Yes, there are some lower loss small cables, but they might not be easy to find in small quantities.
Several decades ago I switched to using 8m of RG-174 for my portable dipoles. Yes, according to the tables in the ARRL Antenna Book, I might lose 1/4 of my power on 10m, but the benefit of getting the antenna up higher overcame the potential losses. (And the losses were lower than the speaker cable that I had been using, besides giving a better impedance match.)
This was before the EFHW designs with a transformer were popular (I used a manual antenna tuner back then). I carried dipole wires for all bands of interest, and could choose what bands to add each time I set up the antenna, depending on the specific circumstances. So I didn’t always need to set up enough wire for 40m or 80m, but I had them available.
But recently I’ve picked up some other sorts of cables, and done more analysis on the losses of the different types. Here is my summary of feedline types so far: I have some further work that doesn’t appear there yet.
Several findings…
A major issue is that the data for RG-174 (and some other small cables) aren’t always correct. I’ve recently double checked the data from my old ARRL Antenna Book and the stated losses are too high above about 10 MHz, and too low below that, compared to the Belden 8216 data sheet and my actual measurements of several samples. (Some of the error may be in interpolating the charts.)
The main source of loss in such small cables, especially in the lower HF range, appears to be in the stranded Copper Clad Steel (CCS) used in the center conductor. The result is that the loss curves tend to flatten out in the lower HF range, as the skin depth of the copper in the center conductor becomes inadequate. Tables that interpolate the losses based on a 50 MHz loss specification are not going to be accurate. You can see this effect in the ratio of the 40m loss to the 10m loss.
I’ve also found that some less expensive RG-174 appears to use less copper in the center conductor, which increases losses a little further, even though the cable is marked with the same type number.
Now, we often think of 75 ohm cables as having lower losses than 50 ohm cables, but that doesn’t necessarily apply to small types that need to use CCS center conductors for strength, because the center conductors are smaller. For example, RG-174 / RG-316 (50 ohms) and RG-179 (75 ohms) are about the same diameter, but the losses in RG-179 are higher, especially on 40m.
Some cable options:
Probably the most available low-loss small coax is LMR-100. I haven’t compared it to the other types yet. I have a bit of a concern with using a foam-dielectric cable outdoors for portable use where it may get stepped on, but that might not be an issue in practice.
RG-174 isn’t bad, as long as the lengths are kept reasonably short. RG-316 is the PTFE version. Both are solid dielectric 50 ohm cables.
Belden 88907 is a thinner version of RG-58 / UR-43. A bit lighter and less bulky, with similar losses. Might not be worth the effort of trying to find it, but worth trying if you do. Might have to add some heatshrink tubing to get connectors to fit.
Belden 9221 is a foam-dielectric 75 ohm cable the same size as RG-174, but somewhat lighter. I found a cheap spool at a hamfest because it was 75 ohms, and am trying out using a 1/2 wavelength for 40m (about 15m) in a 20/40m dipole kit. That’s more cable than I would normally carry, however, so I don’t really save anything. Would be better for just 20m. (If you have an antenna tuner, then the cable impedance shouldn’t matter, and you can use other lengths.) My measured cable losses are a bit higher than the data sheet.
Belden 735A1 is another 75 ohm cable, a bit larger (3.3mm) and heavier than RG-174, but with lower losses (partly due to a pure copper center conductor). This would be my choice for a 75 ohm cable when I wanted to get my antenna up higher than my typical 8m length. Again, might not be easy to find, and possibly only as a 305m spool. (Another hamfest find.)
The best 50 ohm cable that I have found is that used by Tektronix for internal wiring of their equipment. I don’t have a part number for it - probably a custom order. PTFE insulation, with stripes of various colo(u)rs. Size of RG-174, but solid copper center conductor gives it lower losses. It helps to have former Tek employees at local hamfests!
I wouldn’t recommend going smaller than RG-174 for any significant length: both RG-178 (50 ohms) and RG-179 (75 ohms) have higher losses. I use some RG-178 in a balun to get more turns in a small core.