Topics

locked sticky Understanding USB and PC noise

robertcharris@...
 

USB cables radiate considerable noise greatly raising the background noise floor  gawd awful to say the least.  EBAY is one source having listings of triple shielded USB to Micro cables, some as short at 6 inches.  What a difference a cable makes!  Background noise radiated from the USB cable is practically non existent.  Now the downside:  A simple long wire antenna readily picks up considerable noise, even from the short triple shielded USB cable in close proximity to the USB cable.  I've enjoyed great success drastically (understatement)  reducing USP cable noise feeding the Airspy HF+ with a Mini Whip placed at some distance away from the receiver and laptop computer feeding the Airspy HF+.  To put this is perspective---the original unshielded cable radiates noise so severely, that the considerable noise is transmitted to the mini whip placed fifty feet away.  By the way, ferrite "clampons" have proven useless in reducing USB cable noise radiation. Last, but not least, ditch the switching power adapters for powering the laptop.  These things are a disaster in terms of noise.  Go battery power and a laptop computer if possible. Another suggestion, turn off the AGC and add reduce gain on the slider control.  There is a point at which "spurs" across the visible spectrum suddenly disappear if IMD (false signals) is taking place. (Eliminate the noise, and what a pleasure it is listening around to high quality signal content!

markobyrnes@...
 

I used a ft140-31 ferrite ont he USB and one on the Coax, amazing difference. 

The Macbook touchpad was also a severe source of noise all drastically eliminated with the ferrites. 

Roberto Zinelli
 

Yes, computers generate a lot noise! Should be nice have the possibility to switch ACG off and gain control also on AIRSPY-HF  controlled by SPYSERVER.

73 IW4ENS
Roberto

Il giorno 15 apr 2018, alle ore 17:25, robertcharris@... ha scritto:

USB cables radiate considerable noise greatly raising the background noise floor  gawd awful to say the least.  EBAY is one source having listings of triple shielded USB to Micro cables, some as short at 6 inches.  What a difference a cable makes!  Background noise radiated from the USB cable is practically non existent.  Now the downside:  A simple long wire antenna readily picks up considerable noise, even from the short triple shielded USB cable in close proximity to the USB cable.  I've enjoyed great success drastically (understatement)  reducing USP cable noise feeding the Airspy HF+ with a Mini Whip placed at some distance away from the receiver and laptop computer feeding the Airspy HF+.  To put this is perspective---the original unshielded cable radiates noise so severely, that the considerable noise is transmitted to the mini whip placed fifty feet away.  By the way, ferrite "clampons" have proven useless in reducing USB cable noise radiation. Last, but not least, ditch the switching power adapters for powering the laptop.  These things are a disaster in terms of noise.  Go battery power and a laptop computer if possible. Another suggestion, turn off the AGC and add reduce gain on the slider control.  There is a point at which "spurs" across the visible spectrum suddenly disappear if IMD (false signals) is taking place. (Eliminate the noise, and what a pleasure it is listening around to high quality signal content!

robertcharris@...
 

It is great that you touched on using ferrites on the antenna coax cable as being effective.  Absolutely, computer noise (common mode) will travel the coax shield to the antenna. Another fix, in many situations, is to wind a simple toroid ferrite in transformer fashion to magnetically couple antenna signal to the coax to the antenna. There is quite a bit of information on doing so on YouTube, as one information source & plenty of others at our fingertips using search engines. Unfortunately, USB communications noise, between the Airspy and computer (differential), are not effectively dealt with with ferrites, at least in my experience.  For a few bucks, a short triple shielded USB is worth a try. Every set up and circumstances are different. Sometimes one solution will work, more aggravating situations needing more horse power and multiple remedies. Here is a tip:  Using a mini whip antenna attached to twenty, or thirty feet of coax,  wrap aluminum foil around the Mini Whip.  That should kill all signals and background noise.  If noise is hitching a ride from the computer via coax shield, you will find out real fast.

prog
 

On Mon, Apr 16, 2018 at 07:41 am, <robertcharris@...> wrote:
It is great that you touched on using ferrites on the antenna coax cable as being effective.
The RF shielding of the HF+ delegates all the RF hygiene to the end user and doesn't contribute to it.
Don't connect active E-field antennas that use the coax as ground to a source of noise like a PC, and if you do so, make sure you have a very high impedance between the PC's ground and your coax at the frequency of interest. Ferrites are good down to a certain frequency, but for the lower bands (LW, VLF) you will need something more radical like galvanic isolators.

markobyrnes@...
 
Edited

How about Galvanic isolator ? https://www.bonito.net/newsroom/the-galvanic-isolator-gi300/?lang=en

I have Bonito MA205 active antenna and I have it grounded and don't have any issues. I also use H155 coax which helps.

The greatest issues people will face is reducing local QRM from electronic devices.

Bryon NF6M
 

I wonder, as far as RFI coming in on the USB cable itself, would it help to add some RF bypass capacitors from the USB data/power lines to ground within the HF+? 

It seems like that would essentially short any stray RF to ground at the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?

Dana Myers
 

On 4/16/2018 4:17 PM, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it help to add some RF bypass capacitors from the USB data/power lines to ground within the HF+?

It seems like that would essentially short any stray RF to ground at the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?
USB signaling rates are, in fact, RF rates. Any that effectively suppresses RF
will also suppress USB.

Cheers,
Dana  K6JQ

jdow
 

You have not been listening. "Ground is not ground." That is one of Joanne's Laws. It simply means you are deluded if you think the ground over there is the same as the ground over here even if the distance between here and there is in fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be bypassed to ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it help to add some RF bypass capacitors from the USB data/power lines to ground within the HF+?
It seems like that would essentially short any stray RF to ground at the connector, while allowing DC signals and power to remain unaffected.
Or would that distort the rise/fall of DC transitions on the data lines?

Joe M.
 

Is there a list of "Joanne's Laws"?

Joe M.

On 4/16/2018 8:51 PM, jdow wrote:
You have not been listening. "Ground is not ground." That is one of
Joanne's Laws. It simply means you are deluded if you think the ground
over there is the same as the ground over here even if the distance
between here and there is in fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be
bypassed to ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it
help to add some RF bypass capacitors from the USB data/power lines to
ground within the HF+?

It seems like that would essentially short any stray RF to ground at
the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?


---
This email has been checked for viruses by AVG.
http://www.avg.com

jdow
 

Not really. It often takes events happening to remind me. The two which come to mind are "Ground isn't" and "Oscillators don't". At least reasonably linear oscillators do not oscillate. Visit your mental way back machine and remember the old regenerative receiver, "an almost oscillator". If you adjust the regeneration control a little too far it oscillates. What actually happens is as you get more and more regeneration cranked in the gain goes skywards and the bandwidth gets very narrow. So you get more noise in a narrower bandwidth as the regeneration goes up. Eventually the noise looks like a "pure" signal and the level in the amplifier is sufficient to reduce it's gain enough that it does not get more regenerative. This intuition led me to deriving some very good oscillator noise spectra "estimates." It also explained why SAW, Surface Acoustic Wave, delay line oscillators have funky looking noise spectra way out from the carrier at about 1/delay It reaches a baseline. Goes below the baseline then goes back up to the baseline and above it a little then back down looking vaguely like a sine(x)/x function raised above zero. It all related to the phase of the noise feedback in the SAW delay line very very close to but not really oscillator. That one is probably the most out of the box idea I had considering the state of the art at the time.

{^_-}

On 20180416 18:18, Joe M. wrote:
Is there a list of "Joanne's Laws"?
Joe M.
On 4/16/2018 8:51 PM, jdow wrote:
You have not been listening. "Ground is not ground." That is one of
Joanne's Laws. It simply means you are deluded if you think the ground
over there is the same as the ground over here even if the distance
between here and there is in fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be
bypassed to ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it
help to add some RF bypass capacitors from the USB data/power lines to
ground within the HF+?

It seems like that would essentially short any stray RF to ground at
the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?



---
This email has been checked for viruses by AVG.
http://www.avg.com

jdow
 

Oh, another is "Conductors radiate". There is no conductor possible that does not radiate with the possible exception of the center conductor of a piece of superconducting coax. Even dummy loads radiate. (They can reach the LA area from the Seattle area, as a matter of fact, if you feed them enough power. "Gee, your signal's weak tonight!" was the consensus of the group nattering with Art Bell on ham radio after his show one evening. That was a fun group while it lasted. Art and his humongous loop antenna were a long time source of discussion on that on the air gathering of hams. It certainly was not a net or a formalized rag chew. It was fun shooting the breeze more and less technically until some bozos destroyed it by jamming us. And the FCC STILL has not removed the creeps from the air.)

{^_^}

On 20180416 18:18, Joe M. wrote:
Is there a list of "Joanne's Laws"?
Joe M.
On 4/16/2018 8:51 PM, jdow wrote:
You have not been listening. "Ground is not ground." That is one of
Joanne's Laws. It simply means you are deluded if you think the ground
over there is the same as the ground over here even if the distance
between here and there is in fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be
bypassed to ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it
help to add some RF bypass capacitors from the USB data/power lines to
ground within the HF+?

It seems like that would essentially short any stray RF to ground at
the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?



---
This email has been checked for viruses by AVG.
http://www.avg.com

Patrick
 

Hi there !

Assuming one use quality USB cables and / or ferrites, is it necessary to keep the HF+ far from the computer or does it make no difference if they are side by side ?

Thanks
Patrick

2018-04-17 4:32 GMT+02:00 jdow <jdow@...>:

Oh, another is "Conductors radiate". There is no conductor possible that does not radiate with the possible exception of the center conductor of a piece of superconducting coax. Even dummy loads radiate. (They can reach the LA area from the Seattle area, as a matter of fact, if you feed them enough power. "Gee, your signal's weak tonight!" was the consensus of the group nattering with Art Bell on ham radio after his show one evening. That was a fun group while it lasted. Art and his humongous loop antenna were a long time source of discussion on that on the air gathering of hams. It certainly was not a net or a formalized rag chew. It was fun shooting the breeze more and less technically until some bozos destroyed it by jamming us. And the FCC STILL has not removed the creeps from the air.)

{^_^}

On 20180416 18:18, Joe M. wrote:
Is there a list of "Joanne's Laws"?

Joe M.

On 4/16/2018 8:51 PM, jdow wrote:
You have not been listening. "Ground is not ground." That is one of
Joanne's Laws. It simply means you are deluded if you think the ground
over there is the same as the ground over here even if the distance
between here and there is in fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be
bypassed to ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it
help to add some RF bypass capacitors from the USB data/power lines to
ground within the HF+?

It seems like that would essentially short any stray RF to ground at
the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?





---
This email has been checked for viruses by AVG.
http://www.avg.com








jdow
 

Normally speaking the distance from a noise source reduces signal by the square of the distance. For the near field, next to your computer, for example, signal strength reduces with the cube of the distance. You need a well shielded computer - in and out. Ferrites on the power cord should not be ruled out. Ferrites on all audio and video cables should not be ruled out. Ferrites on any other I/O cables should not be ruled out. These may include pairs of ferrite beads, one big one with several wraps of the cable through a 3/8" hole and one smaller that works well for UHF and above. The idea is to prevent radiation from escaping the computer, monitor, or other peripherals. Make the cables high impedances for RF. Then at the antenna make sure it is "balanced" even if you use coax. A set of ferrite beads or transformer and beads can keep the cable's common mode impedance high to reduce pickup. But, hey, why take just my word for it. Visit Jim Browns excellent write-ups. Jim, K9YC, fought some REALLY high EMI environments as a professional audio engineer in a building in Chicago with several very high power transmitters on top. I'd say he figured out a thing or two worthy of attention, "http://audiosystemsgroup.com/RFI-Ham.pdf". Good judicious grounding and good EMI isolation can make your life a lot better with regards to sources you control. There is not much people can do about a neighbor's plasma TV or the like unless they are on (very) good terms with their neighbor. Sometimes you can get the FCC in the US to act.

{^_^}

On 20180416 23:43, Patrick wrote:
Hi there !
Assuming one use quality USB cables and / or ferrites, is it necessary to keep the HF+ far from the computer or does it make no difference if they are side by side ?
Thanks
Patrick
2018-04-17 4:32 GMT+02:00 jdow <jdow@... <mailto:jdow@...>>:
Oh, another is "Conductors radiate". There is no conductor possible that
does not radiate with the possible exception of the center conductor of a
piece of superconducting coax. Even dummy loads radiate. (They can reach the
LA area from the Seattle area, as a matter of fact, if you feed them enough
power. "Gee, your signal's weak tonight!" was the consensus of the group
nattering with Art Bell on ham radio after his show one evening. That was a
fun group while it lasted. Art and his humongous loop antenna were a long
time source of discussion on that on the air gathering of hams. It certainly
was not a net or a formalized rag chew. It was fun shooting the breeze more
and less technically until some bozos destroyed it by jamming us. And the
FCC STILL has not removed the creeps from the air.)
{^_^}
On 20180416 18:18, Joe M. wrote:
Is there a list of "Joanne's Laws"?
Joe M.
On 4/16/2018 8:51 PM, jdow wrote:
You have not been listening. "Ground is not ground." That is one of
Joanne's Laws. It simply means you are deluded if you think the ground
over there is the same as the ground over here even if the distance
between here and there is in fractions of an inch.
In a word, no. Ground isn't ground. And data does not like to be
bypassed to ground giving it round edges and reduced amplitude.
{^_^}
On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it
help to add some RF bypass capacitors from the USB data/power
lines to
ground within the HF+?
It seems like that would essentially short any stray RF to ground at
the connector, while allowing DC signals and power to remain
unaffected.
Or would that distort the rise/fall of DC transitions on the
data lines?
---
This email has been checked for viruses by AVG.
http://www.avg.com

Patrick
 

Thanks for these precious tips.
I have hundreds of spikes on VHF/FM I don't manage to suppress ... I added several ferrites, purchased a quality USB cable, put the HF+ "away" from the desktop (2m), no improvement at all. I suddendly realize that I did not add a ferrite clamp on the monitor cable (a jack connector for audio), not sure it could help but I'll give it a try.

I also noticed that the number of spikes increases / reduces according the the yagi azimut. Maybe the problem comes from the outside only ... (I live in a buidling)

Have a nice day,
Patrick



2018-04-17 9:25 GMT+02:00 jdow <jdow@...>:

Normally speaking the distance from a noise source reduces signal by the square of the distance. For the near field, next to your computer, for example, signal strength reduces with the cube of the distance. You need a well shielded computer - in and out. Ferrites on the power cord should not be ruled out. Ferrites on all audio and video cables should not be ruled out. Ferrites on any other I/O cables should not be ruled out. These may include pairs of ferrite beads, one big one with several wraps of the cable through a 3/8" hole and one smaller that works well for UHF and above. The idea is to prevent radiation from escaping the computer, monitor, or other peripherals. Make the cables high impedances for RF. Then at the antenna make sure it is "balanced" even if you use coax. A set of ferrite beads or transformer and beads can keep the cable's common mode impedance high to reduce pickup. But, hey, why take just my word for it. Visit Jim Browns excellent write-ups. Jim, K9YC, fought some REALLY high EMI environments as a professional audio engineer in a building in Chicago with several very high power transmitters on top. I'd say he figured out a thing or two worthy of attention, "http://audiosystemsgroup.com/RFI-Ham.pdf". Good judicious grounding and good EMI isolation can make your life a lot better with regards to sources you control. There is not much people can do about a neighbor's plasma TV or the like unless they are on (very) good terms with their neighbor. Sometimes you can get the FCC in the US to act.

{^_^}

On 20180416 23:43, Patrick wrote:
Hi there !

Assuming one use quality USB cables and / or ferrites, is it necessary to keep the HF+ far from the computer or does it make no difference if they are side by side ?

Thanks
Patrick

2018-04-17 4:32 GMT+02:00 jdow <jdow@... <mailto:jdow@...>>:


    Oh, another is "Conductors radiate". There is no conductor possible that
    does not radiate with the possible exception of the center conductor of a
    piece of superconducting coax. Even dummy loads radiate. (They can reach the
    LA area from the Seattle area, as a matter of fact, if you feed them enough
    power. "Gee, your signal's weak tonight!" was the consensus of the group
    nattering with Art Bell on ham radio after his show one evening. That was a
    fun group while it lasted. Art and his humongous loop antenna were a long
    time source of discussion on that on the air gathering of hams. It certainly
    was not a net or a formalized rag chew. It was fun shooting the breeze more
    and less technically until some bozos destroyed it by jamming us. And the
    FCC STILL has not removed the creeps from the air.)

    {^_^}

    On 20180416 18:18, Joe M. wrote:

        Is there a list of "Joanne's Laws"?

        Joe M.

        On 4/16/2018 8:51 PM, jdow wrote:

            You have not been listening. "Ground is not ground." That is one of
            Joanne's Laws. It simply means you are deluded if you think the ground
            over there is the same as the ground over here even if the distance
            between here and there is in fractions of an inch.

            In a word, no. Ground isn't ground. And data does not like to be
            bypassed to ground giving it round edges and reduced amplitude.

            {^_^}

            On 20180416 16:17, Bryon NF6M wrote:

                I wonder, as far as RFI coming in on the USB cable itself, would it
                help to add some RF bypass capacitors from the USB data/power
                lines to
                ground within the HF+?

                It seems like that would essentially short any stray RF to ground at
                the connector, while allowing DC signals and power to remain
                unaffected.

                Or would that distort the rise/fall of DC transitions on the
                data lines?





            ---
            This email has been checked for viruses by AVG.
            http://www.avg.com













jdow
 

I have a very large number of spikes. Many are from the APC Matrix 5000 UPSs we have to ride out brownouts. Others trace from the 100 kW (peak) solar array we have. Other noise is telco DSL leakage and powerline insulator leakage. Making it quiet around here will take a lot of work. Distance isolation will help. But that will lead to a lot of transmission line. Win some, lose some. In an urban or very dense suburban environment "Noise is." In fact a LOT of "noise is."

{O.O} I wanna cry.

On 20180417 00:53, Patrick wrote:
Thanks for these precious tips.
I have hundreds of spikes on VHF/FM I don't manage to suppress ... I added several ferrites, purchased a quality USB cable, put the HF+ "away" from the desktop (2m), no improvement at all. I suddendly realize that I did not add a ferrite clamp on the monitor cable (a jack connector for audio), not sure it could help but I'll give it a try.
I also noticed that the number of spikes increases / reduces according the the yagi azimut. Maybe the problem comes from the outside only ... (I live in a buidling)
Have a nice day,
Patrick

2018-04-17 9:25 GMT+02:00 jdow <jdow@... <mailto:jdow@...>>:
Normally speaking the distance from a noise source reduces signal by the
square of the distance. For the near field, next to your computer, for
example, signal strength reduces with the cube of the distance. You need a
well shielded computer - in and out. Ferrites on the power cord should not
be ruled out. Ferrites on all audio and video cables should not be ruled
out. Ferrites on any other I/O cables should not be ruled out. These may
include pairs of ferrite beads, one big one with several wraps of the cable
through a 3/8" hole and one smaller that works well for UHF and above. The
idea is to prevent radiation from escaping the computer, monitor, or other
peripherals. Make the cables high impedances for RF. Then at the antenna
make sure it is "balanced" even if you use coax. A set of ferrite beads or
transformer and beads can keep the cable's common mode impedance high to
reduce pickup. But, hey, why take just my word for it. Visit Jim Browns
excellent write-ups. Jim, K9YC, fought some REALLY high EMI environments as
a professional audio engineer in a building in Chicago with several very
high power transmitters on top. I'd say he figured out a thing or two worthy
of attention, "http://audiosystemsgroup.com/RFI-Ham.pdf
<http://audiosystemsgroup.com/RFI-Ham.pdf>". Good judicious grounding and
good EMI isolation can make your life a lot better with regards to sources
you control. There is not much people can do about a neighbor's plasma TV or
the like unless they are on (very) good terms with their neighbor. Sometimes
you can get the FCC in the US to act.
{^_^}
On 20180416 23:43, Patrick wrote:
Hi there !
Assuming one use quality USB cables and / or ferrites, is it necessary
to keep the HF+ far from the computer or does it make no difference if
they are side by side ?
Thanks
Patrick
2018-04-17 4:32 GMT+02:00 jdow <jdow@...
<mailto:jdow@...> <mailto:jdow@...
<mailto:jdow@...>>>:
    Oh, another is "Conductors radiate". There is no conductor possible
that
    does not radiate with the possible exception of the center
conductor of a
    piece of superconducting coax. Even dummy loads radiate. (They can
reach the
    LA area from the Seattle area, as a matter of fact, if you feed
them enough
    power. "Gee, your signal's weak tonight!" was the consensus of the
group
    nattering with Art Bell on ham radio after his show one evening.
That was a
    fun group while it lasted. Art and his humongous loop antenna were
a long
    time source of discussion on that on the air gathering of hams. It
certainly
    was not a net or a formalized rag chew. It was fun shooting the
breeze more
    and less technically until some bozos destroyed it by jamming us.
And the
    FCC STILL has not removed the creeps from the air.)
    {^_^}
    On 20180416 18:18, Joe M. wrote:
        Is there a list of "Joanne's Laws"?
        Joe M.
        On 4/16/2018 8:51 PM, jdow wrote:
            You have not been listening. "Ground is not ground." That
is one of
            Joanne's Laws. It simply means you are deluded if you think
the ground
            over there is the same as the ground over here even if the
distance
            between here and there is in fractions of an inch.
            In a word, no. Ground isn't ground. And data does not like
to be
            bypassed to ground giving it round edges and reduced amplitude.
            {^_^}
            On 20180416 16:17, Bryon NF6M wrote:
                I wonder, as far as RFI coming in on the USB cable
itself, would it
                help to add some RF bypass capacitors from the USB
data/power
                lines to
                ground within the HF+?
                It seems like that would essentially short any stray RF
to ground at
                the connector, while allowing DC signals and power to
remain
                unaffected.
                Or would that distort the rise/fall of DC transitions
on the
                data lines?
            ---
            This email has been checked for viruses by AVG.
http://www.avg.com
FM RFI.JPG

smswk
 

Maybe i am wrong but:

i have never seen any discussions like that when talking
about any other software defined radio.

The number of spikes where they do not have to be
or the extreme efforts of USB filtering seem to be an exclusive HF+
problem.

The HF+ has a very good shielded case.
Terminate your antenna inputs and use the best USB cables you can get.
What will you see ?
Unwanted spikes and unwanted noise.

And you can go fürther with filtering. Using ferrit cores,
trying it with RC Circuits for theUSB shield, external linear PSU,
special D+ / D- Data Line Filters etc.

It is possible to reduce the problems above but they are still there.

Don`t get me wrong. I really like the HF+ and it is maybe the best and most sensitive
SDR you can get for just a few bucks.

When your antenna inputs are terminated and you add filters like you have never done
before in your life and the prolems are still there:
Stop to spend money on trying external improvements.
In my opinion prior to firmware upgrades almost every week a solution for this has to be found.

vy73`s
Stefan






-----Ursprüngliche Mitteilung-----
Von: jdow <jdow@...>
An: main <main@airspy.groups.io>
Verschickt: Di, 17. Apr 2018 10:28
Betreff: Re: [airspy] USB cable noise and Airspy HF+

I have a very large number of spikes. Many are from the APC Matrix 5000 UPSs we
have to ride out brownouts. Others trace from the 100 kW (peak) solar array we
have. Other noise is telco DSL leakage and powerline insulator leakage. Making
it quiet around here will take a lot of work. Distance isolation will help. But
that will lead to a lot of transmission line. Win some, lose some. In an urban
or very dense suburban environment "Noise is." In fact a LOT of "noise is."

{O.O} I wanna cry.

On 20180417 00:53, Patrick wrote:
Thanks for these precious tips.
I have hundreds of spikes on VHF/FM I don't manage to suppress ... I added
several ferrites, purchased a quality USB cable, put the HF+ "away" from the
desktop (2m), no improvement at all. I suddendly realize that I did not add a
ferrite clamp on the monitor cable (a jack connector for audio), not sure it
could help but I'll give it a try.

I also noticed that the number of spikes increases / reduces according the the
yagi azimut. Maybe the problem comes from the outside only ... (I live in a
buidling)

Have a nice day,
Patrick



2018-04-17 9:25 GMT+02:00 jdow <jdow@... <mailto:jdow@...>>:

Normally speaking the distance from a noise source reduces signal by the
square of the distance. For the near field, next to your computer, for
example, signal strength reduces with the cube of the distance. You need a
well shielded computer - in and out. Ferrites on the power cord should not
be ruled out. Ferrites on all audio and video cables should not be ruled
out. Ferrites on any other I/O cables should not be ruled out. These may
include pairs of ferrite beads, one big one with several wraps of the cable
through a 3/8" hole and one smaller that works well for UHF and above. The
idea is to prevent radiation from escaping the computer, monitor, or other
peripherals. Make the cables high impedances for RF. Then at the antenna
make sure it is "balanced" even if you use coax. A set of ferrite beads or
transformer and beads can keep the cable's common mode impedance high to
reduce pickup. But, hey, why take just my word for it. Visit Jim Browns
excellent write-ups. Jim, K9YC, fought some REALLY high EMI environments as
a professional audio engineer in a building in Chicago with several very
high power transmitters on top. I'd say he figured out a thing or two worthy
of attention, "http://audiosystemsgroup.com/RFI-Ham.pdf
<http://audiosystemsgroup.com/RFI-Ham.pdf>". Good judicious grounding and
good EMI isolation can make your life a lot better with regards to sources
you control. There is not much people can do about a neighbor's plasma TV or
the like unless they are on (very) good terms with their neighbor. Sometimes
you can get the FCC in the US to act.

{^_^}

On 20180416 23:43, Patrick wrote:

Hi there !

Assuming one use quality USB cables and / or ferrites, is it necessary
to keep the HF+ far from the computer or does it make no difference if
they are side by side ?

Thanks
Patrick

2018-04-17 4:32 GMT+02:00 jdow <jdow@...
<mailto:jdow@...> <mailto:jdow@...
<mailto:jdow@...>>>:


    Oh, another is "Conductors radiate". There is no conductor possible
that
    does not radiate with the possible exception of the center
conductor of a
    piece of superconducting coax. Even dummy loads radiate. (They can
reach the
    LA area from the Seattle area, as a matter of fact, if you feed
them enough
    power. "Gee, your signal's weak tonight!" was the consensus of the
group
    nattering with Art Bell on ham radio after his show one evening.
That was a
    fun group while it lasted. Art and his humongous loop antenna were
a long
    time source of discussion on that on the air gathering of hams. It
certainly
    was not a net or a formalized rag chew. It was fun shooting the
breeze more
    and less technically until some bozos destroyed it by jamming us.
And the
    FCC STILL has not removed the creeps from the air.)

    {^_^}

    On 20180416 18:18, Joe M. wrote:

        Is there a list of "Joanne's Laws"?

        Joe M.

        On 4/16/2018 8:51 PM, jdow wrote:

            You have not been listening. "Ground is not ground." That
is one of
            Joanne's Laws. It simply means you are deluded if you think
the ground
            over there is the same as the ground over here even if the
distance
            between here and there is in fractions of an inch.

            In a word, no. Ground isn't ground. And data does not like
to be
            bypassed to ground giving it round edges and reduced amplitude.

            {^_^}

            On 20180416 16:17, Bryon NF6M wrote:

                I wonder, as far as RFI coming in on the USB cable
itself, would it
                help to add some RF bypass capacitors from the USB
data/power
                lines to
                ground within the HF+?

                It seems like that would essentially short any stray RF
to ground at
                the connector, while allowing DC signals and power to
remain
                unaffected.

                Or would that distort the rise/fall of DC transitions
on the
                data lines?





            ---
            This email has been checked for viruses by AVG.
http://www.avg.com
















FM RFI.JPG

PMM
 

I would not say the HF+ is any worse than any other rtl type dongle for usb noise but the HF+ is more sensitive and a lower noise floor so more noise appears above the noise floor you just see more of what you don't see with an inferia rtl dongle.

Yes ferrites help, but a good cable has more impact but I try to keep away from ferrites on antenna cable.

My 1st line of defence is an AudioQuest Jitterbug ment for usb audio dacs. This places some filtering on the datelines and it is noticeable but sadly not the outer shield. Really a full galvanic isolate needed but high speed ones expensive.

2nd is a short but quality usb 6" extender cable with a ferrite on I know if directly plugged All good then any additional noise is the main cable or antenna I can weed out bad cables with this shorty.

Then the long cable to the HF plus with 2 ferrite each end.
If I swap that long cable for my other long cable eek😱 but if I plug into the 6" bit all is OK so that other cable very leaky.

Then I have a bonito GI300 galvanic isolator on the aerial end.

And all that makes good 😊

I had All that with airspy mini / rtl sdrs / e4000 but certainly the rtls/e4000 were limited in dynamic range/snr 

Project currently on hold due to work/lurgy/she that must be obeyed is my magloop and with HF+ it is a different league I see so much more with hf+ which the rtls and surely with upconvertor just fails to see on the same magloop and usb cables.

Hope that makes sense 😊




On Tue, 17 Apr 2018, 19:39 smswk via Groups.Io, <smswk=aol.com@groups.io> wrote:
Maybe i am wrong but:

i have never seen any discussions like that when talking
about any other software defined radio.

The number of spikes where they do not have to be
or the extreme efforts of USB filtering  seem to be an exclusive HF+
problem.

The HF+ has a very good shielded case.
Terminate your antenna inputs and use the best USB cables you can get.
What will you see ?
Unwanted spikes and unwanted noise.

And you can go fürther with filtering. Using ferrit cores,
trying it with RC Circuits for theUSB  shield, external linear PSU,
special D+ / D- Data Line Filters etc.

It is possible to reduce the problems above but they are still there.

Don`t get me wrong. I really like the HF+ and it is maybe the best and most sensitive
SDR you can get  for just a few bucks.

When your antenna inputs are terminated and you add filters like you have never done
before in your life and the prolems are still there:
Stop to spend money on trying external improvements.
In my opinion prior to firmware upgrades almost every week a solution for this has to be found.

vy73`s
Stefan






-----Ursprüngliche Mitteilung-----
Von: jdow <jdow@...>
An: main <main@airspy.groups.io>
Verschickt: Di, 17. Apr 2018 10:28
Betreff: Re: [airspy] USB cable noise and Airspy HF+

I have a very large number of spikes. Many are from the APC Matrix 5000 UPSs we
have to ride out brownouts. Others trace from the 100 kW (peak) solar array we
have. Other noise is telco DSL leakage and powerline insulator leakage. Making
it quiet around here will take a lot of work. Distance isolation will help. But
that will lead to a lot of transmission line. Win some, lose some. In an urban
or very dense suburban environment "Noise is." In fact a LOT of "noise is."

{O.O}  I wanna cry.

On 20180417 00:53, Patrick wrote:
> Thanks for these precious tips.
> I have hundreds of spikes on VHF/FM I don't manage to suppress ... I added
> several ferrites, purchased a quality USB cable, put the HF+ "away" from the
> desktop (2m), no improvement at all. I suddendly realize that I did not add a
> ferrite clamp on the monitor cable (a jack connector for audio), not sure it
> could help but I'll give it a try.
>
> I also noticed that the number of spikes increases / reduces according the the
> yagi azimut. Maybe the problem comes from the outside only ... (I live in a
> buidling)
>
> Have a nice day,
> Patrick
>
> ​
>
> 2018-04-17 9:25 GMT+02:00 jdow <jdow@... <mailto:jdow@...>>:
>
>     Normally speaking the distance from a noise source reduces signal by the
>     square of the distance. For the near field, next to your computer, for
>     example, signal strength reduces with the cube of the distance. You need a
>    well shielded computer - in and out. Ferrites on the power cord should not
>    be ruled out. Ferrites on all audio and video cables should not be ruled
>     out. Ferrites on any other I/O cables should not be ruled out. These may
>     include pairs of ferrite beads, one big one with several wraps of the cable
>   through a 3/8" hole and one smaller that works well for UHF and above. The
>    idea is to prevent radiation from escaping the computer, monitor, or other
>    peripherals. Make the cables high impedances for RF. Then at the antenna
>     make sure it is "balanced" even if you use coax. A set of ferrite beads or
>    transformer and beads can keep the cable's common mode impedance high to
>     reduce pickup. But, hey, why take just my word for it. Visit Jim Browns
>     excellent write-ups. Jim, K9YC, fought some REALLY high EMI environments as
>   a professional audio engineer in a building in Chicago with several very
>     high power transmitters on top. I'd say he figured out a thing or two worthy
>  of attention, "http://audiosystemsgroup.com/RFI-Ham.pdf
>     <http://audiosystemsgroup.com/RFI-Ham.pdf>". Good judicious grounding and
>     good EMI isolation can make your life a lot better with regards to sources
>    you control. There is not much people can do about a neighbor's plasma TV or
>  the like unless they are on (very) good terms with their neighbor. Sometimes
>  you can get the FCC in the US to act.
>
>     {^_^}
>
>     On 20180416 23:43, Patrick wrote:
>
>         Hi there !
>
>         Assuming one use quality USB cables and / or ferrites, is it necessary
>         to keep the HF+ far from the computer or does it make no difference if
>         they are side by side ?
>
>         Thanks
>         Patrick
>
>         2018-04-17 4:32 GMT+02:00 jdow <jdow@...
>         <mailto:jdow@...> <mailto:jdow@...
>         <mailto:jdow@...>>>:
>
>
>      Oh, another is "Conductors radiate". There is no conductor possible
>       that
>              does not radiate with the possible exception of the center
>         conductor of a
>              piece of superconducting coax. Even dummy loads radiate. (They can
>         reach the
>              LA area from the Seattle area, as a matter of fact, if you feed
>         them enough
>              power. "Gee, your signal's weak tonight!" was the consensus of the
>         group
>              nattering with Art Bell on ham radio after his show one evening.
>         That was a
>              fun group while it lasted. Art and his humongous loop antenna were
>         a long
>              time source of discussion on that on the air gathering of hams. It
>         certainly
>              was not a net or a formalized rag chew. It was fun shooting the
>         breeze more
>              and less technically until some bozos destroyed it by jamming us.
>         And the
>       FCC STILL has not removed the creeps from the air.)
>
>              {^_^}
>
>              On 20180416 18:18, Joe M. wrote:
>
>                  Is there a list of "Joanne's Laws"?
>
>                  Joe M.
>
>              On 4/16/2018 8:51 PM, jdow wrote:
>
>                      You have not been listening. "Ground is not ground." That
>         is one of
>                     Joanne's Laws. It simply means you are deluded if you think
>       the ground
>                      over there is the same as the ground over here even if the
>         distance
>                      between here and there is in fractions of an inch.
>
>                      In a word, no. Ground isn't ground. And data does not like
>         to be
>                      bypassed to ground giving it round edges and reduced amplitude.
>
>                   {^_^}
>
>                      On 20180416 16:17, Bryon NF6M wrote:
>
>                          I wonder, as far as RFI coming in on the USB cable
>         itself, would it
>                          help to add some RF bypass capacitors from the USB
>         data/power
>                          lines to
>                          ground within the HF+?
>
>                       It seems like that would essentially short any stray RF
>       to ground at
>                          the connector, while allowing DC signals and power to
>         remain
>                          unaffected.
>
>                          Or would that distort the rise/fall of DC transitions
>         on the
>                          data lines?
>
>
>
>
>
>                      ---
>                      This email has been checked for viruses by AVG.
>         http://www.avg.com
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> FM RFI.JPG
>
>







Leif Asbrink
 

Hi All,

There is a common misunderstanding about the concept of "ground."
My comment here is a bit out of topic, but maybe it could help
to promote understanding.

When we design a PCB we typically have one symbol, GND for
the two different types of ground.

1) Current dump.
2) Voltage reference.

The voltage reference should carry no currents. It should be routed
in a way that minimizes the area of the loop made up from
the signal trace and the voltage reference to minimize the
pick up of magnetic fields.

The current dump is typically as noisy as Vcc because they are
coupled by decoupling capacitors. It does not matter much if
the current dump GND has a significant noise voltage because
the nearby Vcc would have the same noise voltage.

For analog circuits, OP-amps, A/D converters etc it is trivial.
For digital circuits, frequency dividers etc. the current dump
of the chip has to be the same as the voltage reference. If you
think about it, all is trivial.

Designers who do not want to think about it use multilayer boards
with one layer for gnd and another layer for Vcc. That is usually
good enough and then one does not have to worry about the two
kinds of ground are different things.

How does this relate to "USB cable noise and Airspy HF+" ?

In a typical receive site there is nothing like GND=voltage reference.
When we connect a HF+ to a PC computer we would typically get
a substantial RF voltage on the metal enclosure of the HF+.
(Voltage referenced to an ideal ground point which would have
a zero voltage referenced to the far fields at the location.)

The voltage on the HF+ enclosure will cause a current on the
antenna cable. The source might be the USB, but more likely
the computer, its power supply or other switch power units.
The typical receive site has low energy lamps and other modern
things that generate fields that couple to the receive system.

Nothing of that should enter the HF+ directly, (try with a dummy
load,) but the current sent out on the screen towards the antenna
might get into the antenna and then appear in the HF+.

ISOLATE THE ANTENNA!!
It will takre a lot of time, but if things are unclear, it would
probably be helpful to look at these videos:

http://www.youtube.com/watch?v=ItLkn8r4s3E

http://www.youtube.com/watch?v=zsZTX7MQSGQ

http://www.youtube.com/watch?v=kgMbaJDFu9M

http://www.youtube.com/watch?v=C65u7Pmz7a0

73

Leif

You have not been listening. "Ground is not ground." That is one of Joanne's
Laws. It simply means you are deluded if you think the ground over there is the
same as the ground over here even if the distance between here and there is in
fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be bypassed to
ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it help to add
some RF bypass capacitors from the USB data/power lines to ground within the HF+?

It seems like that would essentially short any stray RF to ground at the
connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?

Leif Asbrink
 

Hi Joanne,

Some conductors do not radiate. I think.

I remember reading about a transmission line with
a single conductor surrounded by a thin dielectric.

At both ends a circular ground plane with a diameter of
a half or maybe a couple of wavelengths. Exremely low
losses as I recall. (dielectric and ohmic, not radiation
as I recall.)

There are more trivial exceptions. The coaxial cable
does not radiate, but also the twisted pair does not
radiate provided it is not run close to metal or dielectrics
that could disturb the balance.

There are many more. The three-phase cables that give us
50/60 Hz power do not radiate (when balanced) We can
use that technology on RF also...

73

Leif

Oh, another is "Conductors radiate". There is no conductor possible that does
not radiate with the possible exception of the center conductor of a piece of
superconducting coax. Even dummy loads radiate. (They can reach the LA area from
the Seattle area, as a matter of fact, if you feed them enough power. "Gee, your
signal's weak tonight!" was the consensus of the group nattering with Art Bell
on ham radio after his show one evening. That was a fun group while it lasted.
Art and his humongous loop antenna were a long time source of discussion on that
on the air gathering of hams. It certainly was not a net or a formalized rag
chew. It was fun shooting the breeze more and less technically until some bozos
destroyed it by jamming us. And the FCC STILL has not removed the creeps from
the air.)

{^_^}

On 20180416 18:18, Joe M. wrote:
Is there a list of "Joanne's Laws"?

Joe M.

On 4/16/2018 8:51 PM, jdow wrote:
You have not been listening. "Ground is not ground." That is one of
Joanne's Laws. It simply means you are deluded if you think the ground
over there is the same as the ground over here even if the distance
between here and there is in fractions of an inch.

In a word, no. Ground isn't ground. And data does not like to be
bypassed to ground giving it round edges and reduced amplitude.

{^_^}

On 20180416 16:17, Bryon NF6M wrote:
I wonder, as far as RFI coming in on the USB cable itself, would it
help to add some RF bypass capacitors from the USB data/power lines to
ground within the HF+?

It seems like that would essentially short any stray RF to ground at
the connector, while allowing DC signals and power to remain unaffected.

Or would that distort the rise/fall of DC transitions on the data lines?



---
This email has been checked for viruses by AVG.
http://www.avg.com