Proposed fix for Harmonics Issue


Howard Fidel
 

After much thought, and reading all the threads, and speaking to Allison (who has not commented to me on this yet) I think I have a good solution to the harmonics issue, not the spurs. We know that the problem is capacitance between input and output of the filters. We estimated that it is 2 pF per a relay. For 80 meters, we have 4 relays, or 8 pF capacitance from input to output. You will note that the filter has a capacitor at its output. This acts as a voltage divider. The bigger we make this cap, the small the energy will be that comes through, since the larger the cap is, the more attenuation we get from the voltage divider. To get a filter with more capacitance we need to lower the impedance of the filter, so I used 25 ohms. This means that we need to rewind T11 to get a 25 ohm output impedance. The redesigned filter for 80 meters uses the same inductors as we now have, we just need to change the capacitors to the new values, and add a transformer to the output. If done on the board, this is the only artwork change. It can also be put in line with the output coax, and then no rework of the PCB artwork is needed. The simulation shows that there is a peak in the filter response caused by the 8 pF cap on the original design at 12.5 MHz and -53 dB. With the new filter this peak is at 10.5 MHz and is - 64 dB. I think a 10 dB improvement is all we need. The same can be done for each of the other filters, and the improvement will be the same, since the capacitance through the relays goes down as the frequency goes up (fewer relays). I have attached the simulation files and the filter schematic. I have to design a transformer. 

Howard

Join BITX20@groups.io to automatically receive all group messages.