1. WSJTX | Main
  2. Topics

Locked Interesting double-decode

Bill Lederer
 

Jim:

Dude, you were going neck-and-neck with me on 80 and 20!  Miss those days.

w8lvn


On Sat, Feb 22, 2020 at 3:15 PM Jim Brown <k9yc@...> wrote:
On 2/22/2020 11:42 AM, Bill Lederer wrote:
> So i am under the impression that switching mode power supplies don't
> operate at mains frequency, or at least their switching will spit off an
> array of frequencies due to the switching. AC Mains I can understand,
> but I think a switching mode power supply will emit some range of
> frequencies that may not be related to the mains frequency.

Right. SMPS start by rectifying AC, there's a minimal capacitor filter,
then a square wave generator running typically running in the range of
10-40kHz chops the DC, which is then rectified and filtered by a much
smaller capacitor. RF noise results from the harmonics of the square
wave, and often extend to 10M. A very high fraction of the RF noise we
hear is generated by SMPS. I estimate that the average home has several
dozen, and wiring connected to them, both on the AC side and the DC
side, radiates the noise as the result of common mode current.

Lots of detail in this tutorial.

http://k9yc.com/KillingReceiveNoise.pdf

BTW -- Bill, you and I met around 2003-4 sitting next to each other on
NSRC Field Day. Your BIC time was an inspiration to me!

73, Jim K9YC





--
--w8lvn--
parker@masterdigital.com <parker@...>
 

20200207_121045 -10 0.3 2741 ~ TI9A W5CIA EM40
20200207_121045 -4 0.3 3221 ~ TI9A W5CIA EM40
20200207_121045 2 0.3 2141 ~ TI9A W5CIA EM40
20200207_121045 3 0.3 1301 ~ TI9A W5CIA EM40
20200207_121045 5 0.3 1541 ~ TI9A W5CIA EM40
20200207_121045 6 0.3 1781 ~ TI9A W5CIA EM40
20200207_121045 7 0.3 1421 ~ TI9A W5CIA EM40
20200207_121045 6 0.4 1950 ~ TI9A N0FW -07 <---- clean signal

This is not caused 1) on the receive end, 2) by plane scatter, nor 3) by any other reflections in the near or far field.

I am using a Flex 6700 connected to the computer over LAN only, and the station is apparently using a 160m horse fence antenna 3,800 ft from me. I know the cause, but the problem persists.

Parker W5ADD



Parker W5ADD
Paul <w8aef@...>
 

I see that effect quite often on 6m from a station that is on the other side of the Phoenix airport.
 
Paul, W8AEF


From: WSJTX@groups.io [mailto:WSJTX@groups.io] On Behalf Of Martin G0HDB
Sent: Saturday, February 22, 2020 1:16 PM
To: WSJTX@groups.io
Subject: Re: [WSJTX] Interesting double-decode

On Sat, Feb 22, 2020 at 07:57 PM, Philip Rose wrote:

So, I see the same signal being decoded several 100 Hz apart on 20m, is this a Doppler effect from what is probably the busiest air route in Europe?
 
I still contend it's over-driving of transmitters. I keep my ALC less than 0dB on datamodes.
The spacing of a Doppler-shifted signal on 20m is, in my opinion, extremely unlikely to ever reach several 100Hz; the plane causing the Doppler would probably need to be moving at hypersonic speeds!  I have to confess I've never seen multiples of the same signal (other than those using FT8 Fox & Hounds mode or the MSHV software) with spacings of the several 100Hz you mention.

In my experience, over-driving a transmitter's audio input usually results in harmonics of the eight FT8 tones that appear as one or more 'bar-code' signals elsewhere across the audio spectrum.  Sadly this is very common.

As for ALC, I believe FT8 uses a constant-amplitude waveform so I'm not sure what effect ALC might have on that; I wouldn't expect running with a degree of ALC to have any adverse effect on the purity of the transmitted signal.  However, I recall there have been many previous discussions on this so I probably need to refresh my understanding of this aspect!  Does ALC have any adverse effect on the quality of an RTTY signal, which is in effect a 2-tone FSK signal rather than the 8-tone (G)FSK signal used by FT8?

--
Martin G0HDB

Denny - K0TT
 

I see this occasionally on HF, usually when the geomagnetic field is disturbed. I assume it's caused by multipath or doppler distortion. 
Jim Brown
 

On 2/22/2020 12:58 PM, Jim Shorney wrote:
Noise blanker IMD can cause multiple decodes based on the frequency of repetition rate of your impulse noise.
YES!

73, Jim K9YC
Jim Brown
 

On 2/22/2020 11:48 AM, Bill Lederer wrote:
So that would be multipath then likely?
Yes. There can also be multi-path on the HF bands -- it's the fundamental cause of both "selective fading," and "picket-fencing," the difference being the period of the fading directly related to wavelength. It's the addition of two copies of the same signal that have taken different paths, and are thus shifted in time from each other. That time variation results in a phase difference that is proportional to frequency.

73, Jim K9YC
Jim Brown
 

On 2/22/2020 11:42 AM, Bill Lederer wrote:
So i am under the impression that switching mode power supplies don't operate at mains frequency, or at least their switching will spit off an array of frequencies due to the switching. AC Mains I can understand, but I think a switching mode power supply will emit some range of frequencies that may not be related to the mains frequency.
Right. SMPS start by rectifying AC, there's a minimal capacitor filter, then a square wave generator running typically running in the range of 10-40kHz chops the DC, which is then rectified and filtered by a much smaller capacitor. RF noise results from the harmonics of the square wave, and often extend to 10M. A very high fraction of the RF noise we hear is generated by SMPS. I estimate that the average home has several dozen, and wiring connected to them, both on the AC side and the DC side, radiates the noise as the result of common mode current.

Lots of detail in this tutorial.

http://k9yc.com/KillingReceiveNoise.pdf

BTW -- Bill, you and I met around 2003-4 sitting next to each other on NSRC Field Day. Your BIC time was an inspiration to me!

73, Jim K9YC
Jim Shorney
 

Noise blanker IMD can cause multiple decodes based on the frequency of repetition rate of your impulse noise.

73

-Jim
NU0C

On Sat, 22 Feb 2020 12:15:41 -0800
"Martin G0HDB" <marting0hdb@...> wrote:

On Sat, Feb 22, 2020 at 07:57 PM, Philip Rose wrote:



So, I see the same signal being decoded several 100 Hz apart on 20m, is
this a Doppler effect from what is probably the busiest air route in
Europe?

I still contend it's over-driving of transmitters. I keep my ALC less than
0dB on datamodes.
The spacing of a Doppler-shifted signal on 20m is, in my opinion, extremely unlikely to ever reach several 100Hz; the plane causing the Doppler would probably need to be moving at hypersonic speeds!  I have to confess I've never seen multiples of the same signal (other than those using FT8 Fox & Hounds mode or the MSHV software) with spacings of the several 100Hz you mention.

In my experience, over-driving a transmitter's audio input usually results in harmonics of the eight FT8 tones that appear as one or more 'bar-code' signals elsewhere across the audio spectrum.  Sadly this is very common.

As for ALC, I believe FT8 uses a constant-amplitude waveform so I'm not sure what effect ALC might have on that; I wouldn't expect running with a degree of ALC to have any adverse effect on the purity of the transmitted signal.  However, I recall there have been many previous discussions on this so I probably need to refresh my understanding of this aspect!  Does ALC have any adverse effect on the quality of an RTTY signal, which is in effect a 2-tone FSK signal rather than the 8-tone (G)FSK signal used by FT8?

--
Martin G0HDB
Philip Rose
 



On 22 Feb 2020 20:15, Martin G0HDB <marting0hdb@...> wrote:
On Sat, Feb 22, 2020 at 07:57 PM, Philip Rose wrote:

So, I see the same signal being decoded several 100 Hz apart on 20m, is this a Doppler effect from what is probably the busiest air route in Europe?
 

So this was sarcasm. 
I still contend it's over-driving of transmitters. I keep my ALC less than 0dB on datamodes.
The spacing of a Doppler-shifted signal on 20m is, in my opinion, extremely unlikely to ever reach several 100Hz; the plane causing the Doppler would probably need to be moving at hypersonic speeds!  I have to confess I've never seen multiples of the same signal (other than those using FT8 Fox & Hounds mode or the MSHV software) with spacings of the several 100Hz you mention.

In my experience, over-driving a transmitter's audio input usually results in harmonics of the eight FT8 tones that appear as one or more 'bar-code' signals elsewhere across the audio spectrum.  Sadly this is very common.

I was suggesting that over-driving the pa would cause IMD constructs. I don't really understand the difference between digital modes. Just keep the ALC below zero.

As for ALC, I believe FT8 uses a constant-amplitude waveform so I'm not sure what effect ALC might have on that; I wouldn't expect running with a degree of ALC to have any adverse effect on the purity of the transmitted signal.  However, I recall there have been many previous discussions on this so I probably need to refresh my understanding of this aspect!  Does ALC have any adverse effect on the quality of an RTTY signal, which is in effect a 2-tone FSK signal rather than the 8-tone (G)FSK signal used by FT8?

--
Martin G0HDB


--
73 Phil GM3ZZA
Bill Lederer
 

That is a great article.

Thanks.


On Sat, Feb 22, 2020 at 1:58 PM Martin G0HDB <marting0hdb@...> wrote:

[Edited Message Follows]
[Reason: Added link to the article by GM4FVM]

Nope, it's AC mains-related.  If there's any significant 50Hz or 60Hz ripple coming out of the PSU (especially an SMPSU) on the DC line that's powering the rig, or if there's anything else in the shack that might be injecting a ripple signal somewhere in the audio signal chain, then this will be superimposed on the transmitted audio and will appear as a decodable 'sideband', in the case of the K5DDE decodes the spacing of the spurious signal is 2 x the AC line frequency.

It's not uncommon to see more than two spurii, I think I've seen up to four, all with spacings of exact multiples of the local AC mains frequency, on the signals emanated by some Eastern EU stations...!  I can even see a signal 100Hz away from a nearby amateur's very strong FT8 signal; even though his unintentional sideband is approaching 40dB down on his fundamental signal I would prefer it if it weren't there at all.

I've just found a very informative article on the subject; you''ll find it at:

http://gm4fvm.blogspot.com/2019/11/ft8-and-100hz-ripple-strange-one-for-me.html

--
Martin G0HDB


--
--w8lvn--
Philip Rose
 

Can we agree that this discussion is futile. There are numerous reasons why this happens.

73 Phil GM3ZZA.

On 22 Feb 2020 19:56, Reino Talarmo <reino.talarmo@...> wrote:

Hi Phil,

Of course it does not. There are multiple reasons for additional sidebands such as Doppler shift due to airplane reflections, if also the direct signal is available (multipath). Then normally difference frequency is changing.

By the way this is simple to simulate and results are close even on the relative power difference.
If there is a switching power supply, it may still pass 120 Hz spikes through and narrow spikes do give a closer amplitude distribution in the case of multiple sidebands.

Power supplies of tube amplifiers are seldom switching supplies and if rectifying diodes are not properly bypassed by a snubber, just a capacitor is fine, then those recovery peaks can propagate into anode voltage as well.

73, Reino oh3mA

 

From: WSJTX@groups.io [mailto:WSJTX@groups.io] On Behalf Of Philip Rose via Groups.Io
Sent: 22. helmikuuta 2020 21:31
To: WSJTX@groups.io
Subject: Re: [WSJTX] Interesting double-decode

 

 

 

On 22 Feb 2020 19:21, Reino Talarmo <reino.talarmo@...> wrote:

Hi,

That is normally generated by a power amplifier with less than perfect battery voltage regulation so that a full wave rectification leaves a 120 Hz sawtooth wave riding on top of the battery voltage. If the power amplifier is driven full, then result is 120 Hz sawtooth wave amplitude modulation. Sometimes there are multiple sidebands with 120 Hz spacing, in Europe 100 Hz spacing.

73, Reino oh3mA

 

 

Doesn't explain when the difference isn't 120 or 100 Hz.

 

Phil GM3ZZA

From: WSJTX@groups.io [mailto:WSJTX@groups.io] On Behalf Of Bill Lederer
Sent: 22. helmikuuta 2020 20:30
To: WSJTX@groups.io
Subject: [WSJTX] Interesting double-decode

 

Folks:

Here is something I have seen a couple of times.  A station is decoded on two frequencies at the same time, at vastly different db levels.  Here are two  examples from one station
today:

200222_181830     7.074 Rx FT8     -8  0.1 1356 CQ K5DDE EN90
200222_181830     7.074 Tx FT8      0  0.0  650 CQ W8LVN EN62                       
200222_181830     7.074 Rx FT8     14  0.1 1476 CQ K5DDE EN90


200222_181900     7.074 Rx FT8     -8  0.1 1355 W8LVN K5DDE EN90                      a2
200222_181900     7.074 Rx FT8     17  0.1 1476 W8LVN K5DDE EN90

On the order of 25 db difference, 120 hz difference.

Curious what this might be caused by.

Thanks,
w8lvn

 


--
73 Phil GM3ZZA



--
73 Phil GM3ZZA
Martin G0HDB <marting0hdb@...>
 

On Sat, Feb 22, 2020 at 07:57 PM, Philip Rose wrote:

So, I see the same signal being decoded several 100 Hz apart on 20m, is this a Doppler effect from what is probably the busiest air route in Europe?
 
I still contend it's over-driving of transmitters. I keep my ALC less than 0dB on datamodes.
The spacing of a Doppler-shifted signal on 20m is, in my opinion, extremely unlikely to ever reach several 100Hz; the plane causing the Doppler would probably need to be moving at hypersonic speeds!  I have to confess I've never seen multiples of the same signal (other than those using FT8 Fox & Hounds mode or the MSHV software) with spacings of the several 100Hz you mention.

In my experience, over-driving a transmitter's audio input usually results in harmonics of the eight FT8 tones that appear as one or more 'bar-code' signals elsewhere across the audio spectrum.  Sadly this is very common.

As for ALC, I believe FT8 uses a constant-amplitude waveform so I'm not sure what effect ALC might have on that; I wouldn't expect running with a degree of ALC to have any adverse effect on the purity of the transmitted signal.  However, I recall there have been many previous discussions on this so I probably need to refresh my understanding of this aspect!  Does ALC have any adverse effect on the quality of an RTTY signal, which is in effect a 2-tone FSK signal rather than the 8-tone (G)FSK signal used by FT8?

--
Martin G0HDB
Reino Talarmo
 

No it is an AM due to full wave rectification, poor DC filtering and overdriving power amplifier, at least in most cases.

73, Reino oh3mA

 

From: WSJTX@groups.io [mailto:WSJTX@groups.io] On Behalf Of Philip Rose via Groups.Io
Sent: 22. helmikuuta 2020 21:57
To: WSJTX@groups.io
Subject: Re: [WSJTX] Interesting double-decode

 

 

 

On 22 Feb 2020 19:43, Martin G0HDB <marting0hdb@...> wrote:

On Sat, Feb 22, 2020 at 07:30 PM, Philip Rose wrote:

Doesn't explain when the difference isn't 120 or 100 Hz.

It's unusual but not impossible to see Doppler shifting aka aircraft scatter of signals on especially the higher HF bands; in such cases the spacing of the secondary signal will almost certainly not be a multiple of the AC line frequency.  The effect is much more prominent, and more easily and more often observed, on 6m and higher.

--
Martin G0HDB

So, I see the same signal being decoded several 100 Hz apart on 20m, is this a Doppler effect from what is probably the busiest air route in Europe?

 

I still contend it's over-driving of transmitters. I keep my ALC less than 0dB on datamodes.

 

72 Phil GM3ZZA

 


--
73 Phil GM3ZZA
Philip Rose
 



On 22 Feb 2020 19:43, Martin G0HDB <marting0hdb@...> wrote:
On Sat, Feb 22, 2020 at 07:30 PM, Philip Rose wrote:

Doesn't explain when the difference isn't 120 or 100 Hz.
It's unusual but not impossible to see Doppler shifting aka aircraft scatter of signals on especially the higher HF bands; in such cases the spacing of the secondary signal will almost certainly not be a multiple of the AC line frequency.  The effect is much more prominent, and more easily and more often observed, on 6m and higher.

--
Martin G0HDB
So, I see the same signal being decoded several 100 Hz apart on 20m, is this a Doppler effect from what is probably the busiest air route in Europe?

I still contend it's over-driving of transmitters. I keep my ALC less than 0dB on datamodes.

72 Phil GM3ZZA


--
73 Phil GM3ZZA
Reino Talarmo
 

Hi Phil,

Of course it does not. There are multiple reasons for additional sidebands such as Doppler shift due to airplane reflections, if also the direct signal is available (multipath). Then normally difference frequency is changing.

By the way this is simple to simulate and results are close even on the relative power difference.
If there is a switching power supply, it may still pass 120 Hz spikes through and narrow spikes do give a closer amplitude distribution in the case of multiple sidebands.

Power supplies of tube amplifiers are seldom switching supplies and if rectifying diodes are not properly bypassed by a snubber, just a capacitor is fine, then those recovery peaks can propagate into anode voltage as well.

73, Reino oh3mA

 

From: WSJTX@groups.io [mailto:WSJTX@groups.io] On Behalf Of Philip Rose via Groups.Io
Sent: 22. helmikuuta 2020 21:31
To: WSJTX@groups.io
Subject: Re: [WSJTX] Interesting double-decode

 

 

 

On 22 Feb 2020 19:21, Reino Talarmo <reino.talarmo@...> wrote:

Hi,

That is normally generated by a power amplifier with less than perfect battery voltage regulation so that a full wave rectification leaves a 120 Hz sawtooth wave riding on top of the battery voltage. If the power amplifier is driven full, then result is 120 Hz sawtooth wave amplitude modulation. Sometimes there are multiple sidebands with 120 Hz spacing, in Europe 100 Hz spacing.

73, Reino oh3mA

 

 

Doesn't explain when the difference isn't 120 or 100 Hz.

 

Phil GM3ZZA

From: WSJTX@groups.io [mailto:WSJTX@groups.io] On Behalf Of Bill Lederer
Sent: 22. helmikuuta 2020 20:30
To: WSJTX@groups.io
Subject: [WSJTX] Interesting double-decode

 

Folks:

Here is something I have seen a couple of times.  A station is decoded on two frequencies at the same time, at vastly different db levels.  Here are two  examples from one station
today:

200222_181830     7.074 Rx FT8     -8  0.1 1356 CQ K5DDE EN90
200222_181830     7.074 Tx FT8      0  0.0  650 CQ W8LVN EN62                       
200222_181830     7.074 Rx FT8     14  0.1 1476 CQ K5DDE EN90


200222_181900     7.074 Rx FT8     -8  0.1 1355 W8LVN K5DDE EN90                      a2
200222_181900     7.074 Rx FT8     17  0.1 1476 W8LVN K5DDE EN90

On the order of 25 db difference, 120 hz difference.

Curious what this might be caused by.

Thanks,
w8lvn

 


--
73 Phil GM3ZZA
Martin G0HDB <marting0hdb@...>
 

On Sat, Feb 22, 2020 at 07:48 PM, Bill Lederer wrote:
So that would be multipath then likely?
Yes, it's multipath but with the indirect or secondary path frequency-shifted because of the Doppler effect resulting from the movement of the aircraft.  It's the same effect that's used in bistatic radar systems.

--
Martin G0HDB
Philip Rose
 

Jim, 
I didn't say that, I quoted someone saying that. Hence the > leading the line.

And I agree with you it might be as much a problem in the receiver path as the transmitter path, the fact that only one of several strong signals suffer this indicates it's probably the transmitter path.

73 Phil GM3ZZA

On 22 Feb 2020 19:20, Jim Brown <k9yc@...> wrote:

On 2/22/2020 11:08 AM, Philip Rose via Groups.Io wrote:
> Here is something I have seen a couple of times.  A station is decoded
> on two frequencies at the same time, at vastly different db levels. Here
> are two  examples from one station

This can be caused by overload in the receive audio path between radio
and computer sound card, and by overdrive in the station being
double-decoded. If the difference is 120 Hz, it might be caused by power
supply currents injected into the audio path, usually via the signal
return (cable shield) when station equipment is not properly bonded.
Double decodes like this can also be caused by Doppler, common for
reflections from aircraft on higher bands (we see a lot on 6M).

73, Jim K9YC




--
73 Phil GM3ZZA
Bill Lederer
 

So that would be multipath then likely?

On Sat, Feb 22, 2020 at 1:43 PM Martin G0HDB <marting0hdb@...> wrote:
On Sat, Feb 22, 2020 at 07:30 PM, Philip Rose wrote:

Doesn't explain when the difference isn't 120 or 100 Hz.
It's unusual but not impossible to see Doppler shifting aka aircraft scatter of signals on especially the higher HF bands; in such cases the spacing of the secondary signal will almost certainly not be a multiple of the AC line frequency.  The effect is much more prominent, and more easily and more often observed, on 6m and higher.

--
Martin G0HDB


--
--w8lvn--
Martin G0HDB <marting0hdb@...>
 

On Sat, Feb 22, 2020 at 07:30 PM, Philip Rose wrote:

Doesn't explain when the difference isn't 120 or 100 Hz.
It's unusual but not impossible to see Doppler shifting aka aircraft scatter of signals on especially the higher HF bands; in such cases the spacing of the secondary signal will almost certainly not be a multiple of the AC line frequency.  The effect is much more prominent, and more easily and more often observed, on 6m and higher.

--
Martin G0HDB
Bill Lederer
 

So i am under the impression that switching mode power supplies don't operate at mains frequency, or at least their switching will spit off an array of frequencies due to the switching. AC Mains I can understand, but I think a switching mode power supply will emit some range of frequencies that may not be related to the mains frequency.

w8lvn


On Sat, Feb 22, 2020 at 1:37 PM Martin G0HDB <marting0hdb@...> wrote:
Nope, it's AC mains-related.  If there's any significant 50Hz or 60Hz ripple coming out of the PSU (especially an SMPSU) on the DC line that's powering the rig, or if there's anything else in the shack that might be injecting a ripple signal somewhere in the audio signal chain, then this will be superimposed on the transmitted audio and will appear as a decodable 'sideband', in the case of the K5DDE decodes the spacing of the spurious signal is 2 x the AC line frequency.

It's not uncommon to see more than two spurii, I think I've seen up to four, all with spacings of exact multiples of the local AC mains frequency, on the signals emanated by some Eastern EU stations...!  I can even see a signal 100Hz away from a nearby amateur's very strong FT8 signal; even though his unintentional sideband is approaching 40dB down on his fundamental signal I would prefer it if it weren't there at all.

--
Martin G0HDB


--
--w8lvn--

1 - 20 of 28