DIY Induction Heating with Pll

Discussion in 'The Projects Forum' started by sina_always_stuck, Mar 11, 2011.

  1. sina_always_stuck

    Thread Starter New Member

    Mar 5, 2011
    6
    0
    Hi there,
    My name is Sina, 25 from Iran. I've been interested in this induction heating thing and would like to build one. Its quite a long time searching the net and experimenting to make it work.
    I've been stuck with the pll issue. My configuration is Hbridge which is fed from a simple power supply(not split) so the lowest point is ground. The two signals I try to maintain a phase lock of 90 degrees are the Inverter voltage and tank cap voltage. The inverter voltage is fed from pin 4(VCO out) to pin 3(COMP in) of 4046 as inverter voltage representative. But in order to get a square wave as a representative of cap voltage I'm trying to use a voltage comparator lm393. My question is how to connect lm393 across the cap and meanwhile protecting its inputs from excessive voltage that appear across the cap?

    I tried to put a shunt resistor in series with LC tank and use the voltage drop across it as an indicator of inverter current phase angle but the voltage across it is very odd in its shape and if converted to square wave does not show the phase angle of the inverter current.

    Also I am very interested to use the phase comparator 2 of the 4046, this phase comparator tries to lock zero angle between its input signals which is alright if I was able to use the voltage across the shunt resistor. but in this case I should change the phase angle of the cap 90 degrees and then try to lock zero angle between it and the inverter voltage.
    Any help, suggestions and schematics would be very welcomed. Please let me know if I should send schematics or if my explanations are not complete also I can be available for a chat at any time if this would be an easier way for anyone to help.
    Thanks in advance.
    Best Regards
    Sina
     
  2. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Why do you think phase locked loops are necessary for a heater? All you need for an induction heater is a large coil with lots of current to induce eddy currents in the metal you want to heat, the frequency stability isn't really an issue.
     
  3. sina_always_stuck

    Thread Starter New Member

    Mar 5, 2011
    6
    0
    its a battle between metals and mosfets so we should try to make it very efficient and frequency adjust has the most influence. we can make more heat by less power if frequency is adjusted.
    thanks
    sina
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    You will get much further with a simple frequency counter and a variable oscillator. I think I understand what you are wanting, to make a digitally synthesized function generator. Those tend to be pretty high dollar pieces of equipment (for a reason). For your application I don't think you need that degree of accuracy, not that you would get it without crystal ovens and what not.

    I am not an expert, not even close, but my understanding of induction ovens is not about frequency, it is about raw power you are feeding into the induction coils.
     
  5. sina_always_stuck

    Thread Starter New Member

    Mar 5, 2011
    6
    0
    its 4046 ICwhich has a variable oscillator(VCO voltage controled oscilator) but it does not need frequency counter to operate it needs two signals and is able to lock a phase between them and when it locks the frequency is adjusted. I am able to capture one of these signals and to feed it to 4046 but the problem is how to capture the other signal.
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    If all you need is a variable frequency there are much better and easier ways. I'm fairly sure you've gone down a wrong path for reasons that make no sense. I don't think you know much about electronics, the only reason you would need a frequency counter is to do precision measurements. A phase locked loop does not assure accuracy, or any other characteristic I can fathom that you would need.

    I understand phase locked loops, and what they do, very, very well. I've made my share back when. I also understand frequency synthesizers, I used to repair and calibrate them.

    Where do you think the two signals are coming from, resonance from the items in the induction furnace? I am reasonably certain such does not exist. A PLL will track a frequency (one, not two), lock onto it and duplicate it, you can use it to receive FM radio signals, and AM signals. You can use it to synthesize a different frequency from a fixed time base. You can also duplicate any frequency you want with a very simple oscillator without using PLL, and it is much simpler.

    BTW, what frequency ranges are you expecting to need? The article in Wikipedia refers to line frequency (50Hz or 60Hz) up to 400Hz or higher.

    Again, it isn't the frequency, it is the power. The same article shows a 600KW furnace. At 240VAC this works out to 2,500 amps.

    http://forum.allaboutcircuits.com/showthread.php?t=51264

    I wish you well, but since I don't have any direct experience I'll bow out. Maybe someone with more experience can help more.
     
  7. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Bill, according to this article, resonance in induction heaters can exist, and parallel resonance can significantly improve the efficiency (power out/power in). If this is true, a PLL can help to keep the heater frequency at resonance by detecting and correcting the phase between the drive current and the voltage across the tank circuit. The loop won't need an external reference frequency input, just a VCO and a phase detector that has as its inputs the current and voltage waveforms.
    Even with heavy loading from the object being heated, the phase between current and voltage (assuming the equivalent circuit is a simple parallel RLC) will still be very close to zero degrees at resonance, although the phase/frequency slope may be fairly flat.
    At resonance, the circulating current in the tank can be much higher than the drive current if the load is light.
     
  8. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Interesting article. Like I said, I'm no expert. They are using a tank circuit to resonate the coil for maximum current, but I am not sure how a phase locked loop would pick up on that. Plenty of oscillator designs would oscillate at resonance automatically. I remember looking at current vs. voltage for resonance on a oscilloscope in my college course. A simple phase comparator wouldn't do it, though if you have something to compare the phase difference between voltage and current to bring the oscillator to resonance that would qualify as a phase locked loop. The phase comparitor is where it is at in this case, it would be somewhat different than the norm.

    The article also mentioned a second phase shift. If the LC circuit is between two H bridges you can control the power being delivered by adjust the phase of the H bridges compared to each other. If they are 180° out of phase then it is maximum power, while 0° is no power. Pretty basic stuff, but interesting.

    Maybe someone with a better background than mine can help out here.
     
  9. sina_always_stuck

    Thread Starter New Member

    Mar 5, 2011
    6
    0
    Hi,
    Thanks for the replies. I thought nothing would be better than visually express my point of confusion. Here I post a schematic and tried to clarify my question. please have a look an let me know if I should provide more details.
    Thank you again
    sina
     
  10. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    So where are you picking off the signals? That was the point of my last post. Phase detectors are easy, you are wanting to find where the resonance point is and oscillate on that frequency. If the resonance frequency shifts you want to track it.
     
  11. sina_always_stuck

    Thread Starter New Member

    Mar 5, 2011
    6
    0
    One of the signals is the cap voltage and the other the inverter voltage. The question is where and how to connect the voltage comparator leads to capture them especially the cap voltage. In respect to your question asking where am I going to pick the signals, I say its exacly what I am looking for. Where in the schematic the voltage comparator leads should be connected? Do we need to put some additional component like an inductor to be able to capture signal A and B?
     
    Last edited: Mar 14, 2011
  12. Experimentonomen

    Member

    Feb 16, 2011
    331
    46
Loading...