Transmitter Design Updates

[NZ0I]

More changes have been made to the first prototype transmitter design. One serious error has been corrected in the earlier version, and several simplifications have been made. The serious error was in the low-pass filter (LPF) for the 80m transmitter design. Actually, several issues have existed for a long time now:

Error 1: LPF Filter Values Wrong

The inductor values in the schematics have been shown with values that are off by an order of magnitude. They should have been in the range of 2.6uH, not 26uH. The winding numbers and core types were correct, however.

Error 2: Comparison Errors with ON7YD Design 

The baseline LPF design used for performance comparison has been the one used by ON7YD shown here: https://qsl.net/on7yd/atx80.htm. The FCC requires that all spurious emissions be at least 43dB below the power of the fundamental frequency. ON7YD reports measuring the 2nd harmonic as 45dB below the fundamental, and the 3rd and 4th harmonics as 68dB below the fundamental, with all measurements taken at 3.3W of transmit power. Those results should be more than adequate to achieve the FCC's requirements. Modeling the ON7YD design in SPICE (and using the correct component values) suggests that we should expect to achieve only 38 dB below the fundament for the 2nd harmonic, and better than 50 dB below the fundamental at all higher harmonics, assuming we use ideal components. So the SPICE model shows the ON7YD design to be at least in the ballpark of meeting FCC requirements. Testing of actual transmitter hardware has, so far, provided results falling well short of ON7YD's tests, short of modeled predictions, and around 20 dB short of FCC requirements. But that transmitter hardware has not used ON7YD's board design, the components were not identical to those used by ON7YD, and the power level and test setup also differ from what ON7YD used - suggesting that the layout, component tolerances, and test setup might be critical to achieving the needed performance.

Due to the above analysis, I've changed the LPF designs for both 2m and 80m to 7th order Chebychev filters, which theoretically should achieve the target 43dB attenuation of the 2nd harmonic and higher. I'll also try to take greater care with board layout to avoid degrading filter performance. I'm thinking it might be important to remove ground plane layers from around the LPF so as to prevent stray capacitance from affecting filter performance. Other thoughts?

2-Band Transmitter design: https://drive.google.com/open?id=1yk6tQTWNo6adcF94_UHlC6f1EtOyIl8k

The BOM will be updated shortly.