switch mode power supply charger

Thread Starter

Dabu WhiteHacker

Joined Sep 5, 2017
68
i opened up a switch mode power supply and got the following circuit
Schematic_charger_Charger_20181216123046_Dabu WhiteHacker.png

after searching on internet on internet on how swmp works i got following idea about it
asd_Dabu WhiteHacker.png

but sill there are some things that i could not understand
1. why there is a diode on output, isn't the output already rectified because as i know output is high frequency dc pulse.
2. why there are two starting diode R9 and R10 instead of single resistor
3. why there are R6 and R7 on output of transistor
4. i really don't know in detail on how snubber network, feedback system and Ring Choke Oscillator but i will search them myself so if you could only give only on what is what it would be very helpful
 

Dodgydave

Joined Jun 22, 2012
11,285
The two resistors R9, R10, are probably the same value and maybe it was cheaper or easier to use these values at the time of design, yes one resistor would do the same job.

The diode D3 is the Half Wave rectifier with C5 as the smoothing capacitor.
 

Thread Starter

Dabu WhiteHacker

Joined Sep 5, 2017
68
The two resistors R9, R10, are probably the same value and maybe it was cheaper or easier to use these values at the time of design, yes one resistor would do the same job.

The diode D3 is the Half Wave rectifier with C5 as the smoothing capacitor.
as i mentioned earlier, transformer is outputting high frequency dc signal then why do we need diode if current is following in only one direction as current in dc voltage only flows in one direction
 

ebp

Joined Feb 8, 2018
2,332
First, that thing you call a transformer (and almost everyone does, including me) is not a transformer in the normal sense. In a true transformer, current in the primary winding flows simultaneously with current in the secondary. There is no significant energy storage, although there is some.

This circuit works very differently. The "transformer" is actually used as an inductor with two windings. During part of the switching cycle when Q1 is on, energy is being stored in the inductor. No current flows in the output because of the diode on the secondary side. Without the diode, there would be true transformer action and charge would be removed from the output capacitor during this interval. When Q1 turns off, the stored energy flows through the diode and into the output capacitor and the load. Ideally, 100% of the stored energy can be delivered to the output side, but due to "leakage inductance" (to me it makes more sense to think of this as magnetic flux that isn't coupled perfectly rather than "leakage") some of the stored energy can't be delivered and makes a very high voltage spike on the input winding. The snubber components are there to dissipate (waste!) enough of this energy to keep the voltage across the transistor from being destructive.

While the main intent of the magnetic component is to store and deliver energy, true transformer action does exist and is used to make the circuit self-oscillate. The extra winding that connects to the base of the transistor plays a critical role. When the transistor is starting to turn on, the polarity from that winding is such that it reinforces the drive for the transistor and turns it on harder. Eventually the magnetic component begins to "saturate" - the magnetic flux is no longer able to increase. When this happens, the base drive winding stops reinforcing the base drive and the transistor begins to turn off and the magnetic field begins to "collapse." This reverses the polarity on the base drive winding, driving the transistor to switch off rapidly. When the stored energy has been delivered on the output side, the transistor can again start to turn on and the cycle repeats. Feedback from the output side via the optocoupler influences the points at which the transistor turns on and off. The optocoupler input is driven via a zener diode. This makes a simple but not very good voltage regulating circuit.

The more sophisticated version of this circuit is the flyback converter where pulse width modulation at a fixed frequency (usually) is used and the magnetic component is not driven to saturation. The high peak magnetic flux associated with saturation causes excessive losses in the magnetic core and makes the method generally unsuitable for higher powers. It would work, but the "transformer" would run very hot.

If you look around the web, you'll find articles on the Apple little cube chargers and the imitations. The Apple charger uses a very well designed PWM'd flyback circuit with very good regulation. The cheap knock-offs use the crude self oscillating circuit (and horrendous, dangerous construction!)
 
Top