I have finally started experimenting with the revised antenna tuner to match a nominal flat 450Ω balanced feedline (actual Zo=438Ω) to tuned feeders running up the tower. To-day I began with the 40m tuner, and was able to get a good null with the MicroMatch impedance bridge.
The problem is to determine that the load to the balanced line really is 438Ω + j0 (or 450Ω +j0, close enough for government work). I have a couple of dummy loads made up of Glo-bar composition resistors. My small one uses nine 50Ω 5 or 10 watt Glo-Bars wired in series. The larger one uses three banks each of four paralleled 600Ω Glo-bars, wired in series. In each case, with all the resistors wired in series, there appears to be enough residual reactance to skew the readings from the predicted, at least on 40m.
I have considered using a bank of something on the order of 5K 2w carbon composition resistors wired in parallel, to make up a 20-watt or so 450Ω resistor, and then submerging the whole thing in a jar of mineral oil, à la Cantenna, to give enough wattage rating to make appropriate measurements, without a string of anything in series to accumulate inductance. I could dig through my collection of hundreds of new and used resistors, but that would be a real PITA, and no guarantee I would find enough of the correct value to complete the project, so I considered purchasing new ones. I checked out Mouser and Digikey. Looks like the largest carbon composition resistors now available is 1-watt. Neither company lists anything at 2 watts.
Film resistors are listed at up to 2 watts, and their tolerances are much closer than anything you can find in carbon comp, but film resistors are made up of a layer of carbon or metallic film deposited on some kind of ceramic or composition form, with a spiral groove cut along the length. IOW, they are similar to wirewound resistors, except that the conductor is a flat film of carbon or metal, instead of wire. I don't believe that would be satisfactory for what I want to do, since it would have a small amount of inductance instead of a pure resistance. Also, my experience with film resistors is that they are very fragile, unlike carbon comps; just a little bit of overload and they flame out in a puff and flash, and become useless.
More and more common ordinary components are becoming unobtanium. The way it looks, in a few years we won't be able to buy any kind of newly manufactured electronic component except pre-packaged dipped-in-epoxy digital widgets. It appears that useful rf type analogue components are rapidly disappearing off the face of the earth.