attenuation as my output.
What do you mean with "infinite" attenuation?

I'd say a scope is no useful method to show signals with attenuation of 40dB and more. It's beyond resolution.

Klaus
 
I'm too missing a clear question.

We see a transient reaction on the sine burst. You may want to check in a simulation if the response is already expectable according to the ideal filter transfer function or if it's caused by non-ideal circuit parameters.

You should also mention the filter sampling frequency.
 
hello klausST, I would like to acheive a SNR of -72dB at the output. Is it possible that i plot the output signal in matlab and check my results?

- - - Updated - - -

The filter sampling frequency is 10Mhz. I require an attenuation of -72dB or less at the stopband frequency. I have taken the coefficients of a chebyshev filter. Iam really unaware how to determine the coefficients on my own.
 

actually 1.65v is my DC common mode. So my output is going below that at the stopband frequency so Iam worried about this situation
 

actually 1.65v is my DC common mode. So my output is going below that at the stopband frequency so Iam worried about this situation
Yes, you see an AC output signal, it's swinging around 1.65V, just normal operation.

I guess you don't fully understand why you see a damped oscillation after abruptly starting the 500 kHz input. But that's normal operation too, it happens also with a continuous time filter and isn't related to the SC implementation.

You need to measure the stop band attenuation in steady state, after the transient oscillation has decayed.
 

Yes i did measure my output at around 500us. Its constant at almost 1.6499976v
 

I would measure input and output AC voltage (e.g. Vpp) after sufficient delay and calculate the attenuation.
 
But Iam not getting an AC output. Its almost constant at DC reference of 1.65v(1.6499976V to be preceise)
 

Hi,

But Iam not getting an AC output. Its almost constant at DC
What else do you expect from a low pass filter with 72 dB attenuation?

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top