Researching coax loss

As I'm in the process of rebuilding my antennamast and hardware I'm searching for the best options available. But not against any cost. Now I have measured almost no loss from PL-259 connectors on HF I think the best way to reduce loss is to buy a good coax cable. There are several options for this, the lower the loss the higher the price. But does 1 dB or less loss make your system a lot better? Another thing you have to think of is the amount of power and the length of coax you are going to use. You can imagine if you use several kilo watts you can't do with a simple (thin) coax cable. And when your mast is several hundred meter from your shack it will add considerable loss. But anyway, I'm not planning to use more power as 100W. The cables I will use are maximum 50m long. Do I need to have expensive coax for my system? I always had good results with Aircell7, but when I did a small research on the coax in 2013 I had it seems that my cheaper Belden H-1007 has better mechanical strength.

So, again some research revealed I have several options now. Available still is the original Aircell7. But Belden has the H-2007 as replacement of the H-1007. Messi+Paoloni has the Ultraflex 7. All the same kind of cable with losses close to each other and all with a diameter slightly over 7mm.
The H-2007 is the cheapest of these 3. Some specs of losses on 28MHz @50m 7mm coax:

H-2007 @28MHz: 1,4dB
Aircell7 @28MHz: 1,5dB
Ultraflex7 @28MHz: 1,5dB

Something else is very important, the outer screen on the Aircell7 has a density of 70%. The Ultraflex7 has a density of 83% and the H-2007? But the screen has more density compared to Aircell7 as my 2013 research showed. Cost is important. But losses are important as well. What will I gain with better coax. Some specs of expensive coax cables losses on 28MHz @50m.

Hyperflex10 @28MHz: 1,03dB
Hyperflex13 @28MHz: 0,7dB
LMR400 @28MHz: 1,1dB

The price of the Hyperflex/Ecoflex and LMR400 coax is about 3,8x higher per meter compared to the H-2007. And does it help at 50m coax length? Certainly it will help but I don't know if you can see that practically on HF. It is only 0,7-0,3 dB less loss. Just to get an idea I calculated output power in relation to the type of coax I prefer to use:

(preferable) 50m H-2007 @28MHz 100W input (1,4dB loss) 72.4W
50m Hyperflex13 @28MHz 100W input (0,7dB loss) 85.1W
50m RG58 @28MHz 100W input (3,15dB loss) 48,4W

The tool I use for calculating loss can be found here:

http://www.sengpielaudio.com/calculator-amplification.htm

Is a price of 3,8x the price of H-2007 worth the 12,7W extra at the antenna? For the power purists yes, they have 100W and they want it at the antenna. Personally I don't care about those 12,7W. It is a different story when you have a high SWR and tune you antenna with a internal tuner or tuner in the shack. Because your radio will send 100W but most is heated up in the coax if SWR is not good, in that case a thick lossless coax helps a bit ;-). As regular readers know I got a remote autotuner at the feedpoint of my vertical and cancel out the most of the extra coax loss.

When I started on CB I used RG58 coax, I remember I always had a good SWR. But my signal with the legal 500mW was weaker compared to neighbour stations. I replaced the coax for RG213 later and it made a considerable difference, although my SWR was not as good as I thought.

The above is true but puzzles me. Or the 10m RG58 must have been very very bad at that time. Normally it should have a loss of approx 0,63dB. RG213 would have a loss of 0,12dB.
With the 500mW we were allowed to use the power at the antenna with RG58 is 432mW and with the RG213 it is 486mW. You should hardly notice this, however I did notice at both TX as RX. So investigating (you can google everything these days) showed me that coax cable degrades over years. Obviously the RG58 was old and the RG213 was brand new. I guess the RG58 I used in those days was degraded badly and had considerable more loss as the 0,36dB it should have...

Now I still had some leftover coax cables from my own tower and from PD0FF's tower. A great opportunity to measure loss and compare cables with my MFJ259B analyzer. I did it before to measure loss in connectors and although it isn't professional equipment it does a good enough job for amateurradio needs.

Then it became evident to me that you shouldn't buy expensive cables if you only use HF. The differences are only there on VHF/UHF and higher. On HF a good H2007 cable is just as good as expensive Aircom+ or Ecoflex10. Even better on lower bands. See my measurements below:

Band	Loss Aircom+ l=10m (dB)	Loss H2007 l=10m (dB)
160m	0,1	0
80m	0,2	0,1
60m	0,3	0,2
40m	0,2	0,3
30m	0,2	0,3
20m	0,3	0,3
17m	0,3	0,4
15m	0,4	0,4
12m	0,4	0,4
10m	0,4	0,4
6m	0,4	0,6

All relative of course. Both cables are about 12 years old and the Aircom+ could be degraded a bit more? Factory specs show that the H2007 should have a loss of 0,3dB on 50MHz and Aircom+ a loss of 0,2dB at 10m length. However that is only theoretically and if new and not used. It will be could to test my length of coax before I install it between the shack and tower.

8-3-2018: Made some calculating changes thanks to a comment from PA1PRD. Tnx Erik!