[CQ-Contest] Power Inverters

[Tom Rauch]

Hi Bob,

 > As for a square wave into a transformer, the math will show that the drop
 > from horizontal to vertical becomes dampened as a function of time, so as
 > to "round out" the wave pattern. It is no longer and abrupt drop, as would
 > be patterned in a square or rectangular pattern.
 > Bob Perring

I agree that most switching supplies don't care if the input is a sine
or square wave (as long as peak voltage is within range of the
regulation system), but I disagree strongly with the suggestion a
conventional isolation transformer behaves as a rounding device,
especially through hysteresis.

There are two primary loss mechanisms associated with the core
in a transformer. One relates to eddy currents, and is reduced by
insulating small sections of the core from each other so eddy
currents are significantly reduced. The second is hysteresis.

Both of these problems convert electrical energy into heat, both
reduce the inductance of a winding on a core. Both are VERY
undesirable in conventional transformers, and in the conventional
model of a transformer are represented as Rc, a dissipative
resistance that "shunts" the primary.

Hysteresis is responsible for the currents associated with simply
magnetizing the core with no load. Cores are  **intentionally**
selected to have low hysteresis so they do not become hot, and so
the transformer is efficient at low power levels.

If what you say is true, and a transformer "rounds off" a square
wave, the transformer would have to have significant attenuation to
a signal on 180 Hz and above. It would also significantly heat with
the application of a signal at 1/3 the frequency where loss is
extreme.

This wouldn't be a slight attenuation, it would have to have
significant attenuation.

If it is hysteresis, than perhaps you can explain the following:

1.) How does the core change the slope of the output voltage
without reducing the voltage passing through the transformer, if it is
a 1:1 transformer. In other words, how does the sinewave magically
crest 1.414 times higher than the square wave going into the
transformer when the turns ratio is 1:1?

2.) How does the transformer change the slope of the output
waveform without changing the loading on the supply-end during
the cycle of a sine-wave input?

3.) Absent any tuned circuits resonant on 60 Hz, how does
transformer show significant attenuation to all frequencies above
179 Hz, while not showing any attenuation to signals below 61 Hz?

Constant voltage transformers can behave as you say. The output
can be a sine wave, because it has a capacitor in parallel with the
secondary winding. But you'll also find a constant voltage
transformer designed for the same input and output voltage runs
very hot with no load, loads the power source unevenly, and has a
turns ratio greater than 1:1 between primary and secondary. It does
not break any of the rules, because no transformer can.

73, Tom W8JI
w8ji at contesting.com


--
CQ-Contest on WWW:        http://lists.contesting.com/_cq-contest/
Administrative requests:  cq-contest-REQUEST at contesting.com