http://elecraft.85.s1.nabble.com/Re-NOT-the-feedline-was-KPA-1500-faulting-on-6m-SWR-issues-tp7661502p7661541.html
> Based on my experience, balun power ratings are for MATCHED
> conditions. It is rare that hams use a balun in a matched condition.
> Thus a 1:1 balun should see 50 ohms on the input and 50 ohms on the
> output, while a 4:1 balun should see 200 ohms on the output and 50
> ohms on the input. In the case of a resonant folded dipole, a 4:1
> balun is typically operating in a nearly matched condition. All others
> combinations are unknown and random.
>
> I run about 500 watts on all bands. My baluns are rated at 5KW! It
> takes 3 or 4 big hunkin' pieces of ferrite to attain this power
> level. My 6 meter balun is a 1/2 wavelength electrically of RG-213.
> No ferrite!
>
> Buy or build a balun of your choice. Using an IR temperature gun,
> measure the ambient temperature of the core. Run about 1/2 rated
> power carrier for 30 to 60 seconds. Measure the temperature again.
> If it is warm to hot, this is RF producing heat. And likely
> continuing will produce core failure. This is not a good balun for
> your application.
>
> One of my baluns work between the output of my KAT500 and the balanced
> feed line connected to the center of a 256 ft wire. That antenna
> works 160M - 6M with zero issues. Now, I do run a hybrid balun being
> a 4:1 Guanella balun as a transformer, and it is fed with a 1:1 balun
> for common mode rejection.
>
> Most single core, i.e. 2 or 3 cores stacked with 2 to 4 windings are
> not at all a proper balun design A Guanella balun will have 2 cores
> with 2 windings and then another 2 separate cores with another 2
> windings. These are then wired to produce a 4:1 balun with good
> common mode rejection. Most "factory" 4:1 baluns are poorly
> designed and built junk.
>
> See
https://www.dj0ip.de/balun-stuff/ for further references.
>
> 73
>
> Bob, K4TAX