2007-05-26
Peter VK2TPM's blog inspired this project.
I've built J-poles before for 3 metres (back in my pirate days), but never with twinlead. As Peter reported problems with KC7CCR's dimensions, I decided to characterise the twinlead available in VK-land, assuming it might be different enough to be causing the problems.
DSE carries two different kinds, one black for outdoor use (UV resistant) and clear for indoor use only. I had a few metres of each, so I cut 1.520 metres of each and shorted both ends with a piece of resistor leg. Both were dipped as accurately as possible and the velocity factors calculated:
Black 300 Ohm Twinlead (DSE W-2070): 0.826
Clear 300 Ohm Twinlead (DSE W-2071): 0.833
The textbook value is generally agreed as 0.85. I predicted that the black might be a bit slower than the clear, assuming the dielectric might be loaded with conductive material. The difference is fairly small, but it produces a dip-frequency difference of around 700 kHz near 82 MHz. (The length error needed to produce this is more than 15 mm, and the lengths were equal within at least 2 mm, if not better.) Of course dipping is of limited precision, but the black is quite noticeably and repeatably measurable as slower than clear at all the VHF frequencies I tried.
With this information it is possible to calculate the stub length needed for resonance at the harmonic centre of the 2 metre band (145.973 MHz):
Black: 424 mm
Clear: 428 mm
The original design specified 425.5 mm which is in excellent agreement with the black twinlead. Using my impedance bridge and a HT delivering 100 mW as a generator I trimmed down a length of both types of cable and got a resonance near 426 mm for both, the black slightly shorter. The "known" port was left open, and the shorted stub applied to the other, its length trimmed for the best null. The accuracy is limited by the connector length, the shorting bar, etc, but I think this result is in good agreement with the physics. My dipper agreed as well, and a few mm is hard to resolve when the stub pulls the dipper as you approach resonance. The bridge is actually capable of better resolution, especially with the synthesised generator, but the null is fairly broad and challenging to find the true bottom of.
1 comment.