Active bandpass filters with low Q

[bert nobbins]

I have been trying to make an active bandpass filter of the Multiple Feedback type with a low Q, say 0.5, but for any Q less than 0.707 (1/root2) the design equations give nonsense resistor values. Is 0.707 the minimum Q possible with this configuration, and I am attempting the impossible? I have read all the filter books I can but have found no answer to this question.

If it is impossible, could I get a Q of 0.5 with some other kind of bandpass filter??

It only has to work over the audio bandwidth. I would be very grateful for any help.

 

[LvW]

It's very uncommon to use an active band pass filter with a Q as low as 0.5 because you can realize it with a passive network.
However, it should be possible with the MFB topology.
If you have problems in finding the part values, use a filter design program (available via internet).

 

[keith1200rs]

Yes, it can be done. Try Nuhertz Technologies "filter free" software. The Texas Instruments Filterpro also does it. However, it is not much of a filter! Depending on what you are trying to achieve a combination of a LP & HP might be better.

Keith.

 

[bert nobbins]

It's very uncommon to use an active band pass filter with a Q as low as 0.5 because you can realize it with a passive network.quote]

That's a thought... the Wien network has a Q of 1/3, but I really need 0.5 or close to it. Do you know of any passive networks that can achieve that?

 

[LvW]

.........
the Wien network has a Q of 1/3, but I really need 0.5 or close to it. Do you know of any passive networks that can achieve that?

The WIEN network has Q=1/3 only in case of R1=R2 and C1=C2.

When you need a higher Q try the following:

R1=kR2 and C1=C2/k with k=1 for the classical (symmetric) WIEN network.

In this case T1=R1C1=T2=R2C2=T and 2*Pi*Fo=1/T
and Q=T/(2T+T/k).

For k=10 you have for example Q=1/2.1.

Is that close enough to 0.5?