NTSC information confirmation

Discussion in 'General Electronics Chat' started by nDever, Mar 29, 2011.

  1. nDever

    Thread Starter Active Member

    Jan 13, 2011
    154
    4
    I am designing a basic circuit to produce a composite NTSC video signal. Before I get started, I need to make sure that I have aligned all of the facts. Am I correct in assuming these? Correct me if I am wrong if any of these are incorrect.


    • A horizontal scan begins about every 63.5μs
    • The colorburst is about 3.58Mhz
    • The horizontal and vertical sync signals have negative polarity.
    • The white signal level is 100 IRE or 714mV
    • The blanking signal level is 0 IRE or 0V
    • There are 525 scan lines in a progressive and interlaced scan
    Additionally, I have some questions.

    • How do I produce a negative voltage?
    • How should is the vertical sync signal be integrated in the signal; after the 525th horizontal sync signal?
    • What is the formula that determines the color produced? Does it involve phase angle of the colorburst?
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
  3. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    You can get IC's to do this - http://www.epanorama.net/circuits/vga2tv/vga2palntsc.html

    Colors do depend on a phase difference from the oscillator, which gets resynched with every color burst. However, any technician will tell you that NTSC stands for "never the same color".
     
  4. nDever

    Thread Starter Active Member

    Jan 13, 2011
    154
    4
    So, if the color burst is the reference signal (thus is a constant signal) for which the oscillators phase is compare to, what is it that makes the color vary in the same horizontal scan? There can only be one color burst per line, correct?
     
  5. bribri

    Member

    Feb 20, 2011
    133
    5
    discrete components? micro-controller?


    no. generally there are 486 lines of visible picture, put the V sync in the left over space.
    i think that you actually need two V synch pulses per frame due to interlacing but i might be wrong there. interlacing is weird.


    dual +/- power supply. but there are specialized video ic's (opamps and buffers and such) which will run on a single supply.
    i actually think that you can cheat and let your sync pulses bottom out at zero volts and it will still work... but maybe don't quote me on that one.

    does this help?
    [​IMG]

    another way to 'cheat' in terms of colour sub-carrier modulation might be found in the AD725 chip.
     
  6. nDever

    Thread Starter Active Member

    Jan 13, 2011
    154
    4
    Discrete components.
    In a way, this gives me a different perspective of the color model. So now, if I and Q constantly have a 90 degrees phase difference, and the color burst is always 3.58Mhz, is it the luma that changes the instantaneous color?
     
  7. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    The color burst supplies a reference signal to set the oscillator phase. This permits decoding of the phase-modulated sub-carrier information present during the lines.
     
  8. bribri

    Member

    Feb 20, 2011
    133
    5
    i kind of thought that it was an amplitude modulation of what essentially is the luma.

    ...if you figure it out... let me know!


    do check this:
    http://www.ntsc-tv.com/index.html
     
  9. bribri

    Member

    Feb 20, 2011
    133
    5
    ah:

    [​IMG]
     
  10. resistar

    New Member

    Jun 13, 2010
    1
    0
    This is pretty interesting!!! You guys have more links on NTSC and broadcast video???
     
  11. bribri

    Member

    Feb 20, 2011
    133
    5
    i'd be very interested in following your progress on this!
    am also curious about the application of such a device : )
    i've built a number of video synthesizers, but these are generally limited to (the perhaps more forgiving world of) VGA, SVGA, etc.

    my only real contact with colour subcarriers thus far has been through hacking s-video cables.
    actually, if you haven't been doing this already, i'd recommend using a scope to explore some standard (y/c) s-video sources. the fact that your luma and chroma+burst can be observed on separate channels should be most educational.
     
  12. nDever

    Thread Starter Active Member

    Jan 13, 2011
    154
    4
    Does anyone know where I can find a graph representing the displayed colors with respect to the oscillator phase difference?
     
  13. bribri

    Member

    Feb 20, 2011
    133
    5
  14. bribri

    Member

    Feb 20, 2011
    133
    5
    ha!
    i just remembered,
    it's apparently possible to make an NTSC monitor sync with a schmitt-trigger, 4040, germanium diode and a few odds and ends.
     
  15. nDever

    Thread Starter Active Member

    Jan 13, 2011
    154
    4
    Fascinating. Do tell.

    It appears to me that I will have a master frequency which will be input into the 4040. The correct fraction of that frequency will be emitted from one of the outputs of the counter. But the pulsewidth of the outputted signal will be 50%...extremely long.
     
  16. bribri

    Member

    Feb 20, 2011
    133
    5
    i'll have to dig up the schema.
    if i remember rightly, somewhere in the neighbourhood of 15KHz is divide by 262 via 4040 and some AND gates. the other schmitt triggers alter the pulse width i think.
    was kind of a fast and dirty device for making a simple test pattern.... but it's so simple : )
    but not in colour...
     
  17. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,741
    759
    This is interesting thread... Old memories are surfacing...it's been like 10 years since I last said anything about a composite signal to anyone. If you don't teach ...you tend to forget. And I have forgotten most of them..
     
  18. bribri

    Member

    Feb 20, 2011
    133
    5


    here:
    Generator provides 537 NTSC pattern

    http://www.edn.com/archives/1998/120398/25di.htm#generator
    so no colour carriers but pretty cool none the less.
     
  19. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    When you read about the details of NTSC colour TV signals you will see I and Q modulation of fairly low frequency big blobs of colours on the screen. Each colour has a certain phase from the colour burst phase. Details in the picture are high frequency black and white because our vision cannot see small details in colour.

    For the past 15 years, Never The Same Colour has not occured with my analog NTSC TVs. They don't even have a "tint" control anymore.

    Commercials are louder than ever before. TVs should be called Never The Same Volume.
    One Japanese TV had an Automatic Volume Control. It sounded awful.
     
Loading...