I have a device (1980s era broadcast VTR) that takes a video signal from DC to 4.2 MHz and positive FM modulates a frequency bandpass of 14.18 MHz to 20.0 MHz. Now just before the final video record head driver they divide the FM signal by two using an ECL flipflop. This gives us the industry standard 7.09 MHz to 10.0 MHz bandpass. The reason they modulate at 2X the desired bandpass and then divide down is to ensure a 50% duty cycle (or as close as practical).
I want to clear out any energy below 1 MHz at the record side. Meaning I don't want any energy below 1Mhz recorded on the tape at all. (at all means at least -30db)
Filtering the divided down signal is not practical due to issues I won't get into here*. So if I HPF the 14.18 MHz to 20 MHz signal out the modulator circuit at 2 MHz, will that translate to filtering out below 1Mhz after the divide process, again using a ECL flipflop? Or does the divide process re- introduce its own sidebands?
Also we must consider the FM record head power amplifier after the 2X divider no doubt introduces harmonics on its own.
- The record amplifier, /2 ECL flipflop, and the playback preamplifiers are on a PCB that is mounted on the 5in diameter rotary head drum. It spins at 3600 rpm! FM signals are coupled with rotary transformers and the DC voltages are handled by slip rings. Yes, it's balanced with weights. Even when replacing an electrical component, you must consider upsetting the weight and balance. So mounting an LC filter on this board is a huge mechanical problem!
