10-12-2019, 02:14 PM
One correction to my last post, the output sound trap filter is actually an IIR filter. I mistyped FIR filter.
Thinking about this, I suppose there is a chance that with the 1.x thru 3.x converters where the default image was stored individually during final test, even though the source for the image has always been the same, there could be variances in the capture. I can't imagine there could be much of a difference, but it is certainly possible. During final test the RF output from the converter is viewed on a Sony 9-90UB for overall quality and not necessarily the resolution of the 3MHz grate. Another item of note, on these units with the LC sound trap filter, this was manually adjusted while the output was viewed on a spectrum analyzer. There is a built-in test procedure in these units that creates an unmodulated carrier at the sound RF frequency to aid in the adjustment.
With the 4.x converters, barring component tolerances, everything is identical from unit to unit. The test card and the sound filter are now digital and will always be the same. It is even possible to store the default images completely digitally as they are just uncompressed 8 bit data in the serial flash, monochrome for most formats and 4:2:2 color for the color units. Since this flash also shares the FPGA code, care would need to be taken to not overwrite this code.
Thinking about this, I suppose there is a chance that with the 1.x thru 3.x converters where the default image was stored individually during final test, even though the source for the image has always been the same, there could be variances in the capture. I can't imagine there could be much of a difference, but it is certainly possible. During final test the RF output from the converter is viewed on a Sony 9-90UB for overall quality and not necessarily the resolution of the 3MHz grate. Another item of note, on these units with the LC sound trap filter, this was manually adjusted while the output was viewed on a spectrum analyzer. There is a built-in test procedure in these units that creates an unmodulated carrier at the sound RF frequency to aid in the adjustment.
With the 4.x converters, barring component tolerances, everything is identical from unit to unit. The test card and the sound filter are now digital and will always be the same. It is even possible to store the default images completely digitally as they are just uncompressed 8 bit data in the serial flash, monochrome for most formats and 4:2:2 color for the color units. Since this flash also shares the FPGA code, care would need to be taken to not overwrite this code.
