Hello again!
First let me thank those that helped me with my monostable solenoid circuit. It works great. My garden thanks you! That circuit is another building block for this one.
My next adventure will be an inductive annealer for brass casings. I am trying to find the resonant frequency for the heating coil. First, I cannot use my oscilloscope from all the hash coming from the lights, so I fell back on using a trick with a VTVM to "Peak" the output from my signal generator.
I built a tank circuit using a small resistor sized 220uh inductor, and a 10nf capacitor. I used a 10uf coupling cap to isolate the generator, and a 10 ohm resistor to measure the output from a 12v p-p signal. Viola! I got a 3.5 volt deflection to peak as I swept the frequency. 104 Khz.
Next, I calculated my 10 gage wire coil at 376nH, then used a 10uF cap to make a tank and swept it around the Resonant frequency (82 Khz) and got no deflection at all?? What am I missing? The schematic shows the coil being fed from an H bridge configuration and the chip will be a IRS2453. other than grabbing a LCR meter in case I missed the resonant frequency, what else does anyone think?

 

The resistance of your coil is so very low that the Q factor is very high, and, in addition, your resonant frequency may be much different than calculated. AND, use a resistor to isolate the tuned circuit from the generator. I suspect your resonant frequency is much higher.

 

I deleted my last comment because, as Mr Bill stated, the Q is extremely high, and when I tried to ring a similar tank, I failed miserably.

Why not just put a test brass in the coil and adjust the frequency until max current is pulled from the power supply?

 

I just found the best writeup ive ever seen on induction heating...
https://www.richieburnett.co.uk/indheat.html

 

That is indeed a good write up!!
One option might be to use the work coil as the frequency determining element in a power oscillator scheme. The main hazard that I see is generating huge amounts of interference if the frequency is in use by others. The benefit is that the drive would always be at the resonant frequency. The caution is that the voltages and currents will be at hazardous levels even at just 100 watts of heating power.

 

Thanks for the replies. I haven't built the rest of the circuit yet. Only setting up the building blocks. So I really have no idea if it will even have the power to do what I need yet. I was also suggested not to go the H-bridge rout by another who has built heaters before. He suggested a UC3525AN which I just happen to have. That gentleman also tagged that same site. It is one of the best sources of information I have read yet! Thanks.
Right now, I have breadboarded the circuit after this circuit with the exception of no feed back or PWM control. Just fixed Rt,Ct and Rd The results are what I expected, however I'm getting some pretty bad ringing from both outputs. Any ideas what it causing it and how to stop it? The power supply leads are long, but I tried putting a 80UF cap across the rails to see if that was the issue. It made no difference. The overshoot on a 10V sguare wave is 28 volts. and dips down to around 4 volts.
I'm afraid this will raise havoc on my power Mosfets.

 

Thank you. That was the purpose of the VTVM. A strong increase in voltage across the load resistor indicated the same thing. You can actually continue to lower the voltage and "Peak" the frequency to identify the exact frequency. Thats how radio IR coils are aligned on the old radios.

 

WHERE in the circuit of post #6 are you seeing those wave forms?? Are they in the simulation?? or on the breadboard??
That kind of overshoot in a non-inductive circuit is often due to delays in the feedback loop, and suddenly it gets complicated. The circuit shown does not make sense with the waveforms shown.

 

That is the actual output from the breadboarded circuit. Even the outputs do not match each other. I even swapped the scope leads to see if it would change. It didn't
There is no feedback in that circuit, I left it out. Could that be the problem?

 

The lack of stabilizing feedback can certainly allow a lot of distortion, including instability. So now it is obvious that the circuit does need the feedback to function properly.

 

The lack of stabilizing feedback can certainly allow a lot of distortion, including instability. So now it is obvious that the circuit does need the feedback to function properly.

Thank you! I put a 12V zener on the out put and it seemed to help a lot. If I cannot get the ringing manageble, that should help.

 

I was assuming that the whole circuit was assembled, which I do not see a version of the whole circuit anywhere in this thread. And the one circuit I do see does not appear to include any feedback circuit, but possibly a DC substitute, which will do nothing at all to reduce overshoot spikes.