Post by wq1p on Dec 4, 2009 9:07:24 GMT -5
Bob,
In the past I have measured feed line length using a 50 ohm load in parallel with the desired feed line coupled to a SWR meter and a good frequency counter. Obviously as the feed line approaches "infinity" at the 1/2 wave resonance, the process is not sharp, presenting a narrow band of frequencies. OK, but not great. What a pleasure it is to use the AIM4170.
While the AIM results were far more accurate than previous measurements, and extremely useful, that greater accuracy has prompted questions which will hopefully further my understanding of the data and the physics.
First, I noticed a systematic bias wherein the "length" of the cable seems to shorten with progressive 1/4 wavelength measurements, i.e., 1/4, 1/2, 1, etc. I have noticed this with previous measurement tools; however, given their lesser accuracy, i assumed it was a result of the instrument, not the cable. Now I guess it is the result of losses in the cable, but I would love to understand the physics and math involved.
Second, regardless of the limits used, the data was very tight at the 1/2 wavelength and 1 wavelength points, and less tight at the 1/4 wavelength measurement. Some of this phenomena improved with averaging, but the difference in dispersions was intriguing. Obviously, as the limits are expanded, the granularity of measurement decreases (all plots were done with 1000 points), and the actual theta crossing in most cases is interpolated. That explained the narrower dispersion with tighter limits and, to a degree, averaging, but it did not fully explain the dichotomy. My temptation was to use the 1/2 wavelength "length" because of the tighter dispersion, but the shortening phenomena caused me to question which measurement was closer to reality.
A sample of the data collected using a piece of CQ8xMM Cable (stated velocity factor - 0.78)
First resonance (1/4 wavelength)- 0.9873MHz (Range 0.9868 to 0.9877 depending on limits used and whether averaging used). This implies a electrical length of 249.0527 feet
Second resonance (1/2 wavelength) - 1.9952MHz (Range 1.9951 to 1.9953 depending on limits used and whether averaging used). This implies a electrical length of 246.4813 feet
Fourth resonance (1 wavelength) - 4.0152MHz (Range 4.0151 to 4.0153 depending on limits used and whether averaging used). This implies a electrical length of 244.9589 feet
I can email the whole data set, but I could not figure out how to include it here without hand typing it.
While I realize the exact length is not important, since i can "custom calibrate" and effectively move the antenna feed point to the shack (a TERRIFIC feature of the AIM4170), a significant part of my goal is to educate myself and develop an understanding of the math and physics.
Many thanks for your help.
John, WQ1P
In the past I have measured feed line length using a 50 ohm load in parallel with the desired feed line coupled to a SWR meter and a good frequency counter. Obviously as the feed line approaches "infinity" at the 1/2 wave resonance, the process is not sharp, presenting a narrow band of frequencies. OK, but not great. What a pleasure it is to use the AIM4170.
While the AIM results were far more accurate than previous measurements, and extremely useful, that greater accuracy has prompted questions which will hopefully further my understanding of the data and the physics.
First, I noticed a systematic bias wherein the "length" of the cable seems to shorten with progressive 1/4 wavelength measurements, i.e., 1/4, 1/2, 1, etc. I have noticed this with previous measurement tools; however, given their lesser accuracy, i assumed it was a result of the instrument, not the cable. Now I guess it is the result of losses in the cable, but I would love to understand the physics and math involved.
Second, regardless of the limits used, the data was very tight at the 1/2 wavelength and 1 wavelength points, and less tight at the 1/4 wavelength measurement. Some of this phenomena improved with averaging, but the difference in dispersions was intriguing. Obviously, as the limits are expanded, the granularity of measurement decreases (all plots were done with 1000 points), and the actual theta crossing in most cases is interpolated. That explained the narrower dispersion with tighter limits and, to a degree, averaging, but it did not fully explain the dichotomy. My temptation was to use the 1/2 wavelength "length" because of the tighter dispersion, but the shortening phenomena caused me to question which measurement was closer to reality.
A sample of the data collected using a piece of CQ8xMM Cable (stated velocity factor - 0.78)
First resonance (1/4 wavelength)- 0.9873MHz (Range 0.9868 to 0.9877 depending on limits used and whether averaging used). This implies a electrical length of 249.0527 feet
Second resonance (1/2 wavelength) - 1.9952MHz (Range 1.9951 to 1.9953 depending on limits used and whether averaging used). This implies a electrical length of 246.4813 feet
Fourth resonance (1 wavelength) - 4.0152MHz (Range 4.0151 to 4.0153 depending on limits used and whether averaging used). This implies a electrical length of 244.9589 feet
I can email the whole data set, but I could not figure out how to include it here without hand typing it.
While I realize the exact length is not important, since i can "custom calibrate" and effectively move the antenna feed point to the shack (a TERRIFIC feature of the AIM4170), a significant part of my goal is to educate myself and develop an understanding of the math and physics.
Many thanks for your help.
John, WQ1P
