Re: buildd and buildbot
On Sunday 19 May 2019 03:02:00 pm Brian wrote:
> On Sun 19 May 2019 at 09:26:15 -0500, David Wright wrote:
> > But I think you just enjoy reliving your fights.
>
> Not quite. Gene is an engineer. Engineers are simple, uncomplicated
> and practical people. If something malfunctions, it is a fault in the
> principles or the basic laws of Physics. Einstein, bah. Newton rules,
> OK.
The difference between a mechanical and an electrical engineer is that
Einstein does rule. Because mass is a function of velocity in terms of
v** you spent 25 or 30 years listening to a buzz in the audio of any uhf
television station. All because the transmitters of the day all used the
one amplifying device they had in those earlier years, the high power
klystron. Now for the last 15 or so years, they've used a much higher
efficiency tube called a klystrode by most makers. The technology moves
on.
The klystron had one major flaw that proved Einstein was right when he
wrote E=MV2. Because it was 1, a very high gain device that amplified a
small signal by modulating not the amplitude of a stream of electrons,
but changed the velocity of that stream so that nominally 4 feet later
they were all bunched up, and you could pick that power back out of the
beam which was the input signal amplified by 30 db, eg 1 watt in, and
30kw out. But it took 120 kw in dc power to make that electron beam in
the first place, so the power bills were horrendous.
That dc power was in the range of 20 kilovolts and nearly 6 amps in the
beam. So this beam was moving fast enough that E=MV** manifested itself
because the electron gained mass as you sped it up, and lost mass as you
slowed it down, driving it to maximum bunching lowered the average beam
velocity because there was a difference in how much you could speed it
up(less) and how much you could slow it down(more), introducing an FM
component we called Incidental Carrier Phase Modulation, ICPM for its
acronym. Because the last tv set that actually had a true FM receiver
for its sound was made in '48 or '49, since then they have all recovered
the fm sound by useing the difference between the picture carrier, and
the sound carrier which is 4.5 megahertz, the tvs for ntsc all heard the
resultant apparent buzz caused by the video's ICPM in the sound. Now the
uhf transmitters are all using a much more efficient ampliier that
doesn't suffer from that distortion near as much.
But that buzz you may remember from many years past, was caused because
Albert was right. FWIW, the difference in power put in, minus the
nominally 28% pulled back out meant that the beam had to be defocused at
the outout end of one of these tubes so it would spread its heat out
into a big copper funnel that took 65 to 75 gallons of pure water a
minute to cool it, and failure of the pump supplying that water would
let the beam burn thru that copper in about 1/2 a second, filling the
tube with water and making $120,000 for a new one from Varian into $3.00
worth of junk copper, but you had to break the copper away from all the
beryllium ceramic that was also part of that tube. And you don't do that
as one particle in a lung is lung cancer, no if ands or buts. Do not
ever abrade a white, or in later years pink ceramic. Its a death
sentence.
I found that out while I was running what was then KXNE-TV 19 for the
Nebraska ETV commission back it the 70's when a 3 phase, 40 amp Heineman
breaker failed and single phased the the water pump moving that water. I
was standing at the control panel and recognized the sound of a big 3
phase motor with a locked rotor, and my hand was halfway to the off
button about 24" away when the building went dark as the main breaker
cleared like a 12 ga shotgun. One junk klystron. I found a flashlight,
cleared the breakers on the transmitter, reset the building entrance
breaker, and spent the next several hours rigging the remaining aural
tube by rolling the visual magnet dolly (2200 lbs) out and putting the
identical but tired aural in the visual cubicle, and rigging some
cabling to an "N" tee and re-tuned the tube for visual plus a bit extra
at the high edge of the channel so it could amplify enough sound to say
it was back on the air. At about 1/2 normal power. The governor had to
call a quorum of the unicameral twice over that, once to authorize the
purchase of a new one, and once to write a state check to pay for it
when it was ready to ship nearly 2 months later. And because the ICPM
was now being applied to both visual and aural, it canceled in the home
tv's, so the buzz level dropped from -50 db to around -80, so low I
could not get a valid measurement with the gear of the day.
Yup, old Albert was right. E does in fact equal Mass times Velocity
squared. The atom bomb came from substituting the V2 out for a C2, where
C is the speed of light.
There is more to this story, but I think at this late date, I may be one
of the last 2 or 3 people in the country who could pull one of those
tubes out of its shipping crate, dress it up and put it on the air. One
wrong magnet adjustment can wreck it in 10 milliseconds. But its been 35
years since I last did that, at WNPB-24 near Morgantown, and then I was
working with a used tube they had paid ten G's for, and that was bent in
shipment because it wasn't properly packed. All 4 of the WV ETV
broadcast engineers were busy telling me a bent tube would never work,
but when I got back in the truck that night, there were jaws dragging on
the floor and the power meter said 85%. Not one of those guys actually
understood that bent was correctable by shimming the tube to recenter it
in the magnets. They had no clue when I started measureing for center
with a tape measure, and started cutting up a Masonite scrap I found in
the shop to make shims that the weight of the tube (180 lbs) would hold
in place.
Their problem, typical of those who quit learning when they've collected
the sheepskin, hell I was having high fun, me with an 8th grade diploma,
making all those papered engineers look silly. And as you all know, I
am still learning. ;-) But at 84, its slower and thats a :(
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
Reply to: