Bryce,
Since all three clocks are changed when you do the calibration, adjusting to the first clock output does not ensure that the rig is calibrated (as your experience shows).? The other two can shift the frequency.? The 55005kHz value is approximate, with the 5khz offset to place the output of the first mixer in the range of the 45MHz roofing crystal filter between the first and second IF.?
I only know how to calibrate the uBitx by zero-beating an external signal source.? Adjusting the BFO to get the low-frequency response moves the SSB crystal filter response such that the other two clocks will put the signal in the range that can be used for calibration.? The calibration is adjusting the first and second mixer LO input simultaneously through the master clock frequency compensation.? Note it is not changing the frequency of the actual clock to the Si5351 clock generator.? It is doing calculations in the program that offset the setpoint of the clocks (all three) that have the same effect.
I have read procedures that use WSJT-X programs or audio spectrum programs to do the calibration.? This involves changing the input calibration to be 1kHz higher than the target to show the 1kHz audio signal generated on the audio program.? Using this method does not require you to lower the BFO frequency response, as the test tone is in the 300 to 3kHz range of the SSB filter.
Do not try to get too exact.? A scaling factor in the calibration can limit the accuracy to 10 Hertz (I have not done the math, so that this value could be off).? ?My point is that if you get it below 100Hz error, I would call it good.
Let me know if you can get the calibration where you want it.
73
Evan
AC9TU
