RadioBanter - View Single Post

September 29th 04, 04:04 AM

In article ,
(Brian Kelly) writes:

(Len Over 21) wrote in message
...
In article ,

(Brian Kelly) writes:

As you will say later, those "analog" radios have INFINITE resolution.
:-)

Creative PLL and DDS subsystems of today, designed by others,
make it possible for anyone to select 10 Hz increments on any
HF band (30,000 frequencies within 300 KHz) with crystal-
controlled accuracy.

Analog VFOs are continuously variable. Making it possible for anyone
to select an *infinite* number of "increments" within a 300Hz
bandwidth much less your coarse 300 Khz wide example.

Heh heh heh...your bafflegab won't win blind man's bluff, Kellie,
deal yourself a better hand... :-)

Feel free to try to state you can return to that "infinite possible"
setting within a few PPM...all without any old crystal calibrator and
dependent on that "coarse" analog dial. :-)

Whatta lame whack at a twist. I didn't claim any such nonsense did I?

No.

What you *did* claim was that you could operate within 200 Hz of a band edge
and know you were inside, using certain '50s/60s vintage equipment. Which is e
a reasonable claim for the equipment involved.

btw, back in 1975 or so I designed and built a "digital dial". The way it
worked was that it used TTL 74192 presettable up-down counters to count the
tunable oscillator. You'd adjust dip switches for the offset and direction of
each band and mode. Its time base was a 400 kHz xtal, easily zeroed to within a
few Hz of WWV. The thing normally read out in 100 Hz increments but could be
switched to 10 Hz or 1 Hz. Its accuracy was dependent on how well you set the
time base and presets. Could be used with almost any rig. Hooked it up to a
75S3 and got an A in the course. And yes, you could easily reset it to 100 Hz.
10 Hz took a steady hand.

Later I saw a better design. It sampled and counted all the oscillators in a
rig, and displayed the total. No presets to adjust - set the timebase to WWV
and you're done. Could go to 1 Hz if you were willing to have it update once
per second.

Heh heh heh...your bafflegab won't win blind man's bluff Sweetums,
deal yourself a better hand... :-)

Just watch...

And they do it
without generating any phase noise or other forms of crud synthesizers
toss out.

Kellie, define "phase noise" insofar as amateur radio operation is
concerned.

No sweat Sweetums. If I terminate the rcvr input with a 50 ohm dummy
load via a short length of coax and am able to hear any gurgles,
chirps, squeaks, pings, skips or burps when I swish around some freq
or another it's synthesizer crud, i.e., "phase noise" in play. However
with current-tech ham gear internally-generated crud is not often a
big problem these days since it's usually below the atmospheric and/or
electrical QRN noise floor on the band under consideration. Which is
easy enough to check. Welome to the realities of "phase noise "insofar
as amateur radio operation is concerned" Sweetums.

Sort of.

You, for the limits of your technical knowledge, should
call that "incidental FM" which is what the industry term "phase
noise" refers. :-)

Maybe when ham radio ceases to be a hobby and becomes an "industry"
Sweetums.

Then you should examine exactly how low that terrible phase
noise
is. You can use the term "dbc"

Amos: "Oh crap, here he goes again."
Andy: "Nudge me when he runs outta wind willya?" Zzzzz . .

referring to the number of decibels
below the "carrier" (center frequency reference, not a modulated
carrier per se). The "crud" (as you term it) is quite far down in
relative power and certainly won't affect morse code reception of an
on-off keyed station's carrier.

Wrong. Incorrect. Not true at all in the real world of HF radio.

Len has just demonstrated, once more, that he just doesn't get it.

"Phase noise" is a somewhat new buzzword in industry due to the
importance of keeping it low for QAM signals (Quadrature [phase]
Amplitude Modulation, a combination of PM and AM). The cell
phone engineers will know of that importance on keeping the BER
(Bit Error Rate) as low as possible. The amount of work in the last
decade on cellular telephony techniques has been enormous
worldwide. It's only natural that industry advertisements, from sub-
system components to full systems, emphasize a low "phase noise."

Has nothing to do with the subject at hand, which is HF amateur radio.

As far as on-off keyed radiotelegraphy, your mention of "phase noise"
as being "crud" in synthesizer frequency control is akin to making
a big case for gold-plated music system speaker wires. :-)

Wrong again, Len.

It is ignorance to discount the possibility of "crud" being non-existant
in analog mixing frequency generators. Those analog "infinitely-
variable" oscillators are just as prone as anything to "phase noise."
The wrong selection of mixing frequencies will produce spurious
responses...one of the papers I wrote at RCA was on quick
identification of such possible spurs (not the first, but it was a
very quick way to determine them).

Misses the point completely.

(Long pause to let the fog clear)

(Amos nudges Andy) "I thnk it's over, he melted down in his own hot
air bafflegab again, wake up."
Andy: "Are you sure? I can use more Zs."

Here's what *really* happens:

In an ideal superheterodyne, all the oscillators would generate pure, steady
injection signals. In reality, there is always some imperfections in those
oscillator signals. In modern frequency synthesizers, particularly PLL types,
the imperfection takes the form of noise sidebands on the oscillator signal.

Now even first-generation designs had noise sidebands many dB below the desired
LO signal. Someone who doesn't really understand the situation might react as
Len does, saying that such low-level noise can't have any real effect on
receiving the desired signal.

Trouble is, in the amateur HF environment we often want to listen to a weak
signal surrounded by many strong ones, often only a kHz or two away. Good
crystal and mechanical filters make it possible to separate such signals *if*
they can get to the filter in decent shape.

What happens when the LO signal is phase-noisy is that a close-in-frequency
unwanted signal mixes with the LO *noise*, and produces noise in the receiver
output. With a whole bunch of strong signals, the noise can be so high that it
drowns out the wanted signal. This problem is not due to IMD, blocking or other
various nonlinearities in the front end - it's due to phase noise alone. And in
modern ham xcvrs where the signal is first converted up to about 70 MHz and
then converted down to about 8.8 MHz by means of synthesized LOs, the problem
can be severe.

Scenario: Ham in Texas is trying to work Europeans in CQWW on 40 meters. He
points his 3 elements at 90 feet towards EU. Which also means towards a lot of
the Northeast. There's lots of points to be had working the low-power
limited-antenna hams there. But to do it, he has to be able to pick their
less-than-a-microvolt signals out from between the forest of locals and East
Coast legal-limit folks. And the band is busy - a signal every few hundred Hz
from 7000 to 7070 or more. Plus the usual cast of megawatt SWBC above 7100 and
below 7000, blasting away with big antennas aimed right at him. 100,000
microvolt and bigger signals aren't rare in such scenarios.

If his receiver's LO is noisy, he'll hear all that off-frequency RF as noise.
And he won't hear the low-power limited-antenna less-than-a-microvolt stations
he's trying to work. All the Inrads and DSP in the catalogs won't do any good
in such a situation.

That's why phase noise is important to hams.

My FT-847, which is not much as ham xcvrs go, can be tuned in 1 Hz
increments vs. the "make it possible for anyone to select 10 Hz
increments" thingey you cite above.

10 Hz increments is common in installed equipment (including the
ham consumer market) in the past two decades. I know there are
smaller increments...:-)...but I also have to play to the common
denominator of technical expertise in here.

1 Hz is common in modern manufactured amateur equipment. But that's not really
the issue.

10 Hz increments are perfectly fine for SSB voice tuning, as I've
found out with my Icom R-70.

Heh. You can't tune that pore 'ole 3-star boat anchor in 10 Hz
increments Sweetums, the best you can do with the thing is tune it to
the nearest 100 Hz increment yes? Of course you silly old thing. I've
never seen an R-70 in the flesh so tell me, are those actually Nixies
in the display for God's sake?!

R-70 is a pretty good receiver. Almost qualifies as a boatanchor now....

If that old R-70 is your "window" to ham radio I think I'm starting to
understand why you have a dour view of the hobby. You need to get past
the R-70 and try a JRC NRD 545 Sweetums, like the one I have. It'll
change your life.

The bald fact of the mattter is that once more a PCTA caught you
bafflegabbing again Sweetums, wasn't even a decent try so once more no
cigar for you.

When I bought my R-70 (years ago), the three extras at work in the
Van Nuys, CA, store . . .

Amos: "Oh crap, here he goes again."
Andy: "Nudge me when he runs outta wind willya?" Zzzzz . .

. . . didn't know squat how Icom was able to do it
with 10 KHz reference frequencies to the PFDs (factor of 1000:1)
there.

So what?

Turns out Icom has a neat 3-loop PLL arrangment, doesn't
go into DDS or Fractional-N at all. Minimal phase noise and no
discernable "crud" anywhere within full tuning range.

How many points did Len get with it in the last CQWW? Or even the last SS or
Field Day?

Okay, so your spiffy-schmiffy 1 Hz resolution "xcver" is "guarnateed"
accurate because it has a "digital dial?" I don't think so.

Nobody claimed that it was accurate. It *is* precise, however. Big difference.

Exact 1 Hz
settings imply 100 PPB (Parts Per Billion) accuracy of the master
reference oscillator.

Let's see - 1 part per million is 10 Hz at 10 MHz. Or 1000 parts per billion.
So 100 parts per billion is 1 Hz at 10 MHz.

You will NOT be able to hold such accuracy
and be believable to anyone who has worked to such accuracies in
crystal oscillators. Certainly not for the ham consumer market.

Nobody is claiming that kind of *accuracy*. Only that kind of *precision*.

Fella named John R. Vig (unusual surname) is a good name to
remember on what can be done and can't be done with crystal
oscillators. Big name in the frequency control part of electronics
industry, probably not in the pages of QST. :-)

(Amos nudges Andy) "I thnk it's over, he melted down in his own hot
air bafflegab again, wake up."
Andy: "Are you sure? I can use more Zs."

Yup.

Of course there's an easy and quick check of all this. Just tune in WWV and see
what the fancy digidial says when you zero beat the carrier in SSB mode. That
will tell you how accurate the reference oscillator is. Traceable directly to
NIST via the F2 layer. If you're at all careful you can get to the point where
the S meter needle is slowly fluctuating as the frequency/phase difference
wanders...

btw, some years back I was there, at NIST in Boulder. Saw the various standards
and how they keep WWV synchronized. Also visited the WWV/WWVB transmitter site.
Got lots of pictures, too.

You obviously need to spend
considerable time leafing thru the ham catalogs to get up to speed on
the equipment we use before you spout off and continue to goose up
your "coefficient of ignornace" on the subject of ham radio in general
and the equipment we use. Again. Gets boring.

True. I never bothered to memorize advertisements in QST by
heart...like so many PCTA extras do. :-)

Like who? Exactly.

I rather prefer what I've been exposed to since 1963 on frequency
control methods...

Still living in the past...

Amos: "Oh crap, here he goes again."
Andy: "Nudge me when he runs outta wind willya?" Zzzzz . .

beginning with those "cruddy" synthesizers
(without "real" frequencies, only the "synthetic" variety)...and
quartz crystal oscillator accuracy and stability to the 10 PPB
region.

Common ham radio quartz crystals have guaranteed accuracies
to 50 PPM typical.

Typically expressed as .005%. Of course that's the accuracy of the xtal itself,
before being zeroed to a better reference. WW2 FT-243s were typically good for
..005%, we do a lot better now.

That translates to 500 Hz at 10 MHz, by way
of example. 1 Hz accuracy at 10 MHz is 100 PPB, or 500 times
closer.

yadda, yadda, more of the usual . . .

WWV. Trimmer capacitor. Zero beat. What a concept.

Then there are the few "drudges" (like myself) who've
gotten our hands dirty doing the design and testing of synthesizers.

Then there are drudges like me who have ham licenses and and put
technoligies to work on the airwaves whilst all you're allowed to do
is bafflegab about 'em with your keyboard.

I'm sorry that my technical competence seems like "bafflegab"
to you. Some further learning of the radio technical arts would
erase some of your ignorance and lend credence to what I've
said. Like, I could ask you "how's the zeta of your control loop"
and you would be out to lunch, cussing and hollering "bafflegab!"

No Sweetums, not at all, that's not the way I work. You're being silly
again. If by any chance I ran into an arcane topic like that in which
I had any interest whatsoever I'd ask an EE to uncurl it for me.

Engineering 101: Don't reinvent the wheel. They even taught us EEs that one.

Miccolis is across town. Then comes the non-ham PhD EE Dean at one the
universities in this neck of the woods I know well. Or my buddy
another N3/EE who goes back to our high high school days together and
ran GE's gummint relations operations in Valley Forge, etc. etc.

And that's just the first string...

- - - - -

Amos: "Oh crap, here he goes again."
Andy: "Nudge me when he runs outta wind willya?" Zzzzz . .

"Zeta" is the symbol for the response characteristic in a closed
loop of a PLL, Fractional-N, or hybrid PLL-DDS system.

Wunnerful ducky wunnerful:

Time for a radio story...

Back in high school I knew a local ham down Collingdale way who was always
working on a pet project. Same age as me, saw him in school every day. Had all
kinds of grand ideas of how he was going to build the next generation
state-of-the-art ham rig. All solid-state, full features, all bands, all modes,
etc.

Now this kid was no dummy and his ideas were basically very sound. But he
didn't have anywhere near the resources or practical experience to actually
finish anything. He'd draw all kinds of schematics, spin all kinds of yarns and
sometimes even gather some parts. But build a working rig? Never happened. Not
once. When he *did* get on the air, it was with borrowed equipment that he
conned some local ham into lending him "temporarily". Until said local ham had
to come over and take it back. I made the mistake of loaning the kid a QST,
which I never saw again. I learned fast.

Meanwhile, those of us willing to make do with less than "SOTA" were on the air
and having fun and QSOs while he pontificated.

That was about 35 years ago but the lesson is still valid: All this bafflegab
doesn't make one QSO.

For some reason I was reminded of him. He sounded just like Len...

Now take a break from your bafflegabbery Sweetums and let's play in my
field of professional expertise this time. Demonstrate your level of
technical competence by solving a very real-world electronics design
problem. Assume that you have a one inch diameter x 1/16 inch wall x
eight foot long 6061T651 aluminum tube fully restrained at one end
with the other and dangling horizontally in the wind. Calculate the
maximum wind speed which will not produce permanent deformation of the
tube.

That's easy! Would take me about sixty seconds to get the answer.

An
important factor for lock-in and stability and anyone designing
the loop filter for a synthesizer should recognize that common
term.

Of course. Who designed the R-70? I bet it wasn't Len Anderson...

I've never dined in the executive dining room (the counterpart to
your "captain's table" BS) in any electronic corporation

Hee! No surprise at all there Sweetums, there are obvious reasons . .
. ah, never mind!

but I
HAVE designed and made frequency synthesizers. Hands-on
work all the way, from the initial paper work-up to long hours
in the environmental lab...to accuracies in 100 PPB over
full military environment. Interesting, challenging work!

So solving the tube-bending problem is a piece of cake for a
duz-it-all "engineering genius" like you eh Sweetums?

USING modern equipment is NOT involving development or
anything else. Try not to run off at the mouth/keyboard so
hastily. Try not to nit-pick like nits over minor phrases in
postings so that you have an "excuse" to cuss and snarl at
NCTAs. It makes you look like nursie's cousin. :-)

"Try not to nit-pick . . . ?!" WTF . . ? Bwaaahahaha - from the master
of all RRAP nit-pickers!!

The main point is simple: Hams did not need synthesizers to stay in their bands
and subbands. Nor do they need 1 Hz or even 10 Hz accuracy on HF.

73 de Jim, N2EY