I am designing 24V_5A SMPS circuit by keeping reference of a mobile charger circuit...when I power on the circuit the IRF840(power Mosfet) blasted . What is the reason...Is there any mistake in the circuit diagram..?

 

Please reload the schematic image. The resolution is to low to see what parts are being used. (i.e. to blurry to read).

 

On first look , the second primary windings are shorted, but a better picture would be great.

 

I think it will be clear now...kindly correct the circuit and pls give the explaination

 

The Mosfet isn't getting a turn off signal and is practically shorting out the supply via the primary winding, where did you get this circuit from?

 

Mosfet is getting its turn off signal via bc547 transistor(which would turn off through current sensing resistor)

 

Reference link: https://danyk.cz/impulz3_en.html

 

Reference link:

Your circuit is missing a bunch of components compared to the reference.

 

Your circuit is missing a bunch of components compared to the reference.

We have not used optocoupler for voltage regulation....instead we have opted for LM338 regulator

 

instead we have opted for LM338 regulator

So where is that on the schematic diagram?

 

We have not used optocoupler for voltage regulation....instead we have opted for LM338 regulator

in the absence of regulation, a flyback circuit transfers the same amount of energy each cycle, increasing the energy stored in the capacitor, and increasing its voltage until something fails.

 

in the absence of regulation, a flyback circuit transfers the same amount of energy each cycle, increasing the energy stored in the capacitor, and increasing its voltage until something fails.

Is it ok to use lm338 instead of optocoupler as both are doing the same function?

 

This design relies heavily on the -ve feedback from the Aux secondary winding and the combination of the BA159/47u/2k2, whereas you just have an 1N4148 shorting out the feedback winding. Your feedback winding polarity relative to the output winding polarity is reversed - I suspect you aren't getting the feedback turn-off pulse this circuit requires. Did you wind the transformer yourself or are you using an off-the-shelf one?

 

Is it ok to use lm338 instead of optocoupler as both are doing the same function?

No they aren't - the optocoupler plays a key role in capping the voltage generated at the secondary whereas your approach will see significant open-circuit voltages developing at the secondary especially on low loads.

 

Looking at the reference circuit, there's a 3 turn winding which feeds back to the gate of the MOSFET to switch it off. Notice the dot on the winding, denoting the correct phasing (important on all windings).
Without that secondary I can't see how the circuit would oscillate correctly.

 

This is the final circuit with regulator

 

Is it ok to use lm338 instead of optocoupler as both are doing the same function?

No, the reason why they use an optocoupler is to isolate the output from the primary side, and it's a faster way to switch off the transistor.

Your circuit doesn't have any voltage feedback regulation , so the fet will oscillate at maximum output and then burn up.

 

This is the final circuit with regulator
View attachment 294348

On low load the voltage at the input of the LM338 will keep increasing until the LM338 (or the capacitor) fails.
Another point is that the LM338 is a linear regulator - any excess voltage will be dissipated as heat, making it rather pointless to use a switched-mode circuit in the first place.

 

Let’s talk about the elephant in the room: the design of the flyback transformer.
You have not discussed, at all, the transformer details. A line-powered flyback transformer design is FAR from being trivial. More so if you choose a self-oscillating topology.

 

Agreed. Why try to get a self-oscillating circuit working when there is an almost infinite choice of control ICs?