Dealing with excessive noise - SOLVED: Ethernet over Power caused noise

Comments

• jks

  October 2017 edited October 2017

  Ugh, that's just about the worst noise I've ever seen.

  
  

  It might be ADSL or broadband-over-powerline (BPL). Note how the spectrum is obviously "notched" around the 30m, 20m, 17m and 15m ham bands (purple segments in band display). That's the service provider trying to keep hams happy, but failing dreadfully in this case. If you rotate the loop does the noise null at all?

  
  

  I think it might also be several independent noise sources. I've seen those strong 0 - 30 MHz lines that look like static crashes before. I think they come from seriously defective switch mode power supplies.
  

  
  

  
  

  

  
• roach

  October 2017

  Thanks for the quick reply. 

  
  

  Rotating the loop doesn't have any noticeable impact.

  
  

  An interesting point about the broadband over power. I do use Ethernet over power (The little D-link units), maybe that turns the whole house into a noise emitting crapfest.

  
  

  I'm going to have some free time next week so I think I'll put the SDR + antenna (plus all the other bits that would be needed) in the back of the car and go for a drive and see if I can get a nice clean waterfall. 

  If that works, I'll see if I can identify whats causing it locally. Thanks for the ideas. Hopefully I can get something that looks a bit nicer.

  
  
• jks

  October 2017

  Yes, definitely see if you can temporarily run without the Ethernet-over-power units. Going mobile is an interesting idea too.

  
  

  
  
• roach

  October 2017

  
  

  I ran a test with an without Ethernet over power and wow, you were on the money.

  
  

  Left is clearly without EOP. Additionally I was able to make the noise even worse by downloading a large file. 

  
• roach

  October 2017

  Sorry, attached image was not right

  

  
  
  Attachments:
  https://forum.kiwisdr.com/uploads/Uploader/b8/196e1242399f83da273946a6c1d45e.jpg
• G8JNJ

  October 2017

  Hi,

  
  

  Yes that's definitely broadband-over-powerline (BPL) interference, the notches in the noise spectrum on the amateur bands is the big clue.

  
  

  Unfortunately these sort of devices are the scourge of amateur radio operators in the UK, as even though in theory the frequencies used by amateur radio are notched out, the data carriers become regenerated due to inter-modulation produced by other items plugged into the mains supply in the house.

  
  

  They can also interfere with and reduce the speed of broadband internet services provided by VDSL over copper into the home.

  
  

  Worse still in the UK the main Direct to Home Satellite  TV operator has introduced a Set Top Box with a built in BLP modem which seems to be enabled by default. Just one of these devices can produce excessive levels of RF noise over a 0.5KM radius, making short wave reception in urban areas very difficult.

  
  

  John & Roach, with your permission I'd like to use the screen grabs of the BLP interference as part of some (non-commercial) presentation material relating to RF interference if that is OK with you both.

  
  

  Regards,

  
  

  Martin - G8JNJ
• sv1btl

  October 2017 edited October 2017

  Having a very similar setup (Wellbrook ALA1530LNP + KiwiSDR), I faced a "whole HF band" noise. suddenly, some weeks before. 

  I've changed the antenna with an inverted L one, and no noise has appeared. 

  After communicating with Andrew Ikin, the creator of Wellbrook antennas, and after making together some tests to the interface of the Wellbrook unit, he adviced me to change the coaxial as the final test, the final step before sending me a new Head Amp. of the loop. And, YES, there was a feeder failure, and everything went back to normal after changing the coaxial feeder to the loop with a new one and new connectors both to antenna and Kiwi.

  My KiwiSDR is streaming here: http://sv1btl.no-ip.org:8073/
  

  
  

  Screenshots with HFspan of Perseus SDR, using the same system antenna as Kiwi.

  
  

  
  
  Attachments:
  https://forum.kiwisdr.com/uploads/Uploader/5d/fc71ce2307075974fdb9ab435e8823.jpg
  https://forum.kiwisdr.com/uploads/Uploader/b5/14f5bdf550a4aa8d901dbc67ea894a.jpg
  https://forum.kiwisdr.com/uploads/Uploader/3f/a59ee1ed39baf178167b135ac8c0e0.jpg
• jks

  October 2017

  The noise floor is still too high I think, so maybe still go mobile to check. The noise floor should show a value of -90 to -110  on the spectrum scale on the right. Yours seems to show -60 to -70. But the value range depends on the gain of the loop amp.

  
  

  I'd like to see a screen cap where the "WF min" control is adjusted so the noise floor in the waterfall is a mostly dark blue color. That makes it easier I think to see signals versus noise floor. The waterfall & spectrum colors will change when you adjust WF min but the displayed spectrum values compared to the right scale should not. You seems to have a typical SWBC S/N in the spectrum display.

  
  
• jks

  October 2017

  Martin,

  
  

  That reminds me that I've been meaning to create a "gallery" of waterfall / spectrum screen caps of common interference sources. Lots of situations have been identified now: BPL, xDSL, electric fences, different SMPSs, the Ethernet problem, ADC overload, ...

  
  

  
• roach

  October 2017

  @G8JNJ - Feel free to use the images or generate new images from the SDR (while the EoP is still running)

  @sv1btl - what was the new cable you used? I've tried using the suggested type RG58C (specifically this one http://au.rs-online.com/web/p/coaxial-cable-assemblies/1683993/)

  @jks - you are right, it is still a bit high, I'll keep trying to improve it but removing the EoP is going to be number 1!
• sv1btl

  October 2017 edited October 2017

  roach  Nothing else but a new RG58 coaxial, with new BNC connectors and new BNC/SMA adaptor. Take care of the BNC to Head Amp of the loop, to be well waterproof. I guess what could happen is that the BNC at the loop end could
  have got damp with condensation at night then this becomes conductive from the
  screen to the inner and thus causes a noise voltage. Or sometimes you can get a
  conductive crud across the BNC if its a solder type. Its a good idea to use the PVC boot packed with Vaseline or DC4
  grease to keep out the damp.

  The following suggestions are made from Andrew Ikin, the creator of the Wellbrook antenna:

  The problem could be a poor amp. connection to the loop. Or the
  head amp. has failed or the Interface may have a fault.an you try the loop with another 12V power supply first. If the
  noise is still there then tighten the head amp. connection on the loop.

  If the fault is still there then test the Interface:

  Ensure that there is 11Volts being fed to
  the loop. Also make sure that the feeder cable is okay; no shorts or open
  circuits. If the above is okay then test the Antenna Interface.

  Connect a wire antenna to the Receiver and
  measure the gain on a local MW station. Then connect the Interface to the
  receiver and then connect the wire antenna to the Interface BNC via a 0.1uf
  cap. The gain should be 9dB higher.
  

  If
  the Interface is okay; it amplifiers and provides 11V to  the loop, then
  suspect the loop amplifier is faulty."
• G8JNJ

  October 2017 edited October 2017

  Hi Roach,

  
  

  I took a quick look at your KiWi SDR and something does not seem right.

  
  

  If you are using the Welbrook ALA 1530 I would have expected it to show much higher signal levels on the Medium and Long wave broadcast bands (despite the other noise).

  
  

  Edit - I just saw that you have added a filter for these bands - so that explains it.

  
  

  Where is the loop currently located ?

  
  

  Can you check to see what DC current the Wellbook antenna is drawing from your power supply ?

  
  

  I use satellite TV coax for most of may receive antennas, the type that has a foil braid with an outer wire mesh screen. This works very well in this application and is about the most cost effective coax cable you can use.

  
  

  Regards,

  
  

  Martin - G8JNJ
• roach

  October 2017

  @G8JNJ - the current draw from the power supply is 195mA (its fused at 315mA as suggested by Wellbrook) and at the antenna end its getting 11.6v (11.9 into the antenna interface). 

  
  

  The loop itself is sitting in the same room as a lot of other electronics. I've ordered a reel of coax and some other assorted bits. I'm going to move everything out of this room and into parts of the house that should be less noisy as well as put the antenna outdoors. Hopefully that lowers the very high noise floor. I'll report back with results.

  
  

  On the plus side, I have removed the EoP so its a step in the right direction. 
• G8JNJ

  November 2017

  Hi Roach,

  
  

  "The loop itself is sitting in the same room as a lot of other electronics"

  
  

  Ah that explains it - be sure to let us know how you get one once the loop has been moved.

  
  

  Regards,

  
  

  Martin - G8JNJ
• roach

  November 2017 edited November 2017

  So, I decided to take the SDR and antenna out for a little drive and get away from local noise sources. I was about 30mins out from town in a mainly farming area. Everything was running on a 12v battery so no SMPS noise. I’ve attached a photo of the setup. One thing that I did notice is that the AM block is basically required. If I did not have it installed I could hear the same strong radio station repeating many times (on different frequencies). Something else I noticed is that even with the AM block the SDR is still experiencing a lot of ADC overload and there is a lot of repeating noise signals even though nothing should be introducing noise. Does this mean that the antenna might be overloading the SDR? I’ve attached some screenshots of the waterfall while I was out and about.

  Also, I’ve moved the antenna and SDR to parts of the house which are not heavily using electronics but the ADC is still showing a lot of overload. The AM band is clear, but at this point I’m not really sure what to do, perhaps I should reach out to Andrew and see if he has any ideas regarding the wellbrook. Any other thoughts?

  The attached images

  • 1 - The ‘DXpedition’ setup
    
  • 2.  The waterfall with AM filter (still seeing a lot of overload)
    
  • 3. A zoom in on the noise at 21Mhz
    
  • 4. The waterfall with cheap bunny ears instead of the active antenna
    

  
  

  Also the SDR is live at the above mentioned address if anyone want to get a closer look at the noise

  
  
• roach

  November 2017

  Spending some time listening to the weak radio stations (there seem to be a lot ~17Mhz) I can hear a repeating interference pattern. If I look somewhere else, I can see it a lot clearer

  
  

  The attached image is what I'm seeing (It repeats across the whole waterfall If I zoom in). Anyone seen this type of noise?

  
  

  
  

  
  
  Attachments:
  https://forum.kiwisdr.com/uploads/Uploader/9b/ba5636f0f1000e565944d6800b9272.jpg