I think we can pulse width modulate the square oscillator to change the sine amplitude.

Is that correct?

I'm not sure if this paper represents a normal resonant device or something different.





https://www.semanticscholar.org/pap...ayeb/241ddd66a373f1a5792d3708f6f95e99d083db75

That seems to be demonstrated by this Falstad sim, tho' i'm not sure my circuit is correct. Surprisingly, i get full sine amplitude at 50% duty cycle, and lower amplitude below or above 50%. Except, above 50% i seem to get DC offset on the sine. Also weird is that if i remove the R, then changing PW seems to have no effect on sine amplitude.
https://tinyurl.com/22j4coqr

Are there other ways to regulate voltage of a resonant inverter?

 

I think we can pulse width modulate the square oscillator to change the sine amplitude.

Is that correct?

I'm not sure if this paper represents a normal resonant device or something different.

View attachment 292779

View attachment 292780

https://www.semanticscholar.org/pap...ayeb/241ddd66a373f1a5792d3708f6f95e99d083db75

That seems to be demonstrated by this Falstad sim, tho' i'm not sure my circuit is correct. Surprisingly, i get full sine amplitude at 50% duty cycle, and lower amplitude below or above 50%. Except, above 50% i seem to get DC offset on the sine. Also weird is that if i remove the R, then changing PW seems to have no effect on sine amplitude.
https://tinyurl.com/22j4coqr

Are there other ways to regulate voltage of a resonant inverter?

Generally it is regulated by adjusting the frequency away from resonance, and keeping the duty cycle at 50% (because that avoids the DC offset).
It is generally detuned above resonance because the load then looks inductive making it possible to achieve better efficiency by zero voltage switching .

 

 

Generally it is regulated by adjusting the frequency away from resonance, and keeping the duty cycle at 50% (because that avoids the DC offset).
It is generally detuned above resonance because the load then looks inductive making it possible to achieve better efficiency by zero voltage switching .

I see! Also, if you reduced the frequency, then harmonics could start to seep in.
But in my simulation above, i 'm seeing distortion in the current waveshape. My circuit is prolly wrong.
Another concern is that pushing the f up may exceed the frequency rating of the inductor.

 

I see! Also, if you reduced the frequency, then harmonics could start to seep in.
But in my simulation above, i 'm seeing distortion in the current waveshape. My circuit is prolly wrong.
Another concern is that pushing the f up may exceed the frequency rating of the inductor.

By the time you have attached a rectifier, the waveshape is not that good at the best of times.
Most LLC circuits I have seen operate between 50kHz and about 250kHz. Make sure that your inductor’s self-resonant frequency is well above that value.
Also, the resonant inductor is seldom a real physical inductor, it is the leakage inductance of the main transformer.

 

I don't understand the purpose of the 2nd inductor. Is it to create a resonance at a 2nd frequency?

 

Also, the resonant inductor is seldom a real physical inductor, it is the leakage inductance of the main transformer.

In an inverter, why is a transformer needed?
Step up/down?
Or isolation?
Or something else?

i ask because my application doesn't need isolation from mains. The DC source that feeds our inverter already provides isolation from mains.

 

Isolation. LLC resonant is generally used for off-line power-supplies >100W.

 

Isolation. LLC resonant is generally used for off-line power-supplies >100W.

LLC design seems so simple and organic. It's high efficiency and low EMI. Why not use it for <100W?

By the time you have attached a rectifier, the waveshape is not that good at the best of times.

Where is a rectifier in an inverter?

 

LLC design seems so simple and organic. It's high efficiency and low EMI. Why not use it for <100W?

It's low-load efficiency is poor, and it will not regulate with no load.
Flyback converters regulate better at low load, and only need one power transistor.
A LLC will struggle to meet the "standby power" regulations, whereas a flyback will pass them.

 

A LLC will struggle to meet the "standby power" regulations, whereas a flyback will pass them.

Is that true of all resonant inverters? Or just LLC?

Survey of Resonant Converter Topologies (ti.com)

 

Is that true of all resonant inverters? Or just LLC?

I would suspect that it's all of them. At a guess, the amount of energy circulating in the resonant circuit remains reasonably constant, and losses are a fairly constant proportion of that energy.

 

A LLC will struggle to meet the "standby power" regulations, whereas a flyback will pass them.

Do losses increase with frequency? Or decrease?

 

Do losses increase with frequency? Or decrease?

That's complicated.
Transistor switching losses are approximately proportional to frequency.
Inductive core losses vary with the square root of the cube of frequency, and the square root of the fifth power of peak to peak flux excursion, which in turn will increase as frequency decreases.
Then there's losses from the skin effect and proximity effects, which increase with the square root of frequency.
Overall, it depends on which one dominates.

 

Transistor switching losses are approximately proportional to frequency.
Inductive core losses vary with the square root of the cube of frequency, and the square root of the fifth power of peak to peak flux excursion, which in turn will increase as frequency decreases.
Then there's losses from the skin effect and proximity effects, which increase with the square root of frequency.

Is this correct?

• Transistor losses increase with frequency.
• Inductive core losses decrease with frequency.
• Skin and proximity effects increase with frequency.
• Capacitor losses increase with frequency?

 

Is this correct?

• Transistor losses increase with frequency.
• Inductive core losses decrease with frequency.
• Skin and proximity effects increase with frequency.

Yes. I haven't even thought about capacitor losses, but would expect them also increase with increasing frequency.