Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Switch capacitor filter

Status
Not open for further replies.

Syafiq

Member
Hi

I am doing a project on designing a 5th Order SCF. I need to justify why the SCF is much better than the RC filter? I do not really have much knowledge on this field hence could somebody assist me in explaining the obvious differences and how is SCF better than its other similar types? Thanks!
 
SCF's are better adapted to being built as an integrated circuit since it is easy to make closely matched capacitors in an IC process. They have no particular advantage for a discrete design and are generally inferior for that.

Why are you doing a SCF project if you don't know why it is used? Is this a discrete design?
 
SCF's are better adapted to being built as an integrated circuit since it is easy to make closely matched capacitors in an IC process. They have no particular advantage for a discrete design and are generally inferior for that.

Why are you doing a SCF project if you don't know why it is used? Is this a discrete design?

Thanks for the info. I do know the purpose of Switch Caps is to act as resistors in ICs. However, I am unsure of how to make a comparison between SCF and other filters of similar functions. Besides SCFs, are there any other filters which serves a similar function? Like RC filters etc...:confused::)
 
Syafiq, another unique feature of SC filters is the capability to externally control the pole frequency with the clock rate.
More than that, it is possible to realize very low pole frequencies down to 0.1 Hz, which is not possible using active RC configurations.
As a disadvantage, in many cases you have to provide pre-filtering (anti-aliasing) and for some applications also post RC filtering.

W.
 
Syafiq, another unique feature of SC filters is the capability to externally control the pole frequency with the clock rate.
More than that, it is possible to realize very low pole frequencies down to 0.1 Hz, which is not possible using active RC configurations.
As a disadvantage, in many cases you have to provide pre-filtering (anti-aliasing) and for some applications also post RC filtering.

W.

Thanks for d info. Could you explain the meaning of pole frequencies?
 
Thanks for d info. Could you explain the meaning of pole frequencies?

It is common practice to describe the frequency response based on the location of the transfer functions poles.
For second order functions, the pole frequency is approximately equal to the so called "corner frequency" (end/begin of the path band for lowpass/highpass filters)
In particular, for the Butterworth response, the pole frequency equals the 3-dB frequency.
In case of a (second order) bandpass, the pole frequency is identical to the center frequency.

That means: In SC filters you can externally control the corner frequency and/or the center frequency, respectively.
 
It is common practice to describe the frequency response based on the location of the transfer functions poles.
For second order functions, the pole frequency is approximately equal to the so called "corner frequency" (end/begin of the path band for lowpass/highpass filters)
In particular, for the Butterworth response, the pole frequency equals the 3-dB frequency.
In case of a (second order) bandpass, the pole frequency is identical to the center frequency.

That means: In SC filters you can externally control the corner frequency and/or the center frequency, respectively.

Thanks! So my guess is that the purpose of controllin the corner frequency externally is an advantage as compared to normal RC filters? But what is ultimately the reason to control this corner frequency? For the case of 5th order SCF, I believe that the corner freq is at the 3db point too right? Thanks!
 
Thanks! So my guess is that the purpose of controllin the corner frequency externally is an advantage as compared to normal RC filters? But what is ultimately the reason to control this corner frequency? For the case of 5th order SCF, I believe that the corner freq is at the 3db point too right? Thanks!

The edge of a passband depends on the particular requirements. If you like you can specify the corner also at the -1 dB points.
For example, for Chebyshev responses it is common practice to specify the corner frequency based on the allowable ripple within the pathband (examples: 0.1 dB or 1 dB or 3 dB).

However, two questions:
* You spoke about a SC Filter only. I suppose, you mean lowpass?
* Which topology? I assume: Topology derived from passive RLC structures? (Alternative: Cascade of three separate sections)
 
The edge of a passband depends on the particular requirements. If you like you can specify the corner also at the -1 dB points.
For example, for Chebyshev responses it is common practice to specify the corner frequency based on the allowable ripple within the pathband (examples: 0.1 dB or 1 dB or 3 dB).

However, two questions:
* You spoke about a SC Filter only. I suppose, you mean lowpass?
* Which topology? I assume: Topology derived from passive RLC structures? (Alternative: Cascade of three separate sections)

Yup Lowpass SCF. As for the topology, I am unsure of which I am doing. But I'll have to convert my Passive RC into an Active RC. Thanks!
 
According to my task, I have to convert a passive RC to an active RC.
Do you realize that you can build an active RC filter without it being SC? :confused:
 
According to my task, I have to convert a passive RC to an active RC.

Do you realize that a 5th order passive RC filter (see your question in post#1) has a catastrophic bad frequency response?
Something must be wrong.
By the way: Initially, you spoke about an SC filter!
 
Last edited:
Do you realize that a 5th order passive RC filter (see your question in post#1) has a catastrophic bad frequency response?
Something must be wrong.
By the way: Initially, you spoke about an SC filter!

My project is to design a 5th order SCF. But before i could do so, i am requored to understand the fundamentals of how SCF is better than an RCF. Thanks!
 
My project is to design a 5th order SCF. But before i could do so, i am requored to understand the fundamentals of how SCF is better than an RCF. Thanks!

Hello,


I have to agree with Winterstone. One of the main advantages that was already mentioned was the ability to 'tune' the circuit by altering the clock frequency, and that clock frequency might come from a control circuit that is able to adjust the filter cut off for some near ideal frequency point.

For example look at a single stage RC filter with fixed R and fixed C. The cut off frequency is w=1/RC. There's no way to tune that electronically because R and C are fixed values that can not change. Now the first idea is to make R a potentiometer, so that now we have w changing with the position of the pot wiper arm. But that requires mechanical motion to turn the pot knob, so maybe a human has to do it. You could use a motor (this has been done) but that is more complex.
Now place a switch in series with the resistor R. You can change the apparent R value by turning the switch on and off rapidly, and this only requires a clock signal which an electronic circuit can generate. For a numerical example, say you have a 1k resistor and 1uf cap. With the switch in series with R and with the switch being turned on and off at a high rate relative to the signal and at a 50 percent duty cycle the R looks like 2k now. So we went from a 1k and 1uf cap to a 2k and 1uf cap, so w (the cutoff frequency) changed accordingly. If you calculate that out you'll find that the cutoff frequency changed by a factor of 2, which is quite a bit for a filter to change.
The only real drawback is that we also need at least a post low pass filter to get rid of the clock frequency which gets added to the signal because of the switch. And the clock has to be faster than the fastest signal to be filtered.
 
Thanks for the help guys! I'll do more in-depth reasearch in this unfamiliar field and post more queries if needed.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top