I designed a LPF with the requirements below; everything seems to work fine using MULTISIM but when I test the circuit in the lab it gives me different results. <Please see my attachment of my design>

Specifications:
Output impedance; Zo = 50 ohms +-20% at 5 KHz
Input impedance; Zi = 50 ohms +-20% at 5 KHz
3 dB cutoff frequency = 20 Khz
Roll of of at least -6dB

ANY SUGGESTIONS WILL BE GREATLY APPRECIATE IT!

 

Hi,

I can't comment on Multisim, as I don't have or use it. I plugged your parameters into a designer at http//www-users.cs.york.ac.uk/~fisher/cgi-bin/lcfilter and got different values for components. For the caps, use .16 microfarad and the inductor value is .8 mH. I checked that the corner is at 20 Khz, but not the rolloff. With three poles, the response has to be better than 6 dB/octave.

The Zi & Zo values are set by R1 & 2, so worrying about the impedance at 5 kHz is not too useful, unless you want to mess with wirewound - and therefore inductive - resistors.

 

Thanks a lot; that helped a lot but am not sure why the amplitude starts at -6 db and not at 0?

 

Hi,

Not quite right. Signal amplitude at frequencies below the cut-off may be measured and set to 0 dB as a reference level. Then, at the knee (or operating point), the signal amplitude will be down by 3 dB. That is at the 20 kHz frequency. The signal will decrease by 6 dB/octave as the frequency increases past the filter's operating point.

 

Consider where the output is with reference to your two resistors ...