2
\$\begingroup\$

Update: got it working... Fix: replace diodes for proper schottky. Result: got oscillation but no resonans. Fix: replace cap with 1 uF polystyrene. result: got resonans.

But now my problem is i cant get any power out of it! I know the freq is a bit high to properly heat steel, its around 330kHz. But it draws 3.5A running empty peaking at 50V, and only goes to 5A at 20V when i insert a workpiece.

How to upp the effect? or is this generally an ineffective design?

old question I´m getting a bit interested in resonans and found a circuit for a very simple self oscillating inductuiong heating circuit. im new to using powertransistors.

http://www.rmcybernetics.com/images/induction-heater-schematic.jpg

In addition i put a 12V zener diod and an a 10kR between node R1,T1,D1 and ground to protect my gates from overvoltage

My problem being the circuit wont oscillate. One MOSFET turns on, stays on. The circuit draws ca 3A at 12 V. And nothing much more happens.

I have a hard time understanding why..

Im was thinking that the mosfet needs a negative flank with respect to drain in order to turn of proper (enhancement mode?). But the page explaining uses STP30NF10 which are similar to mine. I am using IRF540 enhancement mode Mosfets 100v 28A.

Anyways... below is schematic of my current non-functional stup. .. please help as to what might be causing the problem.

schematic

simulate this circuit – Schematic created using CircuitLab

actually, while doing the drawing s i noticed that my diodes d3,4 are not schottky diodes. but have a 0.6 Vdrop. would that do it? the circuit is theoretically supposed to osc @ 240kHz.

\$\endgroup\$
7
  • \$\begingroup\$ I really don't see how this circuit could oscillate as shown. Even assuming your L2 and L3 are properly coupled to each other, there's no feedback mechanism that would shut off either transistor once it is on. I suspect that it works for rmcybernetics only because he has a significant amount of impedance in the ground lead going back to his power supply. \$\endgroup\$
    – Dave Tweed
    Jul 18 '15 at 19:54
  • 1
    \$\begingroup\$ I think the possible feedback path is C1 and L1 through D1 or D2 (original schematics). The implemented schematic has at least the following issues: L1 is meant to be coupled (autotransformer configuration) not separate coils. 1N4004 are too slow, so go back to schottky diodes. C1 cannot possibly be 2.2F (that must just be a typo). \$\endgroup\$
    – rioraxe
    Jul 18 '15 at 21:25
  • \$\begingroup\$ The gate circuit appears to be symmetrical hence a starting circuit sould be added to ctreate asymmetry so as to start any one of the MOSFET when switched ON. Cap C1 & Choke L1 forms a resonating circuit at 240KHz hence value of C1 to be matched with L1. VTIngole \$\endgroup\$ Jul 19 '15 at 7:13
  • \$\begingroup\$ @DaveTweed i suspected that would be a problem. Im new to this. When i tested the irf540 i needed positive to turn on and it stayed on til it got negative flank. But when i probe the circuit as shown above, the mosfet turns off at zero volt. could you maybe please suggest a simple way to introduce this turnoff-pulse? BJT´s? \$\endgroup\$ Jul 19 '15 at 8:56
  • \$\begingroup\$ @vijayingole L2&L3 represents one centertapped aircoil, the workcoil as it were, and it is this which constitutes the LC-Tank. How would a starting ciruit look like? \$\endgroup\$ Jul 19 '15 at 9:05
1
\$\begingroup\$

Take 2 diodes, connect them in series. connect them across R6 such that they are reversed biased. Connect a ceramic cap of 0.33 micro (approx.) between junction of the diodes and ground. NB I actually have not understood how the MOSFETS would flip -flop as theree is no non-linear magnetic component in the tansformer coil (circuit.L2-L3)..vtingole

\$\endgroup\$
2
\$\begingroup\$

Simply just reduce your L2 Choke 50uH to 15uH ......circuit current will increase. Your Choke value according to your frequency is very high.....Choke resist the change in current .......or you can increase your capacitor of tank circuit to reduce the frequency so that your 50uH choke will be sufficent to supply more current.

\$\endgroup\$
1
  • \$\begingroup\$ One period per sentence will do Hasan. Also when stating inductor/capacitor changes it helps to add some math for proof, and to validate your answer. \$\endgroup\$
    – user105652
    Oct 15 '18 at 2:51
0
\$\begingroup\$

The first circuit is a thoroughly tested design, so the question whether it will work or not does not arise.

As far as striking resonance is concerned, L2 is one of the crucial elements which must be tweaked optimally.

An alternative method would be to try powering the circuit from a lower level, say from 6V onward and gradually increase the voltage until the right resonance is achieved.

The above must be done with a "load" inserted within the work coil.

\$\endgroup\$
0
\$\begingroup\$

D3/4 should be UF4007, also your gate resistors R5/6 should be 470ohm 5watt wire wound. This will ensure a significant difference and create the start up. L1/2 should measure around 3uH each. That should do it.

\$\endgroup\$
2
  • \$\begingroup\$ There is no guarantee this circuit will oscillate. In an age where PWM ICs are everywhere for any need, this circuit is a childish diy. Also for induction heating substantial power is needed. \$\endgroup\$
    – user105652
    Sep 28 '18 at 23:08
  • \$\begingroup\$ This circuit will oscillate as a result of the offset created by the difference in the 5W 470R wire wound gate resistors. Regarding the use of PWM IC's you need to be specific as to what chip you are referring to. As all PWM chips were not created equal. Also the childish diy comment is in itself childish since not everyone knows it all. I thought that this was a forum for people to share some knowledge and experience as opposed to leaving silly remarks. \$\endgroup\$ Oct 23 '18 at 11:31

Not the answer you're looking for? Browse other questions tagged or ask your own question.