Re: red SATA cable corruption
- Date: Wed, 12 Sep 2018 02:08:01 -0400
- From: Gene Heskett <gheskett@xxxxxxxxxxx>
- Subject: Re: red SATA cable corruption
On Tuesday 11 September 2018 23:01:43 Felix Miata wrote:
> Gene Heskett composed on 2018-09-11 22:29 (UTC-0400):
> > On Tuesday 11 September 2018 21:34:27 Felix Miata wrote:
> >> IMO, red because it's red is less likely than red because red was
> >> the
> *Less* likely, not unlikely. IOW, both happen, but red dye cancer less
> often than SATA spec inadequacy.
> > That red dye, if in contact with the wire, eats it up like some sort
> > of acid, you can literally pour the copper out of the ends where
> > you've cut it in two, onto a clean sheet of paper, as a dark brown
> > powder that looks a lot like rouge.
> I'm quite familiar with the above language. Google has found me many
> copies of it, authored by you, dating back at least 10 years.
> > so don't tell
> > me about red being just a sata-1 slow cable.
> I did not. Just the same, I doubt any properly constructed SATA-II or
> -III cable with a red outer jacket would allow its red dye to come in
> contact with the conductors (unlike SATA-I):
> > You can buy them today,
> I know, and pink, blue, green, yellow, black, tan, etc. But that
> particular magenta shade, I'm doubtful you'll find without trying
> *really* hard. If you do, it's likely old stock.
> I did buy red relatively recently, long after my first exposure to the
> problem, but in spite of "them" (technically adapters, not cables)
> being red, because of the difficulty finding any that are not red:
I reacted a little strongly, Felix, but this dye vs cable failure is
something I first ran into 48 years ago when I became the afternoon
bench tech at Norfolk Two Way Radio, run by Vi Bernstraugh in Norfolk
NE. 1/4 of the repairs I was doing was replacing the mic cords in the
various CB radios of the day, mostly made in Japan. Some I replaced
several times over the years I spent working for Vi.
Going off topic...
Mornings and evenings, I was the resident engineer in charge of KXNE-TV,
about 16 miles NE of Norfolk, a uhf channel 19 facility. The care and
feeding of the $125,000 klystrons used as amplifiers back in the day is
a whole new, now old field of science because the beam voltage of 20
kilovolts not only generates soft x-rays, the electron beams velocity
was a good fraction of C speed, making the effects of relativity in
terms of mass * speed, one of that technologies major sources of
distortion because it resulted in the tube being electrically longer at
high power levels like during the NTSC synch pulses.
The real problem is related to the fact that its gain is a result of
modulating the speed of the electrons as they flow by, resulting in a
bunching because the faster ones caught up with the slow ones at the
other end of the tube. You could put 1 watt of drive in the top, and
get 30kw out the bottom. But the 500+ megahertz drive has a symmetrical
waveform. But relativity says the mass gain is a square law, so the beam
speed is slowed more than the opposite push can speed it up. So the
delay from top to bottom increases as the instant power goes up.
The fcc calls this incidental carrier phase modulation, and back in the
day it caused such a synch buzz in the audio that the fcc gave us
permission to do our audio tests with an unmodulated video carrier.
I mention this because a water pump failed and burned a hole in the
visual tubes collector, filling it with water. We had no spare tube, so
I hooked up a "N" tee, and fed the video drive to the aural tube,
retuning it for video bandwidth, and we ran that way for several weeks
while Eimac/Varian made us a new tube, and the NE legislature had to be
called up to get a quorum to pay for it. During this time, the audio
buzz was gone because both video and audio was being subjected to this
relativity effect in equal amounts, cancelling it in the receivers of
the day because they all operated on the difference. The unwanted
modulation was still there, but there was no difference the receivers
This "tube" is about 5 feet long, and weighs 180 lbs. and operates in a
focusing magnet that weighs 2200 lbs. And it takes 70 gallons a minute
to cool the collector bucket at the bottom of it. It has now been
replaced with a tube & technology thats more efficient. Klystrons at
their best were only 28-30%, which meant to get 30 kw peak out of it,
you were putting something north of 120 kw into it in DC power for the
electron beam. With two of them in each transmitter, using 220 kwh per
hour we were Wayne County Public Powers biggest customer by an order of
magnitude. Its possible I may be the last on this ball of rock and
water, that knows how to pull it out of its shipping crate, mount the
cavities to it, set it in the dolly and hook up the cooling plumbing,
and adjust it for full power daily use. There are adjustments that can
destroy it in 10 milliseconds in the hands of an idiot.
That water pump then was equipt with some vacuum relays attached to its
power feed, and if any phase was killed, so was the beam power by brute
force. Interrupting it at the 20kv to the tubes point. A $150 breaker
cost us around $130k USD. Expensive lesson indeed.
This of course predates debian by what, 30 years? So its definitely off
topic. OTOH, those that have stayed with me have learned that
relativistic effects CAN be observed by you and I if you know what you
And it seems they've put a white sleeve over the conductors of late,
which may slow the migration of whatever is in that particular red dye
to the extent it *might* last 10 years.
As another furinstance, wire wrapping wire comes in many colors, but its
the red stuff that breaks first, not by coming unsoldered when used for
point to point in one-off pcb projects, but 1/16" of an inch up into the
unstripped insulation. Putz around with the other colors, and they'll
break at the surface of the solder.
Yeah, I've some experience. And it goes back quite a ways timewise.
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>