I was just going over it while building a new 44ft doublet and I’m suddenly confused by DL6GCA saying you must avoid feedline lengths that are a multiple of electrical half-wavelength. This doesn’t readily make sense to me. A half wavelength feedline simply delivers the same impedance at the antenna feedpoint to the end of the feedine. I would think instead that what you actually want to avoid are feedline lengths that transform/change the feedpoint impedance to a very high value at the antenna tuner terminals (source), thus presenting a load that might be impossible to match. This would require something like a Smith chart calculation for each frequency and feedline length to check impedance at the source, not a simple calculation of electrical length of feedline.
Do I have this right or wrong? Comments appreciated.
Thanks Armin, I now realize I had a misunderstanding on the spreadsheet. Column C is the TOTAL electrical length of radiator PLUS feedline. I previously thought it was the length of just the feedline. It makes obvious sense now that the TOTAL length should not be a multiple of 0.5 wavelengths! (duh)
A multiple of half the wavelength creates a very high impedance at the end of the line of approx. 3kOhm. Your tuner will not be able to transform this impedance.
This impedance is also achieved with the EFHW, but the 64:1 transformer does the job.
Hello, I’m hoping to make a 44 foot doublet. (22 foot each leg) I will use 300 ohm ladder line and want to be able to use the antenna at 40 meters lowest. I’ll use a 1:1 balun to a manual tuner. I found the nice Excel spreadsheet that helps with determining lengths. I tried to translate it to English but still a little confused. The lowest number for feedline length I can enter without getting red numbers showing up is 16.6 meters.
I assume that the 44 foot doublet is pretty popular and many of you already know what a good feedline length would be but I can’t find anywhere on the interwebs one instance of a written length. Am I able to get away with less than 16.6 meters or 54.5 feet of 300 ohm feedline? Can someone please recommend a minimal length? Maybe @DL6GCA ?
And the last column even says that the particular lengths aren’t suitable for use on 40m, even though the number comes up green.
The underlying problem is that the antenna impedance is rather low, and very reactive, on 40m. When you do get a reasonably usable length, it often is unsuitable on the 2nd harmonic. This is a common problem with the 44’ doublet: while radiation is broadside to the antenna, the ease and efficiency of matching it on 40m can be difficult.
One thing you can do is to use the spread sheet to see if you can find two line lengths, such that any band of interest can be matched using one or the other. Then make a way to switch in the extra line when needed, rather than trying to find one length that works for all bands.
Details will also depend on your tuner: some can match a wider range than others. And don’t be surprised if the tuner efficiency is low.
I had some issues matching a twinlead-fed doublet on my roof, and tested several different tuners. The results are here:
Thanks and good article. Seems I really need to make sure the impedance doesn’t get too low.
The issue I have with the Excel chart is ANY value I enter into the “Length of the feed line” column will give me “Nein” value in column E for 7MHz for a 44 foot Radiator length (6.7m leg) with 0.88 VF. I think i’ll just start at 55 feet of feedline and cut off pieces measuring along the way.
The problem is that the antenna is rather short on 40m. You can certainly get it to load up, but it is around the point where the antenna system becomes less efficient (losses in the twinlead and in the tuner) and the impedances become more extreme. Conditions for improving those on 40m might not also work on 20m or some other bands.
You may have to decide whether performance on 40m is more important than having a specific radiation pattern on 10m, because increasing the antenna length is the best way to make it easier to match on 40m.
I’ll have to build some of that feedline and measure the losses. Some of the stuff using PVC insulation instead of PTFE or PE can have pretty high losses, and I’m not sure of the effect of leaving the two center wires in place.
It probably could be built just as well with the old flat 4-conductor telephone cords, although the losses I’ve measured are often worse than RG-174 above 20m or so.
There is a table in the ARRL Antenna Handbook listing the total length of (half the top wire plus the length of the feedline) showing what lengths to avoid. Basically it is lengths that are a half wave or multiples of a half wave on any band the antenna is intended to cover. If the feedline has a Velocity factor less than 1.0 that should be taken into account.
The table may well be in the ARRL Handbook too. Of various editions.
It is also possible to just copy the ZS6BKW dimensions (either the full size of 28m top wire or half that) including the suggested feedline length for 300 or 450 ohm line.
However the approach of starting long and cutting a foot off it and remeasuring SWR on each band will have the benefit of yielding lots of small stubs of feedline which probably have a use for something. When measuring the swr it is also worth trying to match the impedance with your favourite ATU, whether inside a radio or not. If it matches quickly, I’d consider that a possible solution for that band.
