Hi,
The Frequency Division Multiplex (FDM) efforts also resorted in a
technique called the polyphase network. It's a type of network designed
such that the rotation vector rotating in one direction is supported,
but the other direction is suppressed. This is the same as positive
frequencies is supported but negative is suppressed. The principle works
for 2 phases, but you can run on more. This is naturally ideal for
Single Side Band (SSB) operation which is what you typically want to use
for FDM networks as you can pack the voice calls more densely and thus
run for longer cable runs without the bandwidth of the cable being a
problem, which the more wasteful AM would do, as you have double
sidebands then.
Polyphase filters have since become staple-ware filter as it lends
itself to nice sub-band decimation processing in digital processing,
providing suitable for sub-band filter banks such as those being used
for MPEG audio compression for instance.
I rarely see the roots being referenced, but I managed to get hold of
Mike Gingell who's PhD is really the root of this. He was friendly
enough to send me not only the PhD but a few papers and patent which I
host here:
http://rubidium.dyndns.org/~magnus/synths/friends/gingell/
The papers is a gentle way into the way of thinking.
For radio-amateurs, this network may ring a bell:
http://hanssummers.com/images/stories/polyphase/circuit/circuitnetwork.gif
Hans Summers have used it in conjunction with a Tayloe detector for a 80
m receiver:
http://hanssummers.com/images/stories/polyphase/circuit/circuittayloe.gif
Look at his complete article here:
http://hanssummers.com/polyphase.html
Cheers,
Magnus
On 02/04/2016 12:13 AM, Bob Camp wrote:
Hi
Back in the early days of multiplex transmission, frequency division multiplex
was king. There were a
lot of systems. They used a variety of schemes to pack the baseband signals
into sub 2 MHz sort of
arrangements. For quite a while the state of the art was to use various low
frequency crystals to generate
the multitude of careers that spread out from audio up to maybe a MHz or two.
As a result of this, there were a *LOT* of crystals made at weird low
frequencies. Some of these systems lived
on well after the advent of digital multiplex systems. Once you have 400 miles
of system A it’s easier to extend
it a “few more miles” than to re-do the whole thing. Spares for these systems
were made at least into the 1990’s.
While these are really cool looking parts (try a strobe on your 1 KHz bar …)
they rarely are very stable. They are
more interesting as art than as frequency standards.
Bob
On Feb 3, 2016, at 2:36 PM, Dave Brown <[email protected]> wrote:
And many of these glass envelope crystals were made on very low frequencies- I
have several below 20 kHz and one as low as 3 kHz in a B7G based envelope that
is 4 inches long.
http://homepages.ihug.co.nz/~tractorb/3k1.JPG
DaveB, NZ
----- Original Message ----- From: "Alan Melia" <[email protected]>
To: "Discussion of precise time and frequency measurement" <[email protected]>
Sent: Thursday, February 04, 2016 6:30 AM
Subject: Re: [time-nuts] Glass Envelope Quartz Crystals
I think it was unlikely that that it was made "just to see where it would
come out" That is a flexural bar possibly an NT cut. 100KHz standards were
commonly made in this format.
The British GPO had a factory at Mill Hill in N. London making these in
tube-like (valve in UK) enclosures, IO GT and B7G. I have a number of unit
saved from the dumpster (skip in the UK).when the unit closed in the early
60s. Remember there was a surge in telephones in this era and many of these
frequencies were for FDM carriers on trunk sytems. This is pre-synthersiser.
Also many special quality tubes were made for VHF in B7G with two or three inch wires
instead of pins to reduce the socket parasitics. so these were probably still around in
Russian factories to produce components for the "Foxbat" etc.
Alan
G3NYK
----- Original Message ----- From: "Philip Gladstone"
<[email protected]>
To: <[email protected]>
Sent: Wednesday, February 03, 2016 1:21 PM
Subject: Re: [time-nuts] Glass Envelope Quartz Crystals
I dug around in my junk box, and found this:
https://plus.google.com/+PhilipGladstone/posts/JBNLMSq2rsE?pid=6247050011623528018&oid=115465617973526125523
This is (according to the markings) a 71.137 kHz crystal made in 1948. I
suspect that they just measured the crystal after manufacture rather than
actually trying to make a 71137Hz crystal....
After this discussion, I'm feeling the need to fire it up and see whether
it still runs, and what the aging has done to the frequency....
Philip
On 03/02/2016 07:11, Bob Camp wrote:
Hi
There is actually a range of crystals shown in the pictures. The gold
plated 5 MHz
crystal is probably an overtone part. It could be fairly precise. The 25
MHz part is
plated with something like silver. It probably is a *much* lower
precision part. There
likely are long stories that explain just why this or that package got
used in this
or that application.
Bob
On Feb 2, 2016, at 10:04 PM, Jeremy Nichols <[email protected]> wrote:
The OP's picture looks very much like the crystals that HP's "Frequency
and Time" division in Santa Clara (02 was their division number) used to
manufacture back in the 1970s. My picture shows a 1 MHz crystal that HP
used in the predecessor to the HP-105A (perhaps the 101A).
Jeremy
http://s323.photobucket.com/user/Jeremy5848/media/Miscellaneous/Crystal_1140587_zps0jxjpoal.jpg.html
On 2/2/2016 12:24 PM, Don Latham wrote:
You have it right, iovane. At the least, they should be protected from
light,
thermal radiation, and emf. Won'drous things will happen if the
crystal and
its structure are subjected to radiation through the glass. I'd suggest
a foam
wrap in a tin can as a minimum. Put the oscillator cat in there too.
Don
iovane--- via time-nuts
I think that these crystals were designed to be placed in an oven,
which
worked
as a shield too. I have a similar crystal made by Racal in the 60's,
and in my
case it is fitted with the classic octal tube-type plug. It was housed
(still
is) in a heavy massive shimmering chrome-plated cylindrical brass
enclosure, a
beauty to see, It was the timebase of a tube-type synthesizer with
lots of
tubes. Themperature control was achieved by means of a mercury
thermometer in
which mercury actuated a contact when reaching a wire crossing the
capillary
tube.
Antonio I8IOV
Da: Bob Camp <[email protected]>
Data: 02/02/2016 13.15
A: "Discussion of precise time and frequency
measurement"<[email protected]>
Ogg: Re: [time-nuts] Glass Envelope Quartz Crystals
Hi
Since the 25 MHz crystal has already been soldered into a circuit,
putting it
in a
socket is probably not a real good idea. It’s also a leaded part.
Even with
fat pins
sockets can be an issue. With wire leads, you are asking for trouble.
Functionally, there is little there is little difference between a
glass
package crystal
and a metal package. About the only real one is the obvious - one has
a metal
shield
you can (but sometimes don’t) ground.
Bob
On Feb 1, 2016, at 9:58 PM, Daniel Watson <[email protected]>
wrote:
Hi,
I purchased a pair of interesting glass envelope crystals for a
project.
Here are some pictures:
http://syncchannel.blogspot.com/2016/02/glass-envelope-quartz-crystals.html
Does anyone have an idea about what mount/socket I should buy for
these? I
read a previous thread on the list about Bliley crystals using a B7G
mount,
but I'm not sure if that type might work here.
Also, when building up a circuit to make these oscillate, are there
any
specific differences about crystals in this package that I should
keep in
mind?
Thanks much,
Dan W.
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
