In the last half of the 1970's, I worked in the lab at NRK (the Norwegian broadcaster) while at the university. We had a problem covering football games where the red shorts did not follow the players around the green pitch. To remedy this a 'trouser puller' ('buksepådrager') was created. This was a passive filter that had the opposite group delay and amplitude characteristics as a 500 meter coax cable. One man worked for months selecting capacitors and tuning coils, all mounted on a 50 cm long board with solder tabs. It worked perfectly, and versions were made for different lengths of coax.
Thank you for an excellent video which was most interesting. I used to work on microwave links for TV distribution to the transmitters. One reason for the vestigial sideband is that an SSB-plus-carrier signal suffers from quadrature distortion. Transmitted waveforms having a high modulation index are made to be somewhat pointed at the top. The large amplitude waveforms for TV occur at low frequencies, so it is convenient to retain DSB at low video frequencies. The black level clamping occurs during the back porch, which is used as a reference level. A simpler system just clamps with a diode to the sync tip. The sync pulse stretching is useful if the transmit power amplifier is non linear. Most RF amplifiers tend to compress as the drive voltage goes up, and it is desirable to keep picture/sync ratio at 70/30. Not sure if you mentioned gamma correction, but it happens that a full colour signal might exceed 1V PP on some occasions, and I believe the allowable max is a little greater. Maybe the white clipper is to stop this happening to protect the RF amplifier. In addition to group delay we also used to test for differential gain and differential phase, which are diseases of FM microwave links.
Really good to hear an insight from somebody in the microwave side of things. Quite a difficult area to pry into. Also interesting to hear that VSBs had that additional purpose. I've not come across that!
Thanks for making this. I learnt the theory of analogue broadcasting back at university eons ago and worked in broadcast engineering for a time. Brought back a few memories, though most of the analogue tx equipment was already earmarked for the dumpster and I didn’t really get involved with it. It might be old tech but the RF engineering is still as relevant as it ever was.
@19:21 Sync stretching is used to compensate for sync compression in the linear amplifier chain of your typical broadcast transmitter. The sync tips get compressed somewhat depending on how much headroom your amplifier has. In tube amplifiers, for example, as the tube ages, its peak output capability will drop. Stretching the sync back out can compensate for some of that. Usually this is done in a processing amplifier (proc amp) ahead of the modulator.
back in the 80's colour televisions had a Chroma delay pot to bring the colours in time with the luma. the Baird 9000 series was the first I recall with SAW filters, it was literally a glass panel with a led input and a photo detector on the output.
Really interesting. Thanks for putting so much effort into making this. Always fascinating to read comments from folks who used to work in the industry adding context and extra info, so thanks to them too!
The white clip was set to 103% IRE here in Denmark, if the video modulation was to high, you would get video am noise in sound, some older transmitters could not handle more than 105%, so they shut down if the videosignal was to high. Here in DK the transmiters was operated by the Danish Post and Telegraph in the older times, we just delivered the signal to them with correct dc level, and correct sync/videolevels. They must have had the IF modulator locally placed at our MCR area, because when we added NICAM stereo the NICAM encoder also was in the MCR area.
You just got a new subscriber, as a kid from the 70s, analogue TV is so nostalgic for me... I remember (UK based) NICAM 728 digital stereo ... (Near Instantaneous Companded Audio Multiplex, 728 KBPS)..., analogue satellite tv..., it was soooo much fun! I was a subscriber to Television Magazine here in the UK, sadly in the dying days of TV repair shops.
im from a tv repair background so this is far to complex for me but still interesting. tvs had a "black level clamp" or "dc correction" or "black restorer" never heard the term pedistal clamp
@@MePeterNicholls That is my impression too. Pedestal and black level are the same in 625 line video. In NTSC there is a difference between the blanking level around the line sync and the black level. They use the term pedestal for the blanking level which is slightly less than the black level. Of course, this is ancient memory, and I may be mistaken.
I was reading through video demystified by keith jack and on figure 8.21 it shoes rf modulation implementation. It looks like AM, so why are there different frequencies present/prominent during v sync? blanking and field/lines?
I thought that VSB was there to aid in demodulation, so that you could decode the signal by rectifying it, as least for the low frequency portion. SSB modulation requires synchronizing the phase, and SSB radio tuners (for audio, in amateur radio applications mostly) have a separate knob for dialing in that phase, which wouldn't be very practical. VSB makes it possible to infer that phase from the signal. Or that's what I thought at least, you're saying that it's just because of filter imperfections... I would also love to see the difference made by enabling and disabling the group delay compensation feature on an actual TV screen (or a capture setup, whatever). I suppose it'd probably be rather minuscule, but it would still be interesting to "pixel peep" into that.
I won't dispute what you have said. I wasn't an expert on this topic when I recorded the video. It is true that its impossible to design a filter which would leave none of the upper sideband behind. The Tg correction is quite minimal. It's about +/- 100ns depending on frequency.
@@mattstvbarn 100ns is a little over 1 pixel when sampled at the standard 13.5 MHz frequency. So it would definitely be subtle, but probably perceptible. Would still like to see it, but the number is good insight, thanks!
I am so in awe of finding this channel... I am a mere sparky who knows a bit about RF (enough not to use a 30A Junction box to join 75 ohm co-ax, LMAO ... yes i've seen that)... it's a handy string to add to my bow, to be able to get a good signal for people.... sadly even that is dying now, 'no need for a tv aerial, i have wifi'... 😢
What is the standard video system IF frequency in Europe? In the USA. Canada, Mexico it's 70mhz FM. That goes for point to point microwave and satellite systems.
@@renejensen5656 I understand the max possible deviation. But what was the standard IF frequency for FM video in transmission links? In the USA and other NTSC countries, we too had a 15khz max for TV audio and FM radio. But the consumer TV FM audio IF was 4.5mhz. The FM radio IF was 10.7mhz. And the IF for FM video/audio microwave and satellite transmission was 70mhz. So did Europe use 70mhz as well?
@@andydelle4509 Im not sure about our microlink system, i know that we had some orange one, cant remember the manufacture. There was one video and 4 audio carriers on the units. But they went out of service arround 1994 because we started to use satelitte uplink in KU band. And this was DVB-S QPSK digital MPEG, later DVB-S2 in 8PSK MPEG4 and APSK16 MPEG4. We still have 2 almost new DSNG vans in operation, just in case that the mobil network are down. For our daily news standup on location we use a bonding mobil broadband encoders, like Awivest, Live-U etc.
Group Delay? Some of these audiophool's worry about group delay in their speaker cables! I kid you not! Yes, it does occur with audio frequencies on paper. But really?
Wow that has to be comment of the week! I dip in and out of the audiophile space but I have never seen that discussed before. I am wondering what kind of instrumentation could possibly measure Tg at AF ? on a cable !?
In the last half of the 1970's, I worked in the lab at NRK (the Norwegian broadcaster) while at the university. We had a problem covering football games where the red shorts did not follow the players around the green pitch. To remedy this a 'trouser puller' ('buksepådrager') was created. This was a passive filter that had the opposite group delay and amplitude characteristics as a 500 meter coax cable. One man worked for months selecting capacitors and tuning coils, all mounted on a 50 cm long board with solder tabs. It worked perfectly, and versions were made for different lengths of coax.
Thank you for an excellent video which was most interesting. I used to work on microwave links for TV distribution to the transmitters. One reason for the vestigial sideband is that an SSB-plus-carrier signal suffers from quadrature distortion. Transmitted waveforms having a high modulation index are made to be somewhat pointed at the top. The large amplitude waveforms for TV occur at low frequencies, so it is convenient to retain DSB at low video frequencies.
The black level clamping occurs during the back porch, which is used as a reference level. A simpler system just clamps with a diode to the sync tip.
The sync pulse stretching is useful if the transmit power amplifier is non linear. Most RF amplifiers tend to compress as the drive voltage goes up, and it is desirable to keep picture/sync ratio at 70/30.
Not sure if you mentioned gamma correction, but it happens that a full colour signal might exceed 1V PP on some occasions, and I believe the allowable max is a little greater. Maybe the white clipper is to stop this happening to protect the RF amplifier.
In addition to group delay we also used to test for differential gain and differential phase, which are diseases of FM microwave links.
Really good to hear an insight from somebody in the microwave side of things. Quite a difficult area to pry into. Also interesting to hear that VSBs had that additional purpose. I've not come across that!
Thanks for making this. I learnt the theory of analogue broadcasting back at university eons ago and worked in broadcast engineering for a time. Brought back a few memories, though most of the analogue tx equipment was already earmarked for the dumpster and I didn’t really get involved with it. It might be old tech but the RF engineering is still as relevant as it ever was.
@19:21 Sync stretching is used to compensate for sync compression in the linear amplifier chain of your typical broadcast transmitter. The sync tips get compressed somewhat depending on how much headroom your amplifier has. In tube amplifiers, for example, as the tube ages, its peak output capability will drop. Stretching the sync back out can compensate for some of that. Usually this is done in a processing amplifier (proc amp) ahead of the modulator.
Ah thank you. I have waited a long time for that explanation!
back in the 80's colour televisions had a Chroma delay pot to bring the colours in time with the luma. the Baird 9000 series was the first I recall with SAW filters, it was literally a glass panel with a led input and a photo detector on the output.
Really interesting. Thanks for putting so much effort into making this. Always fascinating to read comments from folks who used to work in the industry adding context and extra info, so thanks to them too!
So glad I’ve discovered this channel!
The white clip was set to 103% IRE here in Denmark, if the video modulation was to high, you would get video am noise in sound, some older transmitters could not handle more than 105%, so they shut down if the videosignal was to high. Here in DK the transmiters was operated by the Danish Post and Telegraph in the older times, we just delivered the signal to them with correct dc level, and correct sync/videolevels. They must have had the IF modulator locally placed at our MCR area, because when we added NICAM stereo the NICAM encoder also was in the MCR area.
You just got a new subscriber, as a kid from the 70s, analogue TV is so nostalgic for me... I remember (UK based) NICAM 728 digital stereo ... (Near Instantaneous Companded Audio Multiplex, 728 KBPS)..., analogue satellite tv..., it was soooo much fun! I was a subscriber to Television Magazine here in the UK, sadly in the dying days of TV repair shops.
im from a tv repair background so this is far to complex for me but still interesting. tvs had a "black level clamp" or "dc correction" or "black restorer" never heard the term pedistal clamp
I believe pedestal and black level were interchangeable
I recall the better B&W tv’s having a black level clamp. A PAL tv needs a black level clamp to render the colour properly.
@@MePeterNicholls That is my impression too. Pedestal and black level are the same in 625 line video. In NTSC there is a difference between the blanking level around the line sync and the black level. They use the term pedestal for the blanking level which is slightly less than the black level. Of course, this is ancient memory, and I may be mistaken.
I was reading through video demystified by keith jack and on figure 8.21 it shoes rf modulation implementation. It looks like AM, so why are there different frequencies present/prominent during v sync? blanking and field/lines?
I thought that VSB was there to aid in demodulation, so that you could decode the signal by rectifying it, as least for the low frequency portion. SSB modulation requires synchronizing the phase, and SSB radio tuners (for audio, in amateur radio applications mostly) have a separate knob for dialing in that phase, which wouldn't be very practical. VSB makes it possible to infer that phase from the signal. Or that's what I thought at least, you're saying that it's just because of filter imperfections...
I would also love to see the difference made by enabling and disabling the group delay compensation feature on an actual TV screen (or a capture setup, whatever). I suppose it'd probably be rather minuscule, but it would still be interesting to "pixel peep" into that.
I won't dispute what you have said. I wasn't an expert on this topic when I recorded the video. It is true that its impossible to design a filter which would leave none of the upper sideband behind. The Tg correction is quite minimal. It's about +/- 100ns depending on frequency.
@@mattstvbarn 100ns is a little over 1 pixel when sampled at the standard 13.5 MHz frequency. So it would definitely be subtle, but probably perceptible. Would still like to see it, but the number is good insight, thanks!
@@kFY514 all listed in here. About page 25. github.com/inaxeon/PTV_Preservation/blob/main/PM5580/Docs/PM5580_Manual.pdf
I don’t recall any group delay compensation at the transmitters in the uk
I am so in awe of finding this channel... I am a mere sparky who knows a bit about RF (enough not to use a 30A Junction box to join 75 ohm co-ax, LMAO ... yes i've seen that)... it's a handy string to add to my bow, to be able to get a good signal for people.... sadly even that is dying now, 'no need for a tv aerial, i have wifi'... 😢
Till the wifi doesn't work and i get to install a cat 6 run for a new AP , swings and roundabouts
I would expect that. I have read that it was not ever done in the UK.
What is the standard video system IF frequency in Europe? In the USA. Canada, Mexico it's 70mhz FM. That goes for point to point microwave and satellite systems.
Since the max teoretical are 15khz audio plus emphasis that would make a FM modulated signal arround 35 khz bandwide at testlevel 0dB (0,775mv).
@@renejensen5656 I understand the max possible deviation. But what was the standard IF frequency for FM video in transmission links? In the USA and other NTSC countries, we too had a 15khz max for TV audio and FM radio. But the consumer TV FM audio IF was 4.5mhz. The FM radio IF was 10.7mhz. And the IF for FM video/audio microwave and satellite transmission was 70mhz. So did Europe use 70mhz as well?
@@andydelle4509 Im not sure about our microlink system, i know that we had some orange one, cant remember the manufacture. There was one video and 4 audio carriers on the units. But they went out of service arround 1994 because we started to use satelitte uplink in KU band. And this was DVB-S QPSK digital MPEG, later DVB-S2 in 8PSK MPEG4 and APSK16 MPEG4. We still have 2 almost new DSNG vans in operation, just in case that the mobil network are down. For our daily news standup on location we use a bonding mobil broadband encoders, like Awivest, Live-U etc.
Maybe one day in future when current digital delivery network gets the bandwidth and efficiency of analog, can reintroduce this engineerin
Group Delay? Some of these audiophool's worry about group delay in their speaker cables! I kid you not!
Yes, it does occur with audio frequencies on paper. But really?
Wow that has to be comment of the week! I dip in and out of the audiophile space but I have never seen that discussed before. I am wondering what kind of instrumentation could possibly measure Tg at AF ? on a cable !?
analog FPV/crappy crappy transmitters in the 5ghz realm is still a thing