Elecraft

[P3] How to measure bandwidth of RX signal

Classic

List

Threaded

9 messages Options

NK7Z

[P3] How to measure bandwidth of RX signal

Hi,

All the ESSB talk got me wondering about how accurate the P3 might be
for measuring bandwidth. Could someone review my method for this
outlined below and comment please:

1. Use ATT if needed to avoid very strong signals
2. Set bandwidth to 4 KHz. See note below.
3. Adjust Span on P3 5 KHz per side
4. Shaded area above waterfall indicates K3 bandwidth

Watch P3. Can one trust the waterfall display to show actual bandwidth?
If so, how accurate would it be?

Note:
I believe that the P3 is tapped prior to any BW restrictions, (of merit
to this exercise), so I understand that step 2 is just for my ears only,
and does not affect BW display on the P3. Is that a correct assumption?

--
Thanks and 73's,
For equipment, and software setups and reviews see:
www.nk7z.net
for MixW support see;
http://groups.yahoo.com/neo/groups/mixw/info
for Dopplergram information see:
http://groups.yahoo.com/neo/groups/dopplergram/info
for MM-SSTV see:
http://groups.yahoo.com/neo/groups/MM-SSTV/info

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Don Wilhelm-4

Re: [P3] How to measure bandwidth of RX signal

David,

Your assumption is correct. The P3 should show you the true width of
the signal.

73,
Don W3FPR

On 4/28/2014 10:08 AM, David Cole wrote:

> Hi,
>
> All the ESSB talk got me wondering about how accurate the P3 might be
> for measuring bandwidth. Could someone review my method for this
> outlined below and comment please:
>
> 1. Use ATT if needed to avoid very strong signals
> 2. Set bandwidth to 4 KHz. See note below.
> 3. Adjust Span on P3 5 KHz per side
> 4. Shaded area above waterfall indicates K3 bandwidth
>
> Watch P3. Can one trust the waterfall display to show actual bandwidth?
> If so, how accurate would it be?
>
> Note:
> I believe that the P3 is tapped prior to any BW restrictions, (of merit
> to this exercise), so I understand that step 2 is just for my ears only,
> and does not affect BW display on the P3. Is that a correct assumption?
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Alan Bloom-2

Re: [P3] How to measure bandwidth of RX signal

In reply to this post by NK7Z

> I believe that the P3 is tapped prior to any BW restrictions,

Correct. I think the best way to measure an SSB signal's bandwidth on
the P3 is to turn on peak hold. After a few seconds the spectrum
display will show a pretty accurate picture of the spectral shape of the
signal

Alan N1AL

On 04/28/2014 07:08 AM, David Cole wrote:

Vic Rosenthal

Re: [P3] How to measure bandwidth of RX signal

In reply to this post by NK7Z

'Bandwidth' is measured by specifying some number of dB that a signal is
down at a particular offset from the center frequency. A measurement in
Hz alone is therefore meaningless.

I suspect that the number of dB needed to make the waterfall display of
a signal disappear depends on various things, like the gain setting. So
this is not an accurate way to measure it.

What should work is to use the spectrum display of the P3. If you want
to know, for example, the bandwidth of a signal at 30 dB down, you just
find the points where the 'skirts' of the signal are 30 dB below the
peak. This is easy to do on the P3 which can display the signal strength
in dBm.

On 4/28/2014 7:08 AM, David Cole wrote:

--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco/

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

NK7Z

Re: [P3] How to measure bandwidth of RX signal

Hello,

Between Vic's, and Alan's answer I have what I need, thank you all!!! I
can now measure bandwidth, and tell how many db down a signal is vs.
it's width. Thank you gents!

Peak hold, then measure! :)

--
Thanks and 73's,
For equipment, and software setups and reviews see:
www.nk7z.net
for MixW support see;
http://groups.yahoo.com/neo/groups/mixw/info
for Dopplergram information see:
http://groups.yahoo.com/neo/groups/dopplergram/info
for MM-SSTV see:
http://groups.yahoo.com/neo/groups/MM-SSTV/info

On Mon, 2014-04-28 at 08:37 -0700, Vic Rosenthal K2VCO wrote:

> 'Bandwidth' is measured by specifying some number of dB that a signal is
> down at a particular offset from the center frequency. A measurement in
> Hz alone is therefore meaningless.
>
> I suspect that the number of dB needed to make the waterfall display of
> a signal disappear depends on various things, like the gain setting. So
> this is not an accurate way to measure it.
>
> What should work is to use the spectrum display of the P3. If you want
> to know, for example, the bandwidth of a signal at 30 dB down, you just
> find the points where the 'skirts' of the signal are 30 dB below the
> peak. This is easy to do on the P3 which can display the signal strength
> in dBm.
>

Jim Brown-10

Re: [P3] How to measure bandwidth of RX signal

In reply to this post by Vic Rosenthal

On 4/28/2014 8:37 AM, Vic Rosenthal K2VCO wrote:
> What should work is to use the spectrum display of the P3. If you want
> to know, for example, the bandwidth of a signal at 30 dB down, you
> just find the points where the 'skirts' of the signal are 30 dB below
> the peak. This is easy to do on the P3 which can display the signal
> strength in dBm.

Yes. And as N1AL recommends, use peak hold. I've done spectrum plots of
various signals and conditions with the P3. I usually set the SCALE for
a pretty wide range (like 72 or the full dB), and set the SPAN to about
10 kHz. I use both the Peak Hold Mode and the Average Mode. In both
modes, but especially with Peak Hold, we must take care to differentiate
sidebands from other stations. I try to set the reference level so that
the peak of the signal is at one of the level lines and the noise level
is near the bottom of the display.

With almost any rig, you'll see small signals down the sides of the
slope that are intermod products, and they are usually symmetrical on a
CW or RTTY signal. A clean SSB signal should be well confined by the
bandpass filter limits of the audio and TX filter skirts. Anything that
extends beyond that is IMD. When a rig is badly overdriven, it's common
to see splatter as horizontal streaks extending above and below the
signal frequency.

Last week, I was on 20M SSB trying to work a weak DX station, but what I
heard was massive splatter from a QSO about 5 kHz up the band. One guy
was relatively clean, but the other was not. I broke the QSO to tell him
he was clobbering a weak DX station. Turns out he was an aeronautical
mobile using whatever radio was onboard. He didn't know whether he was
transmitting AM or SSB, but thought it was AM. I assured him that it
was, indeed, USB, and that the problem was that he was likely
overdriving it badly. He went QRT to figure it out.

As a percentage of transmitted signals, I don't find contesters any
better or worse than non-contesters -- we contesters stand out because
during a contest, there are a LOT more of us. :) I'm retired, so I have
access to the ham bands during the day (when I'm not working on
something else). It's not at all uncommon to tune around 30, 40, and 80
during a weekday and hear fewer than a half dozen signals on these bands
(if you hear any at all), and no more than a few dozen on most of the
higher bands. Compare that to a major contest, when there are several
hundred workable signals on every band from any given location for the
duration of the contest, and for stations with better antennas, perhaps
2-4X that number. The winners of contests like the ARRL SSB Sweepstakes
make more than 2,000 QSOs, and you can work a station only once.

73, Jim K9YC
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

NK7Z

Re: [P3] How to measure bandwidth of RX signal

Jim,

Thank you for the extended explanation, it helps! I want to be sure I
understand the rig, the Panadaptor, and how they interrelate. THank
you again!

--
Thanks and 73's,
For equipment, and software setups and reviews see:
www.nk7z.net
for MixW support see;
http://groups.yahoo.com/neo/groups/mixw/info
for Dopplergram information see:
http://groups.yahoo.com/neo/groups/dopplergram/info
for MM-SSTV see:
http://groups.yahoo.com/neo/groups/MM-SSTV/info

On Mon, 2014-04-28 at 09:27 -0700, Jim Brown wrote:

> On 4/28/2014 8:37 AM, Vic Rosenthal K2VCO wrote:
> > What should work is to use the spectrum display of the P3. If you want
> > to know, for example, the bandwidth of a signal at 30 dB down, you
> > just find the points where the 'skirts' of the signal are 30 dB below
> > the peak. This is easy to do on the P3 which can display the signal
> > strength in dBm.
>
> Yes. And as N1AL recommends, use peak hold. I've done spectrum plots of
> various signals and conditions with the P3. I usually set the SCALE for
> a pretty wide range (like 72 or the full dB), and set the SPAN to about
> 10 kHz. I use both the Peak Hold Mode and the Average Mode. In both
> modes, but especially with Peak Hold, we must take care to differentiate
> sidebands from other stations. I try to set the reference level so that
> the peak of the signal is at one of the level lines and the noise level
> is near the bottom of the display.
>
> With almost any rig, you'll see small signals down the sides of the
> slope that are intermod products, and they are usually symmetrical on a
> CW or RTTY signal. A clean SSB signal should be well confined by the
> bandpass filter limits of the audio and TX filter skirts. Anything that
> extends beyond that is IMD. When a rig is badly overdriven, it's common
> to see splatter as horizontal streaks extending above and below the
> signal frequency.
>
> Last week, I was on 20M SSB trying to work a weak DX station, but what I
> heard was massive splatter from a QSO about 5 kHz up the band. One guy
> was relatively clean, but the other was not. I broke the QSO to tell him
> he was clobbering a weak DX station. Turns out he was an aeronautical
> mobile using whatever radio was onboard. He didn't know whether he was
> transmitting AM or SSB, but thought it was AM. I assured him that it
> was, indeed, USB, and that the problem was that he was likely
> overdriving it badly. He went QRT to figure it out.
>
> As a percentage of transmitted signals, I don't find contesters any
> better or worse than non-contesters -- we contesters stand out because
> during a contest, there are a LOT more of us. :) I'm retired, so I have
> access to the ham bands during the day (when I'm not working on
> something else). It's not at all uncommon to tune around 30, 40, and 80
> during a weekday and hear fewer than a half dozen signals on these bands
> (if you hear any at all), and no more than a few dozen on most of the
> higher bands. Compare that to a major contest, when there are several
> hundred workable signals on every band from any given location for the
> duration of the contest, and for stations with better antennas, perhaps
> 2-4X that number. The winners of contests like the ARRL SSB Sweepstakes
> make more than 2,000 QSOs, and you can work a station only once.
>
> 73, Jim K9YC
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]

Bill Frantz

Re: [P3] How to measure bandwidth of RX signal

For a look at good bandwidth control, tune down into the AM
broadcast band with the K3 and P3. Note that AM stations are
powerful enough that you can receive them at the upper end of
the broadcast band even without the extended receive option for
the K3.

When I tuned down there in the San Francisco Bay Area, I could
see stations right next to each other, each with significant
power stretching for nearly the full 10 KHz permitted bandwidth
and a very small sliver of clear bandwidth between each station.
It is a wonderful demonstration of good spectrum utilisation.

Cheers - Bill, AE6JV

---------------------------------------------------------------------------
Bill Frantz |"We used to quip that "password" is the most common
408-356-8506 | password. Now it's 'password1.' Who said
users haven't
www.pwpconsult.com | learned anything about security?" -- Bruce Schneier

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Phil Kane-2

Re: [P3] How to measure bandwidth of RX signal

In reply to this post by NK7Z

On 4/28/2014 8:57 AM, David Cole wrote:

> Between Vic's, and Alan's answer I have what I need, thank you all!!! I
> can now measure bandwidth, and tell how many db down a signal is vs.
> it's width. Thank you gents!

As others will probably tell you as well, Occupied Bandwidth is defined
as the spectrum occupied between the --26 dB points. This defines 99%
of the signal.
-- --
73 de K2ASP - Phil Kane
Elecraft K2/100 s/n 5402

From a Clearing in the Silicon Forest
Beaverton (Washington County) Oregon
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]