William Sommerwerck wrote:
Some of the older US made units from Litton or Tappan had
a giant reed relay, and the HV was actually switched on and
off for the defrost mode. They made a nice sound switching
on and off.

I'm curious as to how my GE works.

If, at 50% (say) you were full "on" for one minute, then off for one minute,
you'd be running in that first minute at a power level that would cause
localized boiling. (This is a problem with soups and oatmeal.)

I do not see that. Therefore, I have (???) to assume that PWM with a cycle
of less than one second is used.


There are a few things you can do.
Use containers with straight vertical sides. The thinnest part of the
material
boils first in a typical bowl. I've been acquiring HUGE ceramic coffee
mugs to cook soup.
The other thing you can do is siphon off the energy with a cup of water.
Varying the amount of water and the power cycling can accommodate most
needs.

In some cases, a microwave browning dish can average the long-duration
PWM of the oven.

I've done some experiments with a dish that looks like a lemon juicer.
The donut shape of the liquid gets energy from all sides and can help
moderate the edge boiling. You'd really not want any thickness of the
subject material to be any thicker than the microwaves can penetrate.