building a bandpass filter with ...

Thread Starter

1233

Joined Dec 26, 2009
10
I am asked to design a bandpass filter give the selectivity, the center frequency, and the passband gain.

Q= 10
fo = 5MHz
Passband gain Ko=1

Edit: using cascaded 1st order stages.

I don't know where to start from.

thanks in advance.
 
Last edited:

ABoul

Joined Mar 30, 2009
15
i'm in a bit of a hurry so this is the best i can do: http://en.wikipedia.org/wiki/Q_factor

go under the section titled "Physical interpretation of Q" (more specifically, the response equation, noting that s = jw) and the first part of the next section. if you play about with the equations, you should get values for R, L and C.

EDIT: by inspection, LC = (2 * pi * 5000000)^2.
 

Thread Starter

1233

Joined Dec 26, 2009
10
thx, I will have a look at that

edit : forgot to mention that I should use cascaded 1st order stages.
 
Last edited:

Thread Starter

1233

Joined Dec 26, 2009
10
I think I should build the filter using inductor, capacitor and resistors only. I am sure that it is not so difficult, but I don't know where to start from
 

Nanophotonics

Joined Apr 2, 2009
383
It's OK to design filters using RLC, but I think it's not practical to build it. The range of values available for L is limited and there are various other disadvantages. I would probably go for opamps.
 

PRS

Joined Aug 24, 2008
989
I am asked to design a bandpass filter give the selectivity, the center frequency, and the passband gain.

Q= 10
fo = 5MHz
Passband gain Ko=1

Edit: using cascaded 1st order stages.

I don't know where to start from.

thanks in advance.
Start by considering the bandwidth: BW = fc/Q

This, obviously is 500kHz
So your f(low is 4.5MHz) and f(high)=5.5MHz.

With the instructions to use "stages" I'd guess you're allowed to use transistors for amplification in order to offset the attenuation of the filter.

A single common emmiter stage should do.
Use the input capacitor to obtain f(low) and at the output put a capacitor in parallel with the load to get your f(high). To calculate the C use the input resistance in series with the source resistancefor fL and and the output resistance in parallel with the load resistance for fH. Put these in the equation C = 1/(2*Pi*R*f).

At least, this is how I understand the problem. Better ask your instructor for more details.
 

Wendy

Joined Mar 24, 2008
23,429
I'm betting not, it just means you have to predict the 1.5db points, and adjust accordingly.

Generally filters use T attenuators to offset differences in terminations. A 6db attenuator will have a 12 db return loss.
 
Top