BIPOLAR SQUARE WAVE LTSPICE SERIES

And, additionally, the L is in series with the output, the C is in parallel with the output. In my circuit diagram I have to use an LC filter with a resonance frequency much higher than 60Hz (in fact 1 Khz), because i will work with a PWM frequency of 20 Khz. I wonder why the current through the grid is not sinusoidal. Of course your switch frequency of the 2 switches is of course also 60 Hz, so that explains it I assume. I was surprised to see that you use a series LC circuit with a resonance frequency at 60Hz (behaving as a bandpass filter for 60 Hz). The L:C ratio must be chosen so it cooperates with your inverter's output. It indicates power going out onto the grid. This indicates current going contrary to the expected direction based on grid voltage. Notice the upper scope trace has a green voltage sinewave, and a yellow Ampere waveform which is inverted. My +134V goes out on the grid for 1/120 sec. The upper led is lit, indicating positive polarity. My simulation has switches opening and closing in response to mains voltage. I raised and lowered my bipolar DC supply, until I got 1A going onto the grid. Starting with 134 VDC, it can create a shape with peaks greater than 170V (theoretically). This creates a sine waveform from a square wave. The capacitor charge builds, eventually blocking current. The inductor causes gradual rise in current. Not sure if this helps, but in case you're sending a square wave, through an LC filter, onto the grid (which your schematic appears to resemble).ДЄ series LC is needed in that case.

Bipolar square wave ltspice