How do you explain that the dark spot remains stationary after sync is lost (tail end of the video), Hugo? That's what caught my eye.
Not saying that you're wrong, Hugo--my eyes could be playing tricks on me. It wouldn't be the first time.
The blanking pulses are not derived from sync directly, they are derived mainly from the H & V scan output stages, so it does not matter if there is a video input to the monitor or not, if there is a H blanking pulse defect it will still be there in the scanning raster in the same place, why it is not there for all H scanning lines I am not sure yet.
This sort of thing relates to the age old question, when a defect is seen in the image, is the defect in the Picture Content or is it in the Scanning Raster. In this case, the fact it remains there, with no signal, means it is in the scanning raster.Which also means something could be physically obstructing the beam, but when I have see that its always a fixed unchanging shadow.
Looking at the form of it, it looks like the H blanking pulses are abnormal just after the start of scan to give that tear drop like effect where the beam is being cut off in that zone. Probably, where the H & V blanking pulses are mixed, there will be a faulty capacitor. It depends on the design, some sets use a dedicated blanking amplifier. So we need the schematic.
Edit : and pondering it more, if the vertical blanking signal was modulating the H blanking, then even de-synchronized with no video input, the defect would stay in the same place, starting just after H and V retrace in the upper left of the image.
This is just an initial hunch of course.