[RFI] Documenting Power Line Noise

[Ford Peterson]

Ed-W1RFI wrote:

**************************
"Related to the testing Ford is doing, I stayed late last night and drafted some test procedures and information that, although intended to apply to BPL, may be useful in any such measurements.

Comments appreciated:

http://www.arrl.org/~ehare/bpl/measurements.html
http://www.arrl.org/~ehare//aria/ARIA_MANUAL_TESTING.pdf

...SNIP...
**************************

Ed, and others.

I have attempted to document my noise floor using the two described techniques.  The first method, using a spectrum analyzer, and the second method using the S meter on my IC-746 with no preamps or attenuation.

The first of two power lines running near my house was brought back on-line over the weekend.  The noise is greatly reduced from what I experienced last winter, but it is still there none-the-less.  My noise floor with the power line off-line is S0 on 160M.  With the power line active, it is S 7 white noise.  

The S meter has been calibrated to a signal generator that has been checked and is reliable to within a couple of dB.  Since I use the same generator to document the S meter and calibrate the Spectrum Analyzer, the relative readings, and more importantly, the changes in the relative readings, should be valid.  

Here is how my IC746 shakes down:

Output of the HP-606A at 1.850 MHz necessary to deflect the S-Meter on the IC-746 with no attenuation and no preamps enabled.  (missing data omitted due to oversight).

 

      dBm
     S Units
     
      -67
     9.0
     
      -70
     8.5
     
      -73
     8.0
     
       
     7.5
     
      -76
     7.0
     
      -80
     6.5
     
       
     6.0
     
      -83
     5.5
     
      -84
     5.0
     
      -85
     4.5
     
      -87
     4.0
     
      -88
     3.5
     
      -89
     3.0
     
      -90
     2.5
     
       
     2.0
     
      -91
     1.5
     
      -92
     1.0
     
      -93
     0.5
     
      -94
     0.0
     


With the power lines active, the IC-746 in 'narrow' CW (500 Hz filters in both IFs), I observe an S meter reading of 7.0.  When the power lines are not active, I observe an S meter reading of zero.  This suggests at least an 18 dB change in the noise floor.

Looking at the Spectrum Analyzer, 3 KHz BW and 10 KHz span, I observe a change in the noise floor of 5 dB corresponding exactly to the absolute levels observed by the S meter.  Power lines active, -75 dB.  Power lines inactive, -80 dB.  Using the CW "wide" position on the receiver, I observe a reading of S 7.0 when the power lines are active, and S 6.5 when they are inactive.  This suggests 4 dB but is well within the margin of error on the S meter.

Changing the Spectrum Analyzer to 1 KHz BW and a 10 KHz span, I observe a change in the noise floor of about 15 dB, which again, is well within the margin of measurement error.  This serves to confirm Ed's methods, and I believe it should serve as a good baseline for my upcoming battle with the utility.

Looking at the SA with a bandwidth of 0.3 KHz and 1 KHz span, the noise floor is down to -94, (18 dB change) which again confirms the S meter versus the SA methods.  This setting more closely resembles the receiving characteristics of the receiver in CW 'narrow' mode (500 Hz).  In fact, with 500 Hz filters in both IFs of the receiver, the effective bandwidth will be somewhat less than 500 Hz anyway.

I have photographed the images of the SA screen and described the images and procedures in a MS Word Document.  Since the power line is inactive for another few days, is there any other base line measurements that should be taken?  

Thanks for the help.

Ford-N0FP
ford at cmgate.com