View Single Post
Old 06-06-2009, 05:34 PM
Altair Altair is offline
Senior Member
Join Date: Feb 2009
Posts: 107
Hello all,

I had some time to look in depth at the original circuit.
I have redrawn it differently just for fun. See attachment 1.

A bridge rectifier is now in place of the power supply, so consider the circuit being powered from AC . I removed the 2 shunts, and also removed D1 & D2 which are now replaced by the bridge.
The diodes are Shottkys, to be able to switch efficiently at higher frequencies.
I also added a coil in series with the heating element, because commercial elements I've got here are very low in inductance.
Of note is that the main difference between this circuit and a standard boost converter is that in the boost, the pulse of energy which is sent to the coil to magnetize it, is not available to the load. It is wasted to ground. Here however, that magnetizing pulse goes through the heating element.
Then of course, the flyback pulse continues heating the element further..

I then made another drawing, with a second element (HE2), that becomes active during the duration of the flyback pulse. According to armagdn03, additional impedance during the flyback pulse would be beneficial. If we make this impedance from an additional heating element, I assume the results would be increased efficiency. See attachment 2.

Of course, in this circuit, we're not running in resonance mode. It's just basic pulsing of a coil.
No tests have been made yet, I don't have all the parts I need right now. I would be interested in hearing what you think of this scheme.

The third drawing is the "What if..." option
Instead of the second element, I put another coil which is in fact the secondary of a transformer. The primary is L1, the first coil. This could be a bifilar wound coil.
The way it works (if it really does !) is that during the charging of L1, the current of L2 is blocked by D1.
Then, during the flyback pulse, L2 would be (supposedly) aiding L1, being of the same polarity. See attachment 3.

Would that really be worthwhile ? I have absolutely no idea ! Maybe there would be a way to integrate a capacitor to make it resonant, but I don't think it is possible to combine the 2 modes (switching & resonant) in the same circuit.

Anyway, do you think putting a coil in series with the heating element as I have done in the first circuit, is a good solution to transfer more energy with each pulse ? I understand that the coil should theoretically have the lowest possible resistance, however as the goal is to make a >heater< unit, thermal losses in the components are irrelevant aren't they ? Or maybe they are, because it would reduce the efficiency of the whole unit...

Questions, questions

Attached Files
File Type: pdf Heater 1.pdf (4.9 KB, 223 views)
File Type: pdf Heater 2.pdf (5.1 KB, 164 views)
File Type: pdf Heater 3.pdf (5.5 KB, 146 views)