Science desk
< March 3	<< Feb \| March \| Apr >>	March 5 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

March 4

How does vertical blanking exist in analog television if it scans horizontally only?

From this slow motion video, it looks like electron sweeps from left to right horizontally but not vertically. How does vertical blanking exist in analog television if it scans horizontally only? Rizosome (talk) 12:08, 4 March 2022 (UTC)[reply]

If by "analog TV" you're meaning an old fashioned CRT display then it's quite simple. There are two oscillators which generate a sawtooth waveform at line frequency and at double frame frequency. The output from the frame oscillator is applied to the top and bottom plates of the tube, and the line frequency to the left and right hand plates. The spot scans from left to right (and slightly downhill). When the sawtooth drops the line flyback circuitry blanks the beam so you don't see a zig-zag. At the end of the half frame the frame oscillator sawtooth drops and it also blanks the spot for a few lines. during this inter-frame period data can be transmitted, this is how teletext worked. As the oscillators flyback there is a sudden change in the current through the oscillator coil and this can be used to generate the EHT, which for a large colour set can be in excess of 25kV. I mentioned half frames, scanning is interleaved to get better quality. Martin of Sheffield (talk) 12:28, 4 March 2022 (UTC)[reply]

Is it possible for the sawtooth waveform to become audible? (to the naked ear) I think they might've hummed at low volume the entire time they were on but it's been so long I forget. Sagittarian Milky Way (talk) 15:20, 4 March 2022 (UTC)[reply]

Yes, with poor design you can get interference into sound. It would be at 50 Hz (roughly G₁) in the UK or 60 Hz (B₁) in the US since the frame frequency is usually half mains frequency. If you were young and with good hearing you could also sometimes hear a high pitched note that was the line frequency, 11 kHz for 425 line and 16 kHz for 625 line in the UK. Martin of Sheffield (talk) 15:58, 4 March 2022 (UTC)[reply]

The 405-line System A had line frequency 10125 Hz (=405•25). The 625-line System B has line frequency 15625 Hz (=625•25). Philvoids (talk) 14:06, 6 March 2022 (UTC)[reply]

Oops, not sure where the extra kHz crept in, and I've thrown away the scrap of paper I used. In case it isn't obvious I was rounding to the nearest kHz. Martin of Sheffield (talk) 14:33, 6 March 2022 (UTC)[reply]

Here is a lovely whitepaper from Texas Instruments: Troubleshooting TI PSR Controllers. (TI has many other white-papers that are specific to the characteristics of individual parts - including old-fashioned power controllers and more "modern" digital controllers that have special protection-circuits designed just to suppress the audible noise; and other vendors surely have similar publications). The learnings from the "old-fashioned" high-voltage cathode-ray tube devices can also be applied to other electronic circuit scenarios; but the old CRT power systems had difficult constraints that made their audible-noise problems worse than modern-day-devices: flyback transformers for CRTs tend to be high-power and high-voltage. These large voltage- and power- swings, occurring at oscillation frequencies within the audible range, can couple into thermally-induced and electromechanically-induced vibrations in capacitors and ferrite-cores. These issues can be mitigated mechanically (the old trick of pouring glue on everything!); or by selecting different kinds of parts; or by changing the characteristics of the electrical impulses so that they would couple less strongly into a mechanical vibration.

Individuals with sensitive hearing can often hear the power-supply switching frequencies in small, low-wattage AC/DC and DC/DC power-supplies - even though it's much smaller amount of power than a CRT flyback transformer, some low-power power-supply designs still leak a relatively large amount of energy in to those same kinds of mechanically-coupled losses that result in acoustic noise. (I can still hear the tinny whine emitted by certain models of AC/DC and DC/DC power supplies commonly used in several small house-hold and consumer-electronics- products like certain WiFi routers - and my ears are getting older every day!)

Nimur (talk) 14:15, 5 March 2022 (UTC)[reply]

You can see on that very video you linked the whole screen area is beeing swept vertically by the horizontal scan line, can't you? So it is scanned in both direction, just with diffeent frequencies/ --CiaPan (talk) 12:31, 4 March 2022 (UTC)[reply]

Specifically, watch that video starting at 2:53. Clearly, each next line swept from left-to-right is lower than the previous one, precisely like how people read a page from a book line by line. In analog television the sweeps work the same, except that each field (a scan of the screen) covers only one half of the lines, alternating between the even lines and the odd lines. A vertical synchronization pulse marks the end of each field and is the first of a sequence of pulses covering the vertical blanking interval: during the brief time needed to make the electron beam sweep up, it is blanked. The details differ depending on whether NTSC or PAL is used, but the principle is not different of how horizontal blanking works. --Lambiam 14:09, 4 March 2022 (UTC)[reply]

I highly recommend the YouTube channel Technology Connections. He's got a very good series on video technology, and goes into excellent detail. He's easy to understand, and has an absolutely biting sense of humor. --Jayron32 12:07, 4 March 2022 (UTC)

What exactly is "double frame frequency"? Rizosome (talk) 15:50, 7 March 2022 (UTC)[reply]

2*<frame frequency> The lines are interlaced so that (very, very, approximately) the TV displays 1 3 5 ... 401 403 405 2 4 6 ... 400 402 404. 625 PAL and NTSC are similar, but with a different number of lines. Therefore the scan is done a double frame frequency so that two scans (the odd and then the even) occur for each frame. Martin of Sheffield (talk) 16:13, 7 March 2022 (UTC)[reply]

The duration of a frame is 40 ms so frame frequency is 1/40ms = 25 Hz. A frame comprises two interlaced fields so field frequency is double frame frequency i.e. 50 Hz. (In USA and countries using 60 Hz mains supply the frame/field frequencies are 30/60 Hz respectively.) The lines scanned in each field are:

Odd field 1 3 5 7 9 11 13 15 17 19 21 23 25 27....625

Even field 2 4 6 8 10 12 14 16 18 20 22 24 26 ...624

All lines are scanned but only the underlined lines 26 to 625 display video. The preceding low-number lines are hidden above the CRT picture area and they are prevented from illuminating by the vertical blanking pulse. It has been common practice to exploit this invisible Vertical blanking interval for test lines or communications such as Teletext. Philvoids (talk) 22:22, 7 March 2022 (UTC)[reply]