[GreenKeys] Fwd: Re: Western Union

[David I. Emery]

On Thu, Apr 25, 2019 at 04:23:59PM -0400, Bruce Gentry via GreenKeys wrote:
> Agreed- but TV was on the air and growing from 1946 onward. In the "good 
> old days" in small towns not on the grid, it was a problem. Many of 
> those stations still used a 60 cycle vertical and 15,750 horizontal rate 
> through the 50s if they  transmitted locally originated B&W exclusively. 
> With a 59.94 vertical rate, a set of poor design or in bad condition 
> could have a bar slowly crawling  the screen with a good steady 60 cycle 
> power source.  Many power systems in the 40s and 50s  may not have been 
> accurate enough for fax machines. The military ones I worked on used a 
> tuning fork oscillator to eliminate power frequency issues.

	Certainly was (AND STILL IS) true that the power line frequency
is not close enough for something like analog fax.   Today is no
different than back then... it can be surprisingly far off over short
periods when the grid is under heavy load - nor is it off by a fixed
amount only slowly varying either.

	At least until quite recently power grid controllers attempted
to keep the number of cycles per day accurate... so electric clocks
would on the average be right... recently in some places in the world it
has been decided that speeding up the grid to catch up after heavy load
has slowed it down costs too much and attempting to keep cycles per
day/week/month right is no longer done there.   And even where it still
is done on really hot days with huge grid wide power demand clocks can
be slow by up to tens of seconds at certain times of day.

	Of course a large part of this is the death of mechanical electric
clocks based on synchronous motors and gearing...

	HOWEVER, not an insignificant number of electronic clocks on
line powered consumer devices don't use accurate crystal oscillators to
keep time, but use the power line as a frequency reference to save a few
pennies on a reasonably accurate crystal rather than a cheap ceramic
resonator as a clock source that may be accurate only to a percent or so
over temperature and time (plenty good enough otherwise for clocking
logic).

	Another curious fact here is that the 59.94/29.97 of NTSC  was a
horrid compromise to avoid having harmonics of the horizontal scanning
frequency interfere with the sound carrier center frequency of 4.5 MHz
in inter-carrier sets and especially to minimize picture interference in
color pictures from the 4.5 sound carrier in the video (sound carrier was
4.5 MHz away  from the video carrier and inter-carrier TVs did not have
a separate IF for the sound and video, but simply extracted the 4.5 MHz
beat between the two carriers out of the video detector which acted of
course as a mixer).

	Pretty obviously it would have trivially inconvenient to
slightly offset the sound carrier frequency of TV stations from the
original 4.5 MHz from the video carrier to avoid this... but some
complete DOFUS decided instead that nothing much would be bothered by
adjusting the basic timing of the whole video clock by 1000/1001 - so
this didn't happen and a permanent headache for video engineers and
equipment and protocol designers and architects was created that
persists to this day.   HD video is still  29.97/59.94 fps in the USA.

	The number of hours, months, years, decades, centuries,
millenia, even eons of expensive engineering (and these days software
engineering) time that has gone ever since into working out ways of
dealing with 29.97 instead of 30 and 59.94 instead of 60 is
incalculable.. particularly when this didn't happen in Europe and much
of the rest of the world which uses 25.0 and 50.0 frames per second
timing nor (for the most part) in Hollywood which used 24 frames per
second displayed 48 or 96 times a second.

	It is essentially impossible to think of any consequences of
moving the offset between the two TV carriers of that era (of VSB analog
TV) that would even have chewed up more than a few months here and there
of engineering time (or service or maintenance or any other time).   At
most before frequency counters were common it might have been a bit more
difficult to adjust this accurately in the field.

	And while existing sets back in 1953 would have had to be very
slightly realigned  for optimum operation very few had filters at 4.5
MHz tight enough to notice...


-- 
  Dave Emery N1PRE/AE, die at dieconsulting.com  DIE Consulting, Weston, Mass 02493
"An empty zombie mind with a forlorn barely readable weatherbeaten
'For Rent' sign still vainly flapping outside on the weed encrusted pole - in 
celebration of what could have been, but wasn't and is not to be now either."