How to increase the peak voltage of a TL494 MSW inverter under load?

[Samueru]

Hi, the Inverter is a Bestek 1000W 12V DC to 110V AC modified sinewave. I use it with a solar setup.

The Inverter works like as shown here:

The input voltage range goes rom 9.3V to 15.V

The battery side H Bridge uses 8 IRF1404. The switching frequency (if my cheapo DSO scope is anything to go by) is 35 kHz.

This H bridge is controlled by 4 small transistors in a daughter board with a TL494 and a LM358, the later I assume does the over voltage/under voltage and overcurrent protection.

The transformer are 3 transformers in parallel, '''MRI10011-115V-02''

The high voltage DC side uses 8 IRF640 for its H bridge.

This inverter, with a 15V input produces 170V peak, the peak voltage changes accordingly to the input voltage, the second I load it to 9W it drops to 160V and that's the problem I have with the inverter, under usage with a battery and moderate loads the peak voltage drops to as low as 122V, to compensate for that the output H bridge increases its duty cycle to maintain 110V RMS (turning it into a square wave), however with the peak voltage that low several SMPS appliances that I have refuse to work.

Normally for example if this was a TL494 based ATX power supply I would intercept the feedback between primary and secondary to change the output voltage, but given that there's no feedback between those (No optocoupler and no extra transformer winding in sight) I have no idea where to start.

There's a youtube video of a guy changing a MSW inverter to Pure sinewave, we had to modify the transformer to increase the peak voltage of the inverter for the modification, and later also had to add feedback to the battery side TL494 because it had issues under load, he used an arduino for that.



Given that the no load peak voltage in my case is high enough since I don't need a pure sinewave, how could I go adding feedback loop without an arduino is this case?

Or if that is too much work, how could I increase the fixed duty cycle a little so that the no load voltage stays at about 185V peak to minimize the issue under load?

Edit: I already tried increasing the high voltage DC bus capacitance and it made no difference.

[Zero999]

Please post a schematic.

Are the high side MOSFETs getting enough gate drive?

[Samueru]

Please post a schematic.

Whole thing? I cant.

Are the high side MOSFETs getting enough gate drive?

I just took a closer look into the inverter and I realized a few things are different. Tell me if I should still take at look if the fets are getting enough gate drive (which I assume is check that the gate voltage doesn't drop) after you read this. The efficiency of the inverter hits 91% with a 250w load. The transformers are always cold to the touch.

I took several pics of the inverter with details: https://imgur.com/a/K6bzuW2

The 3 transformers have a primary with center tap, each transformer has 2 YMP1404L fets in a push pull configuration with +12V going to the center tap. Total 6 mosfets and not 8.

No idea what the blue potentiometer does, It came set to 193 Ohm, changing the value doesn't affect the output voltage.