Jim Lux wrote:
Roy Lewallen wrote:
Jim Lux wrote:

The other factor is core saturation.. flux goes as the
voltage*frequency ... so 10:1 VSWR means you might have 10x voltage..

At HF, the ferrites which are best to use for baluns will go up in
flames due to loss at flux densities well below saturation.

Roy Lewallen, W7EL


Well then, same general idea.. lots of volts = lots of flux = lots of
dissipated energy from the core loss...

Not exactly. The dominant loss mechanisms in "low frequency" ferrites
are essentially linear, and don't cause distortion. Saturation occurs
during a portion of each cycle, resulting in distortion. And saturation
doesn't always imply high loss.

And, of course, not all baluns are made with appropriate materials (e.g.
using a lossy EMI suppression mix might be fine in a "choke"
application, where high Z is keeping the current low.. but terrible in a
transformer type situation, where you have a lot of flux in the core)

A properly designed transformer has very little flux in the core -- the
flux is the vector sum of the flux in all the windings, which in an
ideal transformer is zero. The flux you do have is the magnetizing flux
due to the finite impedances of the windings, which you strive to
maximize just as you do in a choke (and the leakage flux which isn't
coupled from one winding to the others). So the best core material is
generally the same for a broadband choke as it is for a broadband
transformer. EMI suppression materials are very good for both, since
they're engineered to maximize impedance. An exception is where you're
dealing with enough power that the core loss is intolerable. For
example, a well designed choke or transformer might have only 0.5 dB
loss, a generally insignificant amount. But if you apply a kW to it,
you're talking about 120 watts of power dissipation, too much for a
small core. In those cases you have to use lower loss material, which
usually means lower impedance windings and consequent higher flux density.

Roy Lewallen, W7EL