Hey guys,

I have a few true or false questions about the attached active band pass filter. Problem is all the books we have only cover active low/high pass filters and nothing about the bands - and everything I've researched so far hasn't really covered the characteristics of the filter I'm looking at. The questions are as follows:

- The gain is constant from DC to 3180Hz bandwidth.
- Vout = Vin× (RF / R1) for DC signals.
- Attenuation rate is -40dB/decade.
- Signals with v(t) = sin 2000πt are not blocked by the circuit.
- Voltage (VA-VB) is amplified differentially across a large bandwidth.
- Attenuation rate is -20dB/decade.
- An EMI noise signal of 50Hz is cut by the circuit.
- A Noise signal of 350Hz is cut by the circuit.

I'm pretty sure the Attenuation rate is -20dB/decade since its only a first order? Apart from that I don't know where to really look. If you guys could point me in the right direction that would be awesome.

Kind regards,
Max

 

hi Max,
Download this free TI Active Filter Pro tool, very useful tool.

http://www.ti.com/tool/filterpro

 

Thanks mate, I'll give it a go

EDIT: Just had a look. Few problems - we weren't given any values for our components, so I don't know what to really set for the specifications and the topologies provided don't match my circuit. :\

 

[/FONT]- An EMI noise signal of 50Hz is cut by the circuit.
- A Noise signal of 350Hz is cut by the circuit.
[/FONT]

Hola Eric,

"Cut" in this context, does it mean "attenuated"?

 

hi
These two documents should get you started, post your calculations and we can talk you thru the design.

As its homework we can only gives hints/clues.

http://www.google.co.uk/url?sa=t&rc...G6RdMeqMnsK2wwQ&bvm=bv.64542518,d.ZGU&cad=rja

 

Hola Eric,

"Cut" in this context, does it mean "attenuated"?

Hola at,

I would read 'cut' as attenuation

Eric

 

It's just an opamp circuit. Analyze it as an opamp circuit. What is the transfer function?

Now apply your knowledge about transfer functions to the questions being asked.

Give your best shot at each question, explaining your reasoning in terms of the analysis of the circuit as just an opamp circuit. Then we can discuss where you are right and where you are going off the rails.

 

We have never been asked to derive a transfer function. They were always given to us (Still first year). I did however find a schematic for my circuit with said transfer function. I still don't see how I would apply that to questions such as "A Noise signal of 350Hz is cut by the circuit."

Thank you for your patience.

 

We have never been asked to derive a transfer function. They were always given to us (Still first year). I did however find a schematic for my circuit with said transfer function. I still don't see how I would apply that to questions such as "A Noise signal of 350Hz is cut by the circuit."

Thank you for your patience.

Since we are dealing with a filter, cut, to me, means attenuated/blocked/removed.

Furthermore. We are dealing with bandpass filter. This means that there are 3 regions. On a frequency plot from left to right: Region 1 low frequency signals are blocked, Region 2 bandpass region where signal of desired frequencies are passed through, Region 3 high frequency signals are blocked.

You are given one signal that is not blocked, this signal has angular frequency of 2000*pi. w=2000*pi. But w=2*pi*f. So f=1000 Hz. So signal of 1000 Hz is in the bandpass, Region 2, this signal pass through.

350 Hz is less than 1000 Hz and we are asked to block this signal. So it must lie in Region 1, the low frequency signals that are blocked/attenuated/cut/filtered out.

 

Hey guys,

I have a few true or false questions about the attached active band pass filter. Problem is all the books we have only cover active low/high pass filters and nothing about the bands - and everything I've researched so far hasn't really covered the characteristics of the filter I'm looking at. The questions are as follows:

- The gain is constant from DC to 3180Hz bandwidth.
- Vout = Vin× (RF / R1) for DC signals.
- Attenuation rate is -40dB/decade.
- Signals with v(t) = sin 2000πt are not blocked by the circuit.
- Voltage (VA-VB) is amplified differentially across a large bandwidth.
- Attenuation rate is -20dB/decade.
- An EMI noise signal of 50Hz is cut by the circuit.
- A Noise signal of 350Hz is cut by the circuit.

I'm pretty sure the Attenuation rate is -20dB/decade since its only a first order? Apart from that I don't know where to really look. If you guys could point me in the right direction that would be awesome.

Kind regards,
Max

* The input capacitor will block dc so the first question is false
* The second question needs a negative sign in front of it, so false
* You're right about it being a first order circuit so it attenuates at -20dB
*There is no way of knowing the pass band without values for the resistors and capacitors. This means the question about those lower frequencies are unknowable, too.

 

* The input capacitor will block dc so the first question is false
* The second question needs a negative sign in front of it, so false
* You're right about it being a first order circuit so it attenuates at -20dB
*There is no way of knowing the pass band without values for the resistors and capacitors. This means the question about those lower frequencies are unknowable, too.

Somehow, from somewhere, he came up with a circuit with component values (Post #8).

But, PLEASE, try to offer hints and suggestions instead of just answering the questions for him. This is HIS homework, not yours.

 

Somehow, from somewhere, he came up with a circuit with component values (Post #8).

But, PLEASE, try to offer hints and suggestions instead of just answering the questions for him. This is HIS homework, not yours.

I think that signal that is not blocked might be a clue.
Signals with v(t) = sin 2000πt are not blocked by the circuit.
that is signal at 1000 Hz. We could then assume that 1000 Hz is the center frequency of the pass band.

 

I think that signal that is not blocked might be a clue.
Signals with v(t) = sin 2000πt are not blocked by the circuit.
that is signal at 1000 Hz. We could then assume that 1000 Hz is the center frequency of the pass band.

Why would you make such an assumption based on one of eight true/false questions?

 

I think that signal that is not blocked might be a clue.
Signals with v(t) = sin 2000πt are not blocked by the circuit.
that is signal at 1000 Hz. We could then assume that 1000 Hz is the center frequency of the pass band.

But we don't even know the v(t) signal passes through. As I understood it the questions are true or false. Even if it does pass through we don't know the lower corner frequency, which is necessary to know whether or not the 50 Hz for 350 Hz signals will pass. We need RC values.

 

Sorry for interrupting, WBahn. I'll lay off.

 

Sorry for interrupting, WBahn. I'll lay off.

No need to lay off, just try not to give answers outright (though sometimes it comes down to that being the only way, but I don't know if we are at that point yet on this one).

 

The circuit in post 8 was just one I found on the net that had the transfer function like you asked. The circuit in my question has no values and just asks to "pick the correct properties for the above circuit" - but thats what confuses me like PRS said...we need values to determine if 50Hz or 350Hz pass. Thats why I've just been researching this exact circuit thinking that maybe its a common characteristic of this circuit to pass/block those signals... Think I might email my lecturer for clarification.

 

Definitely ask for clarification. Depending on the component values chosen, any particular frequency might be below the passband, in the passband, or above the passband. My guess is that either specific component values where intended to be given, or that the allowable answers included true, false, and indeterminate.