Switching power supplies

[Alen Hodžič]

Hello!
I am trying to design a flyback converter with the help of this schematic by a well known electronic hobbyist for 30 years Diodegonewild. I will paste the circuit description below.
I have read or should I say download countless switching power supply design books. I would like to design this flyback converter with the well known ic the MC34063. (I know I know....) This chip isn't really suited for this but screw it it is what I have on hand currently... I'm a little bit stuck with trying to copy this design with my chip. The datasheet mentions it that it can be used as a flyback converter. But sadly it doesn't have the calculations for the flyback configuration where the UC3843 has all the calculations in the datasheet. My question is how was this converter build by DANIEL.CZ did he study the datasheet formulas.. Which I think he did... I am searching in about 5 different books about switching power supply design and I don't know what would be the best approach for this... Abandon the MC34063 and stick to the UC3843 or should I read about a dozen SMPS design books to try to figure out how to design a MC34063 flyback converter?
I really want to become good in switching power supply design.. I design the boost and buck converter with the MC34063 but I would really like to design a flyback converter...

How do people become good a these things?

HERE IS THE CIRCUIT DESCRIPTION
   Circuit description: The circuit works as a flyback converter. Maximum output power is 150W. Driving circuit is an integrated circuit IO1 - UC3843. The switching element is a MOSFET transistor T1. Supply operates in discontinuous current mode (DCM), which reduces the reverse recovery loss of diodes D4 and D4'. These ultrafast diodes are used to rectify the secondary voltage. The current is sensed using a current transformer Tr2, because of direct sensing would cause excessive loss. The converter works with over 50% duty cycle. It is good for transformer converters with low input voltage (longer pulse of lower current causes less loss of the MOSFET than the shorter pulse of higher current). Because of duty cycle over 50% converter is equipped with a slope compensation with T2 and R1 (as in datasheet of UC3842-5). Spikes are dampened by the primary "lossless" snubber with D2, C7, D3, L1. The advantage over conventional RCD snubber is a much lower power loss (there's no dissipation resistor) and also reduction of dv/dt for MOSFET T1. As T1 you can use any sufficiently fast MOSFET N with UDS = 75V and more, with the conductive resistance below 5mR. Smaller capacity of gate is an advantage. I used IRFB3207Z. T1 is located at a sufficient heat sink. Diode D2 should have a little cooler, possibly with a common heat sink with T1 or be soldered by its pad on the PCB. Other components are without a heat sink. Ferrite core transformer Tr1 is the central column 12x15mm and the cross section is therefore 1.8 cm2. air gap is 0.8mm at sides and 1.6mm in center column, so it is 2.4mm total. First we wind firs half of the secondary winding, that is 40turns of two magnet copper wires of diameter from 0.35 to 0.4 mm (I have it in 2 layers). Then primary winding, that is 6 turns of fifteen wires of diameter 0.5 mm (3 layers, wound using five wires in each). Finally, the second half of the secondary winding as well as the first. Each Windings and winding layers are isolated. Halving the secondary reduces the leakage inductance. L1 is iron dust choke, inductance is not critical. You can also use the finished coil from switching power supplies or active PFC. Diameter of wire can be any of about 0.8 mm or more. Current transformer TR2 is on a small ferrite ring and has a 1 turn in primary and 60 turns secondary. Core dimensions are not critical. If the number of secondary turns was different, you can adjust the value of shunt resistor R2. The working frequency of the inverter is about 40 kHz. Efficiency is more than 90% at full power. Note: For 220V~ mains the amplitude is 310V, for 240V~ mains the amplitude is 340V.

[bmxsesh]

You probably get a DIP coil former like E13/7/4 TDK with 2.54mm leg pitch and wind it by hand, then build a circuit on a breadboard until you get a desired result after that you put that circuit on a PCB, if it's your first experience with transformers
I'm about to make a flyback transformer with LM2588 switching regulator for generating +15V,+10V,+5V,-7.5V  and will probably take this route
Since this is a power circuit the parasitic inductance would not matter

[Berni]

Design of switching power supplies is easier to you first study the basics of the particular topology. Then the datasheet is just part of figuring out how to make the chip give you the correct timings for that topology.

Tho in general mains powered 50 to 400W single output switching power supplies is something that the Chinese are REALLY good at. So there is not much worth in building your own as a good deal of particular skills are required, only for you to in the end spend 3x the price on just components as opposed to simply buying a PSU module from a well regarded name brand like Meanwell.

Where switching power supplys are worth designing your own is when you need to turn 12V into 5V or 3.3V or something on your board. There you can nicely integrate the power supply onto the same board without needing a module. But a 220V supply will usually be on its own PCB anyway. So you might as well just buy it as a module, and have it come with a nice metal shielded case to keep down EMI and prevent people from getting electrocuted if they poke around inside.