|
Guys (and gals),
Please go back and review the complete thread.
The OP is using a "made for application" sound card that Scott
Zimmerman designed ).? It's driven by digital data applications (in
this instant case - VARA FM) that doesn't create audio (bandwidth
or level) that results in a over occupied bandwidth situation.?
Therefore, what Scott Z was/is recommending to the OP is
technically correct for the application.? VARA FM runs comfortably
at 2.25kHz deviation in a +/- 5kHz system.? There is no fear of
wrecking anything here - even on 15kHz channel spacing.
If what you thought was truly the case - I/he/we would agree with
you, and be recommending the RB
Version C was recently released, and includes optimizations to
eliminate any possibility of post filter clipping and even further
reduces clock leakage.?
The big advantage of this module is it allows YOU to set the
low-pass filters cut-off frequency and audio levels to the desired
/ allowed / legal occupied bandwidth - all with easy to set
potentiometers.? This (or something like it) is what people should
be using when taking discriminator / detector audio and putting it
to the naked modulator.
Comments and questions welcomed.
Kevin W3KKC
On 11/2/2022 12:51 PM, Burt K6OQK wrote:
toggle quoted message
Show quoted text
I have to agree with Chris on this.? Running raw discriminator
audio directly into the modulator is going to cause problems.? You
need "something" ahead of the modulator to control the maximum
deviation.? This is one of the reasons I disagree with the idea of
coming out of the discriminator, going directly into the modulator
and setting the system for 5 kHz in and 5 kHz out.
Years ago when I was on the TASMA technical committee one of the
most frustrating problems I regularly faced was with people who
would do this.? In a lot of cases the squelch tail was measured
going out to 20 or more kHz of deviation, while popcorn and noise
from signals that were only partially quieting went out to almost
the same.? They don't understand.? It's just bad practice.
Burt, K6OQK
On Tue, Nov 1, 2022 at 10:29 AM, Chris Boone wrote:
I strongly disagree Scott...
That's a recipe for disaster. Feeding wide band unsquelched
noise to a modulator will cause the transmitter to over
deviate. Remember the squelch noise out of a discriminator is
half the bandwidth of the IF filter which means you have audio
at 7.5 khz + coming out of the receiver.. Anytime I run flat
audio, I always put a clipper/limiter circuit in line to
prevent over deviation, and also to limit the audio bandwidth
to 3 kilohertz maximum. Under FCC part 97 rules we are
required to make sure that we use good engineering practices.
Running raw discriminator audio straight into a transmitter
modulator without any processing violates that rule. I've done
that in microwave systems but that's a special case and you
have the bandwidth to do it. In the crowded VHF/UHF bands,
it's totally ludicrous to do so
?
Chris
WB5ITT
|
Occupied bandwidth is an interesting topic.? For many of us - using an unmodified commercial transmitter - it's an 'academic' topic since the manufacturer has delt with it.
?
That said, amateur radio is about experimentation.? Whether for voice or data applications, if the user is implementing the transmit audio processing, bandwidth is a serious consideration.
?
Knowledge and understanding of bandwidth measurement procedures and results could be a weak area for many hams.
?
In his discussion about Quantar transmit audio, Burt said that he regularly monitors his occupied bandwidth.? I asked him to share his knowledge and expertise by explaining how he does occupied bandwidth measurement.
?
Hopefully this will promote an interesting an instructional discussion.
|
"... weak area for many hams." - amen.? This is at the forefront of many evolving discussions.
I'm no where near the level of expertise and experience of most/all here, but in the band-planning and repeater coordination world I/we need to absorb, understand and reference a lot of this as some/many in the ham community want accommodations for or to "sneak in" some generic piece of ham gear to promote/expand/implement more narrow-band digital gear and modes amid the analog FM world.?
For a stand-alone 'experiment' whatever desired or incidental results of occupied bandwidth may not matter much.? It's when we 'present' our work amid the shared/coordinated spectrum with others (just have to have that new repeater...) that all the specs and standards become essential.
If the ARRL or anyone else has ever addressed this at some point, it needs to be re-visited, and if not, authored and put out there.
Thank you!!
?
|
I agree with Aspinwall's thought.?
Just because someone got a license by memorizing some Q n A doesn't absolve any responsibility for damage.?
You don't get plausible deniability just because you don't know what you're doing.
ARRL and FCC both have clauses that refer to "good amateur practice" and in my mind, this speaks to being
able to test and verify the signal quality of one's emissions.? Site owners also have been instituting tightened
standards for communications equipment.? When these ideas are disregarded, then rules get tightened because
experimenters are making the same bad mistakes over and over again.? I just heard of a site that just did an
inspection and found a guy sneaked in some of his ham friends and now they all owe a whole lot of money for?
back rent and got tossed out the door.? Since this tower owner owns a LOT of sites, look out people.
Note that even if specs and standards are met, there may be problems that still must be corrected.
Experimenting is OK, but when you take it off the dummy load, you are responsible for emissions.? When you put
something on a mountain top,? you might create a lot of damage and hatred for hams before being found out.?
Plausible deniability won't fix it.
The other thing I notice, is that some people come here for advice before they even learn how to test and
troubleshoot for themselves.? That can be a long process of training and experience you can't get from Dear Abby.?
I welcome someone asking the question, because we can't know it all.? Asking for help is a good thing.? We should
try to discern when someone is in over their head and suggest they step back and study up on some things before
trying to save the day with a quick quip.
On Sun, Nov 6, 2022 at 08:20 AM, Jim Aspinwall wrote:
"... weak area for many hams." - amen.? This is at the forefront of many evolving discussions.
I'm no where near the level of expertise and experience of most/all here, but in the band-planning and repeater coordination world I/we need to absorb, understand and reference a lot of this as some/many in the ham community want accommodations for or to "sneak in" some generic piece of ham gear to promote/expand/implement more narrow-band digital gear and modes amid the analog FM world.?
For a stand-alone 'experiment' whatever desired or incidental results of occupied bandwidth may not matter much.? It's when we 'present' our work amid the shared/coordinated spectrum with others (just have to have that new repeater...) that all the specs and standards become essential.
If the ARRL or anyone else has ever addressed this at some point, it needs to be re-visited, and if not, authored and put out there.
Thank you!!
?
|
I dare say we've been facing a new 'dimension', in the realm of hotspots... many new entrants get (an economical) taste of having their own personal portal to "work the world" with a 5 watt talkie and 10mW - 1 watt (?) IP interface (Skype via radio?)? "This is great, now if only I could get more range..."? ?If we thought $20 import talkies were ugly. take a $5 key fob 'radio', add some LMR-400 and a dual-band j-pole on the roof... plus or minus the variable of where they are programmed.? 'better' on the roof of a multi-story apartment building, office... snuck in the cabinet of a radio site.? Many of course have no clue, no Elmer (or not a good one), no test equipment.
Some of us pushed for the piStar s/w engineers to at least restrict some of the frequency programming to respect satellite, weak signal, etc. but 430-440 is the UHF spectrum for much of the non-US world.? That still leaves ill-defined spectrum there, and a lack of understanding/awareness of band plans, coordination in 440-450.? Who hasn't had a hotspot show up on a repeater input?? "but I'm digital, we don't get interference..." - so can't cause any either?? ?I'm due to help a US-based AllStar hotspot vendor document cautions and mitigation methods so their customers don't screw up too badly.
Surely it was "interesting times" going from AM to SSB, AM to FM, wide to narrow, certainly narrow to narrower. Part 97 is weak on (absent current) tech specs.? Measuring occupied bandwidth is a stretch if not 'impossible' for some of us, but hopefully we can do something to mitigate on-air disaster.? We're the senior experts, well, most of y'all are, some of us are expected to be and we do reach out.? It's good that a lot of us are willing, able and are helping.? Someone helped us, and if not us, who to help new folks not mess things up for us now, and the what the future holds.??
Appreciate y'all for my meager ability to realize how not to mess up the airwaves too much, or for long.
|
Are we becoming TOO TRUSTING??? I remember all these CB shops that would charge people money to build sound effect and noise
makers, and rip the modulation limiter out of the radio and tweak the drive and final stages to make 10 Watts (1000 bird watts) by moving
the stage from class B to class Crackhead.? More "Bird watts" splatter and harmonics and make the customer sound like an AH so
everyone on the band would hunt them down for a Molotov cocktail party or a drive-by shooting.?
I hear the cartels are full encrypted linked trunked systems now, so maybe we have been bypassed.
Oh well, back to building my gigawatt spark gap transmitter.? Cant wait. Kirrrrrch Kik Kirrrrrrch Kik????? Kirrrrch Kirrrrch Kik Kirrrrrrrrrrrrch!!!!
True democracy is allowing every mentally impaired person the right to self-actualize, destroy everything, kill their neighbors and eat them.
|
I also share the concerns about hotspots, new users, and those who want to build a repeater when all they have for test equipment is free DMM from Harbor Freight.
?
However, a few may discover a passion for RF and want to 'learn the ropes' - hopefully, with real study, not by watching a 2-minute youtube video.??
?
For those folks and a lot of other amateurs and professionals, bandwidth is a topic they haven't had much experience with.? Mostly we use commercial gear and trust that the manufacturer took care of it.
?
The repeater builder site and this discussion list are the pre-eminent resource to shine a light into these dark corners.
?
In that interest, let's talk about bandwidth in the context of channel spacing, bandplans, and frequency coordination.
?
First - occupied bandwidth [OBW]
?
Occupied bandwidth is a regulatory requirement that is specified for equipment in some regions, such as the USA. [47CFR2.202(a) & 47CFR2.1049]
It is defined by the ITU-R as a maximum bandwidth, outside of which emissions do not exceed a certain percentage of the total emissions such as 1% - in other words, the OBW contains 99% of the total power.?
Thus, the 99% occupied bandwidth is the frequency bandwidth such that, below its lower and above its upper frequency limits, the mean powers radiated are each equal to 0.5 percent of the total mean power radiated by a given emission.
?
There is also the term Authorized BW which is the maximum bandwidth that the modulation is allowed to occupy. This is essentially a reduction in the Channel Spacing
for example, for 15khz channel spacing the authorized bandwidth is 11.25 kHz
?
Then, there is the bandwidth indicated by the emission designator.? For analog FM it is determined by Carson's rule based on a single 3 kHz frequency causing the indicated peak deviation.
?
Thus, for Analog FM (5.0 kHz peak deviation)?
?
Emissions designator: 16K0F3E?
Occupied bandwidth:? 12.59 kHz
Authorized bandwidth: 20 kHz (25 or 30 kHz channel spacing)
?
Accumulating a peak hold trace on a spectrum analyzer will certainly tell a lot about a transmitter's bandwidth utilization, but, in fact - OBW is a defined concept with a specified measurement procedure.
The actual measurement procedure uses a transmitter test pattern comprised of two tones, one at 650 Hz and another at 2,200 Hz The goal is to simulate normal modulation for generating adjacent channel power data.
?
So, if we know the OBW of a transmitter do we know how well it will play with neighbors (i.e. adjacent channels)?
?
Not really!? 99% of the power seems like it wouldn't leave enough to be a problem.? How much trouble can be caused by a half percent on either side?? Maybe a lot.
?
1% is actually just -20 dBc so 1/2% is -23 dBc. If we have 100 watts, -23dBc is a half watt.? If that half watt is in the neighbor's receiver passbband - how much trouble could we cause with a half watt into a DB224?
?
Hmmm.? There must be a better way.? There is.? More on that next time.
|
Touching again on OBW before moving on, in the other discussion I asked Burt to explain the procedure he uses to measure occupied bandwidth.
Burt replied:? "For occupied bandwidth measurements I use an Anritsu MS2721B spectrum analyzer.? I set the span to 100 kHz, the RBW (Resolution Bandwidth) to 100 Hz, and VBW (Video Bandwidth) to 30 kHz, and also use peak store.? With the repeater in use that has a mix of fully quieting and noisy signals I let the analyzer sweep for 10 minutes or longer."
?
The industry procedure uses the same 100 Hz RBW and since the goal of the OBW test pattern is to simulate normal modulation, either should produce comparable results
?
Once the trace is acquired most modern analyzers will compute the 99% OBW value at the touch of a button.? This will provide a baseline record for comparison to industry expectations and for comparison when experimenting with different transmit audio processing.
|
What next?
?
In Burt's discussion, Bob Dengler, N06B, said:? Also, the audio has to have a flat response (aside from the usual 6 dB/octave pre-emphasis).? ?Any extra slope that favors the high frequencies or hard clipping will shift the PSD to the high end, greatly increasing the occupied bandwidth.
?
Bob didn't elaborate and no one picked up discussion of PSD but that is a great next topic.
?
PSD (Power Spectral Density) is the power density per unit bandwidth.? In TIA documents, PSD is referred to as SPD (Spectral Power Density.? ?SPD gives an idea about how the power of the signal is distributed over frequency.
?
A transmitter's SPD can be characterized by a power density measurement of its emissions over a specified frequency span using a spectrum analyzer?which presents a graphic display comprised of a number of discrete data points, each data point representing the amount of power measured in a ¡°frequency bin¡±
?
The TIA has published numerous documents to assist frequency coordinators, engineers, and coverage specialists in understanding how various modulation formats perform.?
These documents contain emission designators and graphs that show the SPD waveform for each modulation format to illustrate actual bandwidth requirements.
?
Life was easier when everything was 16K0F3E.? With the proliferation of formats including Securenet, EDACS, P25, DMR, Tetra, etc.? frequency coordinators face new challenges.
Amateurs have also faced the proliferation of modulation formats with folks experimenting with many of the commercial / public safety formats as well as new amateur digital formats such as D-Star and Fusion.
?
D-Star came first and presented new opportunities for frequency coordinators along with technical confusion - especially since Icom claimed d-Star signals fit in 6 kHz of spectrum.? They published numerous documents containing this claim.
?
Some hams initially accepted the claim.? In the left side of the attached image are some snips from Icom documents.? On the right is a graphic from the Illinois Repeater association that
clearly shows the acceptance of the claim that the entire D-Star modulation fits into 6 kHz of spectrum.
?
Astute hams soon investigated and realized the claim was a misrepresentation.??
?
Hopefully, as we continue the bandwidth discussion it will become clear how to assess the actual bandwidth requirements of various formats - especially in the context of band planning.
?
|
What might follow to your comments, Neil, is that as far as additional transmitter bandwith added by the PL component I think it is not really the additional say +- 0.7KHz, but more like just the PL frequency say +_0.15 KHz. To back that up, once a fellow ham and I did a crude over the air test using a Micor PM modulated transmitter with and without PL. We were about 40 miles apart and he was hearing us with about 50 % noise and his squelch was set such that for every 10 times I said a loud "five", his receiver would squelch out on my voice peak 3 out of 10 times. That did not vary with or without PL. Ever since then I set peak dev on the initiatiing mobile, base or portable to 4.8 KHz assuming perfect deviation symmetry with less than 100cps frequency drift, and then disregard the 0.7 KHz PL deviation where the service monitor would read +_5.5 KHz (unless it was set up for only sensing 300-3000cps. With these settings maximum range is achieved, with just a slight occasional increase in distortion over +-3KHz dev on voice? peaks(how a lot of receivers seem to be rated when measuring audio distortion).
John W1GPO
On 11/11/2022 8:31 PM nj902 <nj0907@...> wrote:
What next?
In Burt's discussion, Bob Dengler, N06B, said:? Also, the audio has to have a flat response (aside from the usual 6 dB/octave pre-emphasis).? ?Any extra slope that favors the high frequencies or hard clipping will shift the PSD to the high end, greatly increasing the occupied bandwidth.
Bob didn't elaborate and no one picked up discussion of PSD but that is a great next topic.
PSD (Power Spectral Density) is the power density per unit bandwidth.? In TIA documents, PSD is referred to as SPD (Spectral Power Density.? ?SPD gives an idea about how the power of the signal is distributed over frequency.
A transmitter's SPD can be characterized by a power density measurement of its emissions over a specified frequency span using a spectrum analyzer?which presents a graphic display comprised of a number of discrete data points, each data point representing the amount of power measured in a ¡°frequency bin¡±
The TIA has published numerous documents to assist frequency coordinators, engineers, and coverage specialists in understanding how various modulation formats perform.?
These documents contain emission designators and graphs that show the SPD waveform for each modulation format to illustrate actual bandwidth requirements.
Life was easier when everything was 16K0F3E.? With the proliferation of formats including Securenet, EDACS, P25, DMR, Tetra, etc.? frequency coordinators face new challenges.
Amateurs have also faced the proliferation of modulation formats with folks experimenting with many of the commercial / public safety formats as well as new amateur digital formats such as D-Star and Fusion.
D-Star came first and presented new opportunities for frequency coordinators along with technical confusion - especially since Icom claimed d-Star signals fit in 6 kHz of spectrum.? They published numerous documents containing this claim.
Some hams initially accepted the claim.? In the left side of the attached image are some snips from Icom documents.? On the right is a graphic from the Illinois Repeater association that
clearly shows the acceptance of the claim that the entire D-Star modulation fits into 6 kHz of spectrum.
Astute hams soon investigated and realized the claim was a misrepresentation.??
Hopefully, as we continue the bandwidth discussion it will become clear how to assess the actual bandwidth requirements of various formats - especially in the context of band planning.
|
Gang, The chart showing the ICOM D-Start bandwidth / PSD claims is nice with one exception...? It doesn't tell you how far down the sideband energy is relative to the reference carrier.? Without that information it's just a pretty picture.? Am I missing something? Burt, K6OQK 
|
There is a real problem when relying on training propaganda and brochures.?
With FM analog, you can see some bandwidth, but some voice peak happens
at some frequency and another pair of sidebands will pop up with sufficient
energy to throw noise into the adjacent channel.? With the different DMR
modes, there are very different bandwidth characteristics.? IF you find the
lab data that was submitted, at least you might see actual traces rather than
a brochure cartoon.? Receiver bandwidth is important too, as a practical matter.
The idea of 15/20/25khz does not fly in So Cal where the best hope was for
20khz rx bandwidth.? How do you get 15khz RX bandwidth without changing
the TX to +- 2.4khz dev and 12.5khz channel spacing.? Calls for mods for
every user radio and attention to frequency accuracy beyond most users.
|
On Sat, Nov 12, 2022 at 10:29 AM, Burt K6OQK wrote:
Without that information it's just a pretty picture
Exactly.? Icom let the marketing guys put that together while the engineers were having lunch. Attached is a comparison of the IRA graphic to real D-Star PSD graph with lines added to show how much of the emission falls outside of 6 kHz of spectrum.
|
John wrote: "What might follow to your comments, Neil, is that as far as additional transmitter bandwidth?added by the PL component ....."
John, thank you for your thought provoking comments.? I believe that you are correct - low frequency modulation components add directly to total deviation but probably have a lesser contribution to total bandwidth than higher voice frequency components
|
Chris and JB posted some excellent comments.? I'd like to return later to address those after covering some additional basics. Those are: Receiver ACRR and Transmitter ACPR.
?
First - receivers.? How well a receiver will tolerate a neighboring signal is based on its selectivity or adjacent channel rejection.??Adjacent channel rejection is the ratio of the level of an unwanted adjacent channel input signal to the reference sensitivity.? ?Because it is a ratio it is commonly referred to as the Adjacent Channel Rejection Ratio (ACRR).
?
ACRR is largely determined by the characteristics of the receiver's IF filter including type of implementation (i.e., ceramic, digital), shape factor, and especially ENBW.??
?
ENBW (Equivalent Noise Bandwidth): is the frequency span of an ideal filter which will pass the same value of noise as the actual filter implementation.? Sometimes ENBW is referred to as Effective Noise Bandwidth.? In the 'old days' - when receivers were totally hardware based - ENBW could be measured with test equipment.? Today, ENBW is typically inferred from the manufacturer's adjacent channel rejection specifications.
?
There is a defined ACRR measurement procedure that employs a test signal modulated by two test tones, one at 650 Hz at a deviation of 50% of the maximum permissible frequency deviation, and another at 2,200 Hz at a deviation of 50% of the maximum permissible frequency deviation.
?
In analysis of co-channel interference, the receiver is often referred to as the "victim".? In many cases, especially like the discussion of amateur use of 15 kHz spacing, the?victim receivers are fielded equipment whose characteristics cannot be modified.
?
Transmitter ACPR (Adjacent channel power ratio)?
ACPR is an increasingly critical measurement for many of the current and future transmission standards.? ?ACPR, sometimes also termed adjacent channel leakage ratio (ACLR), is a test for characterizing the likelihood that a given system may cause interference with a neighboring one.
?
In TIA terms, Adjacent Channel Power Ratio is an "Unwanted Emission" defined as the ratio of the total output power of a transmitter under defined conditions and modulation, to that part of the output power that falls within a specified passband centered on the nominal frequency of either of the adjacent channels above or below the assigned carrier frequency.? The specified passband is the ENBW IF bandwidth of the adjacent channel's "victim" receiver.
?
Once we have acquired a transmitter's SPD data, ACPR can be calculated for any channel spacing and victim receiver ENBW.? ?In some cases, FCC and TIA require ACPR to be greater than 50 dB.? 60 dB is a more typical requirement.
?
Since the subject of amateur band plan channel spacing has come up, let's look at the issue.
?
The attached comparison shows the difference in ACPR for a victim receiver having a 16 kHz ENBW at channel spacings of 15 kHz and 20 kHz.? The transmitter is based on 16K0F3E. The SPD data is the published TIA data for that format.? Note that at 20 kHz channel spacing, the ACPR is a very acceptable 63.3 dB whereas at 15 kHz spacing the ACPR is an unacceptable 27.2 dB.??
A huge difference that illustrates why 15 kHz channel spacing is so problematic requiring consideration of geographic separation of systems and service areas as well as propagation issues.
?
?
|
At 11/12/2022 05:30 PM, you wrote: Chris and JB posted some excellent comments. I'd like to return later to address those after covering some additional basics. Those are: Receiver ACRR and Transmitter ACPR. There is a defined ACRR measurement procedure that employs a test signal modulated by two test tones, one at 650 Hz at a deviation of 50% of the maximum permissible frequency deviation, and another at 2,200 Hz at a deviation of 50% of the maximum permissible frequency deviation. I'm curious as to how the above frequencies & deviations were arrived at. Are they supposed to approximate the occupied BW when human speech is the modulating source? In analysis of co-channel interference, the receiver is often referred to as the "victim". In many cases, especially like the discussion of amateur use of 15 kHz spacing, the victim receivers are fielded equipment whose characteristics cannot be modified. This generally does not apply in SoCal, where the bandplan in the 15 kHz spaced 146-148 MHz segment calls for inverting the offset direction every 15 kHz. That puts the repeater RXs "see" adjacent channel repeater TXs, not user TXs. Those TXs are typically of higher quality than user radios, & since they are coordinated their characteristics can definitely be addressed if needed. Since the subject of amateur band plan channel spacing has come up, let's look at the issue.
The attached comparison shows the difference in ACPR for a victim receiver having a 16 kHz ENBW at channel spacings of 15 kHz and 20 kHz. The transmitter is based on 16K0F3E. The SPD data is the published TIA data for that format. Note that at 20 kHz channel spacing, the ACPR is a very acceptable 63.3 dB whereas at 15 kHz spacing the ACPR is an unacceptable 27.2 dB. I think we've all agreed that 5 kHz deviation is not compatible with 15 kHz channel spacing unless significant geographic separation is used. Here in SoCal we settled on 4.2 kHz some 20 years ago to lessen the amount of distance needed between adjacents. Bob NO6B
|
Hi Bob.
?
Bob said: "I'm curious as to how the above frequencies & deviations were arrived at. Are they supposed to approximate the occupied BW when human speech is the modulating source?"
?
That's exactly what the TIA says about the analog SPD Interference transmitter test pattern: "The goal is to simulate normal modulation for generating adjacent channel power data."
?
These standards are developed by working groups comprised of subject matter experts.? Somewhere in the dim dark past there must have been an interesting technical discussion about goals and alternatives. This may have taken place before widespread internet use so finding an archive could be challenging.
?
Also, Bob said: "Here in SoCal we settled on 4.2 kHz some 20 years ago to lessen the amount of distance needed between adjacents."
Bob, earlier I mentioned your comment: "Also the audio has to have a flat response (aside from the usual 6 dB/octave pre-emphasis).??
Any extra slope that favors the high frequencies or hard clipping will shift the PSD to the high end, greatly increasing the occupied bandwidth."
Do you have any test results you can share to illustrate this?
?
?
Attached is a comparison of ACPR for 5 kHz deviation vs. 4 kHz deviation.? Note that there is over 8 dB of improvement by using the narrower format.
?
|
|
|
One more look at D-Str - as Burt said, a chart without numbers is just a pretty picture.
In their literature, Icom said: "D-Star channel spacing = more repeaters"? AND "Requiring only a 6 kHz channel...D-Star repeaters can be interleaved between existing channels..."
It looks much less like a good idea when you have the numbers!
|
It appears that I need to add a little information to my notes as to how I make Occupied Bandwidth measurements.
For analog systems, all measurements are made relative to the amplitude of the un-modulated carrier. For those of you familiar with broadcast measurements, this is the same as FM NRSC RF Proofs and very similar to AM NRSC measurements.The Laws of Physics are the same. See FCC 73.317 for FM and FCC 73.44 for AM measurements. Without the un-modulated peak carrier reference the results of any occupied bandwidth measurements are meaningless. Measurements should be made under normal operating conditions with modulation containing typical audio modulation, not tones.
One other thing that needs to be kept in mind is that deviation refers to one direction only from the carrier.? If you're talking about both sides, that's "swing (or P-P).? I mention this because some people seem to confuse the two when talking about Occupied Bandwidth.
I'm sure I'll think of more.
Burt, K6OQK
Hide quoted text
?
toggle quoted message
Show quoted text
On Sat, Nov 5, 2022 at 10:39 AM, Burt K6OQK wrote:
For occupied bandwidth measurements I use an Anritsu MS2721B spectrum analyzer.? I set the span to 100 kHz, the RBW (Resolution Bandwidth) to 100 Hz, and VBW (Video Bandwidth) to 30 kHz, and also use peak store.? With the repeater in use that has a mix of fully quieting and noisy signals I let the analyzer sweep for 10 minutes or longer.? It's best if I make these measurements at the sample out of the transmitter, but I can get a fairly good representation of reality from home.
|
Kevin Custer
Burt K6OQK
