A 3.5 to 4.0 MHz 5-Resonator, 0.2 dB Ripple, 50 Ohm Band Reject Filter

Just suppose you are a serious short wave listener, but you live next door to this ham radio operator that was forever using the 80/75-meter ham band. He is just too strong to allow your receiver to operate when he is transmitting. To fix this problem, lets design a band reject filter that rejects the amateur frequency range, but allowed us to listen on either side, right up to the ham band. The topology of this filter is is similar to the band pass transformed filter, but with series and parallel resonant circuits reversed. Thus we have alternating series-tuned circuits connected by parallel-tuned circuits. Each of these circuits places a null in the middle of the ham band, and the design part is to provide an equal ripple response on either side of the rejected band. The design printout for the BR filter, from LCFIL3A.BAS is:

The component values for the 5 resonant circuits are at the bottom of the printout. See the low pass filter description for assistance with the exponential notation of part values.

The schematic show all of the component values:

We can simulate the overall response:

This plot shows the insertion loss in a 50-Ohm system as the red curve, MS21 or "magnitude of S21>" It also shows in blue the magnitude of the reflected wave, MS11. This latter quantity, shown in dB is a measure of the impedance match looking into one port of the filter with the other port terminated in 50 Ohms. The curves correspond to lossless components.

At first glance, this looks like a band pass filter. But, the red curve is the MS21 and it has much attenuation from 3.5 to 4.0 MHz! Again, component losses will cause the most ainsertion loss at the edges next to the stop band.

Click here to return to the main LCFIL3A page.

This page was last updated and Copyrighted 11 December 2013, Robert S. Larkin

Please email comments or corrections to bob 'the at sign' janbob dot com