Waveguide Propagation
Ed Price
frequencies BELOW it's normally useful range?
I would like to construct a high-pass filter to reject the fundamental
frequency of a transmitter, but allow the passage of the second and
higher harmonics.
I am assuming a coax transmitter output, with the coax connected to a
6dB attenuator and then a coax/waveguide transition. Then a section of
straight waveguide, perhaps 18" long, followed by another waveguide/coax
transition. The coax would then be connected to an attenuator and a
spectrum analyzer. (The purpose of the 6dB attenuator is to limit
fundamental power reflected to the transmitter to a maximum of 10% of
the forward power. At the fundamental, the waveguide should yield a
terrible impedance discontinuity, reflecting most of the forward power.)
If I choose a waveguide which would normally just support the relatively
lossless transmission of the second harmonic, how many dB of loss could
I expect at 1/2 the second harmonic frequency (the fundamental
frequency)?
The variables which I would know are the frequencies involved, the
physical width and height of the waveguide, and the length of the
waveguide.
A second question would be what is the effect of the length of the
waveguide? Do I only need to provide a certain minimum length, or will
loss be strongly proportional to waveguide length? Is it possible that
the coax/waveguide transitions alone will provide enough waveguide
length?
To give you a better perspective, imagine that a 4.5 GHz 100 Watt
transmitter is connected to X-band (WR-90 guide? normally used from 8
GHz to 12.5 GHz) waveguide. What would the transmission loss be at 4.5,
5, 6, 7, 8, and 9GHz?
Thanks in Advance,
Ed
Medical Electronics Lab
>
> Can anyone provide a formula for the transmission loss of a waveguide at
> frequencies BELOW it's normally useful range?
Should be exponential below cut off.
> I am assuming a coax transmitter output, with the coax connected to a
> 6dB attenuator and then a coax/waveguide transition. Then a section of
> straight waveguide, perhaps 18" long, followed by another waveguide/coax
> transition. The coax would then be connected to an attenuator and a
> spectrum analyzer. (The purpose of the 6dB attenuator is to limit
> fundamental power reflected to the transmitter to a maximum of 10% of
> the forward power. At the fundamental, the waveguide should yield a
> terrible impedance discontinuity, reflecting most of the forward power.)
Ought to work ok. But the reflected power in the funamental has
to go somewhere. Put in a circulator and water dump or other RF
absorber.
> If I choose a waveguide which would normally just support the relatively
> lossless transmission of the second harmonic, how many dB of loss could
> I expect at 1/2 the second harmonic frequency (the fundamental
> frequency)?
> The variables which I would know are the frequencies involved, the
> physical width and height of the waveguide, and the length of the
> waveguide.
The pass band equations should be in any wave guide book.
>
> A second question would be what is the effect of the length of the
> waveguide? Do I only need to provide a certain minimum length, or will
> loss be strongly proportional to waveguide length? Is it possible that
> the coax/waveguide transitions alone will provide enough waveguide
> length?
You need at least 1, maybe 2 wavelengths. The decay is proportional
to wavelength. The transitions shouldn't have any losses for different
wave lengths (well, to within a few db, but not like what you're
looking for).
>
> To give you a better perspective, imagine that a 4.5 GHz 100 Watt
> transmitter is connected to X-band (WR-90 guide? normally used from 8
> GHz to 12.5 GHz) waveguide. What would the transmission loss be at 4.5,
> 5, 6, 7, 8, and 9GHz?
The formulas are in lots of books, but it goes as
1/exp(wave_number*distance) below cutoff. Once you're above
cutoff, it'll pass 100%.
Patience, persistence, truth,
Dr. mike
Jonathan E. Hardis
> My radio was on and tuned to a local FM station before it turned
> on. There was noise that went away as soon as ( presumably) the
> transmitter was turned on but nothing was being broadcast (but
> the carrier signal). Why did the radio get silent all of a
> sudden? Does it have anything to do with some kind of automatic
> gain?
>
> Thanks for any ideas!
I would think that it has to do with locking on to the carrier signal
(locking, as in a phase lock loop).
- Jonathan
ale2
Benjamin P. Carter
No, it has to do with the nature of FM as opposed to AM. Automatic gain
control is only needed, and only used, for AM radio reception. Without it,
the loudness of an AM station would be proportional to its signal
strength.
FM is different. The loudness of an FM station depends only on the
frequency deviation. This is required by the FCC not to exceed a certain
value, to avoid interference with nearby stations. A station with a strong
signal is no louder, but it has better fidelity, than a station with a
weak signal. The weaker the signal, the more static you will hear. In
the extreme case of no signal (no carrier), the static is maximum. That is
the response of a discriminator (an FM demodulator) to random noise.
--
Ben Carter
mae...@cix.co.uk
On 1998-09-25 jha...@tcs.wap.org(JonathanE.Hardis) said:
Newsgroups: sci.physics.electromag,sci.physics
> on. There was noise that went away as soon as ( presumably) the
> transmitter was turned on but nothing was being broadcast (but
> the carrier signal). Why did the radio get silent all of a
> sudden? Does it have anything to do with some kind of automatic
> gain?
I would think that it has to do with locking on to the carrier
signal (locking, as in a phase lock loop).
- Jonathan
I don't know whether PLL receivers use AGC; but earlier radios have used
AGC since the mid 20s. This turns down the gain of at least one stage of
amplification (and usually mixing), thereby also reducing the noise
which gets amplified, when a stronger signal arrives.
It is remarkably easy to incorporate. Old tube wirelesses usually have a
single or double diode, often combined with a triode or other valve; and
one of those diodes is there to generate a negative bias voltage which
gets bigger for bigger signals.
That is used to bias a "variable-mu" valve which has a control grid of
uneven spacing. When it is made more negative, the closer-would sections
don't let ANY electrons through so the overall Gm gets less.
============ ===== ===== BILL J. ===== ===== ============
GM8APX, qthr Edinburgh, Scotland, UK
++My old fax machine is Y2K-OK, if I subtract 28 years++
Net-Tamer V 1.12 Beta - Registered
Benjamin P. Carter
>earlier radios have used
>AGC since the mid 20s. This turns down the gain of at least one stage of
>amplification (and usually mixing), thereby also reducing the noise
>which gets amplified, when a stronger signal arrives.
A traditional FM radio with vacuum tubes has the following stages:
optional RF amplifier, mixer, IF amplifier, limiter, discriminator, audio
amplifier. The limiter chops off the peaks of the (sinusoidal) IF signal.
The output of the limiter approximates a square wave of constant
amplitude, regardless of the amplitude of the input IF signal. The
discriminator measures the frequency of the square wave. The output of
the discriminator is a voltage proportional to the frequency deviation
and independent of signal strength (even without AGC).
I don't remember whether old FM radios included AGC. It seems to me that
AGC would have been useful but not essential for a cheap FM radio.
--
Ben Carter
ale2
b...@netcom.com (Benjamin P. Carter) writes:
> AGC would have been useful but not essential for a cheap FM radio.
I should have mentioned that the radio in question is a $80 or
so "boom-box" (only a small boom though), and only a few years
old (no tubes, I've been inside it). I read on this newsgroup
that the noise one hears on a
radio tuned in between stations is from the radio's first
amplification stage and it seems it should still be there when
tuned to a station?
Thanks for the suggestions!
dave pierson
One of the features of FM is the ability to broadcast 'quiet'.
(Details get beyond my ability to explain, nought to do with
cancelaation tho...)
Looked at another way, the 'noise' in an _am_ receiver is
(largely) the result of the early stages amplufying their own,
or ambient AM noise. Since the _FM_ detector ignored amplitude
variations, it ignores the internal noise as well. What would be
noise in an AM system simply never gets to the audio stages.
thanks
dave pierson |the facts, as accurately as i can manage,
Compaq Computer Corporation |the opinions, my own.
334 South St |
Shrewsbury, Mass USA pie...@gone.enet.dec.com
"He has read everything, and, to his credit, written nothing." A J Raffles
"....the net of a million lies...." Vernor Vinge
Duncan O'Neal
>In article <bpcEzw...@netcom.com>
>b...@netcom.com (Benjamin P. Carter) writes:
>
>> AGC would have been useful but not essential for a cheap FM radio.
>
>I should have mentioned that the radio in question is a $80 or
>so "boom-box" (only a small boom though), and only a few years
>old (no tubes, I've been inside it). I read on this newsgroup
>that the noise one hears on a
>radio tuned in between stations is from the radio's first
>amplification stage and it seems it should still be there when
>tuned to a station?
>
>Thanks for the suggestions!
No, it is nothing to do with AGC. In fact FM doesn't need AGC since
the signal goes through limiter/gain stages instead.
If the reciever has no carrier to track, the noise could show up from
an ajacent station, the frequency contol might hunt towards but is
unable to lock on -- 'cause as soon as it steps over the line it
retreats' -- (metaphorically)
+ Submitted by: Duncan O'Neal
+-------------------------------------------------------------+
| To send me a wire: BUT NOT TO ADVERTISE! And Not-to-use-
| -on-a-mailing-list: I scribe -- donealsh...@gov.jp
| Drop the m...@gov.jp...bit And insert @ before S.
| D0T between W&W and another before ca.
+-------------------------------------------------------------+
Paul Alan Cardinale
>
> My radio was on and tuned to a local FM station before it turned
> on. There was noise that went away as soon as ( presumably) the
> transmitter was turned on but nothing was being broadcast (but
> the carrier signal). Why did the radio get silent all of a
> sudden? Does it have anything to do with some kind of automatic
> gain?
>
> Thanks for any ideas!
Here's what's happening:
The voltage coming out of an FM detector is propotional to frequency
of the strongest signal going in.
Before the station comes on, there's nothing but background noise going
into the detector. The strongest frequency at any given point in time
is random; thus the output voltage with no signal is random. This
random
(white) noise is what you heard before the station turned on. When the
station turned on, but before the sound started, it was broadcasting an
unmodulated carrier at the center frequency and the output of the FM
detector was a flat 0 volts (silence).
mae...@cix.co.uk
On 1998-09-28 don...@nospam.gov.jp(DuncanO'Neal) said:
do}Newsgroups: sci.physics.electromag,sci.physics
do}On 26 Sep 1998 20:58:37 GMT, al...@NOSPAMpsu.edu (ale2) wrote:
do}>In article <bpcEzw...@netcom.com>
do}>b...@netcom.com (Benjamin P. Carter) writes:
do}>> AGC would have been useful but not essential for a cheap FM
do}radio. >
do}>I should have mentioned that the radio in question is a $80 or
do}>so "boom-box" (only a small boom though), and only a few years
do}>old (no tubes, I've been inside it). I read on this newsgroup
do}>that the noise one hears on a
do}>radio tuned in between stations is from the radio's first
do}>amplification stage and it seems it should still be there when
do}>tuned to a station?
do}>Thanks for the suggestions!
do}No, it is nothing to do with AGC. In fact FM doesn't need AGC since
do}the signal goes through limiter/gain stages instead.
do}If the reciever has no carrier to track, the noise could show up
do}from an ajacent station, the frequency contol might hunt towards
do}but is unable to lock on -- 'cause as soon as it steps over the
do}line it retreats' -- (metaphorically)
do}+ Submitted by: Duncan O'Neal
do}+-------------------------------------------------------------+
do}| To send me a wire: BUT NOT TO ADVERTISE! And Not-to-use-
do}| -on-a-mailing-list: I scribe -- donealsh...@gov.jp
do}| Drop the m...@gov.jp...bit And insert @ before S.
do}| D0T between W&W and another before ca.
do}+-------------------------------------------------------------+
All high gain systems are inherently noisy. Most FM receivers have
"squelch" - a deliberate means of silencing the audio amplifier except
when something Definitely Useful is on the way through. Even electronic
organs have squelch.
============ ===== ===== BILL J. ===== ===== ============
GM8APX, qthr Edinburgh, Scotland, UK
++Websites which say Get a New Browser don't sell much++
Jos Bergervoet
>>old (no tubes, I've been inside it). I read on this newsgroup
>>that the noise one hears on a
>>radio tuned in between stations is from the radio's first
>>amplification stage and it seems it should still be there when
>>tuned to a station?
>>
>No, it is nothing to do with AGC. In fact FM doesn't need AGC since
>the signal goes through limiter/gain stages instead.
The effect of a limiter is the same as AGC. And this is exactly what
explains the behavior. Weak FM stations do give a noisy sound on
reception, because then there is less limiting/gain-reduction.
This should clearly prove that the dissapearance of the noise is due
to gain reduction! It is true that FM noise is less than with AM,
but still in principle it's there.
Jos
Duncan O'Neal
wrote:
Yes, garbage-in ~ garbage-out.
>
>Jos
+ Submitted by: Duncan O'Neal
+-------------------------------------------------------------+
| To send me a wire: BUT NOT TO ADVERTISE! And Not-to-use-
| -on-a-mailing-list: I scribe -- donealsh...@gov.jp
| Drop the m...@gov.jp...bit And insert @ before S.
| D0T between W&W and another before ca.
+-------------------------------------------------------------+
Benjamin P. Carter
> It is true that FM noise is less than with AM,
>but still in principle it's there.
A proper comparison of AM to FM noise requires some mathematical
sophistication. This is a topic you will find in textbooks written for
electrical engineers. It depends on Fourier analyses of the modulated
signals, to which band-limited Gaussian white noise is added. For AM, 100
% modulation occurs when all the energy is in the sidebands, so that the
carrier disappears. For FM, the analysis is more complicated. 100%
modulation is defined as a certain maximum allowable frequency deviation.
The signal is "wideband FM" if that frequency deviation is large compared
to the highest possible modulating frequency; otherwise the signal is
"narrowband FM". Audio FM stations were wideband in the good old monaural
days. Now that FM is stereophonic, the noise is much worse than it used
to be, but it is still better than the crappy audio signal in broadcast
TV, which is truly narrowband FM. The mathematical theory verifies what
you already know about TV sound quality, which is that narrowband FM is no
better than AM. Only wideband FM yields anything like audio high
fidelity in a noisy environment.
--
Ben Carter