View Single Post
Old 24th Aug 2019, 10:09 pm   #17
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Screen Tearing on Pye BX1016 on Video playback

Practically all horizontal AFC (automatic frequency control) systems sample a signal from the H output stage and compare that to the phase of the sync pulse to generate a "filtered DC control voltage" for the H scan oscillator. It is the same concept as a PLL.

Since the H oscillator becomes DC controlled (not triggered by individual H syncs as it was in early TV sets) there is a limit to the speed at which it can respond, say if the H sync phase suddenly changes after head switching in a video signal from a VCR. If the control voltage can change fairly rapidly, say after a few scanning lines the tear in the image is not too bad, but if its 10 or 20 lines or more there can be a big tear.

Often the physical head alignment in some VCR's isn't wonderful and there is a bigger phase error at head switching. In other words the heads are not exactly physically 180 degrees apart which is how the phase error comes about, but as noted the tape back tension also generates a phase error if it is different on the recording and playback machinery. But, on the same machine therefore, it is still possible to have the big tear on a record-playback because of the head alignment. You will notice head drums have a hole leading down beside each head, this is used, with a special tool, to move the head position on the perimeter of the drum. (Don't attempt that if you are not trained to do it, its tricky)

If you look at the components at determine the DC filtered control voltage in the AFC there are generally three. A main filter capacitor, and in parallel with that a larger value capacitor with a series resistor. These two series components across the main filter cap form an "anti-hunt network" to prevent the H osc hunting around when locked.

In most cases it is just a matter of altering the time constant of that filter network ,by reducing the value of the capacitors, to speed up the reaction time of the H oscillator. Usually it does not affect the result with a stable broadcast signal, perhaps only make the H lock system slightly less noise immune. And you should get a much better result with the signal from the VCR.
Argus25 is offline