Prerequisites:

The recommended input voltage is 12~18V. The hardware protection range is 10~20V, beyond this range the power supply will stop working.

Principle explanation:

The circuit topology is similar to a two-tube forward,

Connect the 2596 output to the gate of the MOS tube so that the MOS tube is turned on at the same time and the inductor stores energy.

When the 2596 is turned off, because the inductor current cannot mutate, the gate voltage is pulled down, D1 is turned on and clamped to -0.5V, the MOS tube is turned off, and the inductor current is output through D1 and D2.

D3 and D4 are inductor reverse current clamps.

TL431 generates a voltage reference and cooperates with LM393 comparison to make the circuit work within the specified power supply voltage range to protect the MOS tube.

Add a picture to prove that it’s not just me taking off my pants and farting:

Production process:

Computer test:

Input 12V, output 5V2A

Output ripple

MOS tube gate-source voltage waveform

Test Data:

The ripple in the above table is estimated to be 50% off. I forgot to turn on the bandwidth limit. I measured it again and found that the ripple was reduced by more than half.

I took a look at Xinlong's PDF and found that the official sepic topology efficiency of xl6009 is around 80%. I guess if I replace it with a mos tube buck chip, the efficiency will be on par with him.