wierd Non linearity w/ passive BP filter

Thread Starter


Joined Feb 7, 2006
I have built a BP filter with passive surface mount components (L and C) on a copper board. the operational specs are fo=155Khz and BW=10Khz. I am using this filter in a setup to filter out the spurious tones that are generated by my synthesizer. From my design I expect to get over 40dB suppression at twice the fundamental freq (i.e. ~320KHz) and my measurements are in excellent agreement with this. However, I find that I dont get much suppression( less than 20dB) of my spur tone at three times the fundamental (i.e. 480Khz). However, when I change my input freq into to the filter from 160 Khz to 480 Khz I get suppression close to 60 dB. I find this really strange because my filter characteristic should be independent of the input tone. Its almost as though the input tone at 160Khz strongly interracts with the spurious tone at 480Khz. Do you think this could be a consequence of non linearity associated with the inductors...
the inductors I am using are 10 uH and 500 uH. Is this a problem thats seen in resonant circuits?
How close are the input and output impedance to resistive over the full 500kHz band? What kind of L's and C's are you using? What type and order of filter is it?