Simple 12V, 1A SMPS

Most electronics enthusiasts require DC power supplies to operate various devices and
accessories. The most popular and common supply is a 12V DC supply that can be easily derived
from the household AC supply with transformation, rectification, filtering and stabilization. These
power supplies have a bulky steel- or iron-laminated transformer that provides a safety barrier
for the low-voltage output from the AC input, and reduces the input from typically 230V AC to a
much lower voltage. The low-voltage AC output from the transformer is then rectified by two or
four diodes and smoothed into low-voltage DC by large electrolytic capacitors.

A switched mode power supply (SMPS) offers the same end results at a lower cost and higher
efficiency. For a given output power, an SMPS is lighter and smaller. This is because, if the
frequency of operation is increased, one can get away with using a smaller core cross-sectional
area. Besides, an iron-core transformer works only up to about 10 kHz, and if we need something
in 50-100kHz range, we need a ferrite core.

Circuit and working

Fig. 2 shows the circuit of a simple 12V, 1A SMPS. The circuit is built around a low-power offline switcher
TNY266 (IC1), photo-transistor photo-coupler EL817 (IC2), a flyback transformer (X1) and some other easilyavailable components.
Low-power offline switcher (TNY266). The SMPS here has been designed using a TNY266 chip, which is
affectionately called the 555 of SMPS. This device has a 700V power MOSFET, an oscillator, a high-voltage
switched current source, a current limiting and thermal shutdown circuitry integrated onto a monolithic device.
The start-up and operating power is derived directly from the voltage on the drain (pin 5), eliminating the need
for a bias winding and associated circuitry. In addition, the device incorporates auto-restart, line under-voltage
sense, and frequency jittering.
The drain-source breakdown voltage of the MOSFET in TNY266 is important. During the off period, the
MOSFET sees rectified 317V DC approximately. Additionally, it sees the reflected voltage of the secondary,
which is about 130V AC. It also encounters the ringing voltage from the leakage inductance, and the drainsource capacitance of the MOSFET. Therefore a MOSFET with a Vdss of 650V DC is expected to keep the
necessary safe operating margin. Fortunately, a MOSFET with these properties is included in TNY266.
The 230V AC input is connected at CON1, which is rectified by diode D1. The neon lamp (NL1) glows when
the input supply is present. Resistor R1 limits the current through the lamp. The rectified output goes to the first
terminal (A) of coil L1 and the second terminal (B) is connected to the drain of the inbuilt MOSFET in IC1.
Diodes D2 and D3 are essentially the snubbers, and are used to protect the MOSFET from going above 600V.
Flyback transformer. A flyback circuit is simply a pair of coupled inductors. If a current is passed through one
inductor, it will store energy E = (L.I2), where L stands for inductance in henry, and I for current in
amperes. This energy can later be taken out of the second inductor, which is coupled to the former at a different
volt-current ratio. The flybacks energy storage and extraction mechanism is interesting. The key point is the

polarity of the winding; the secondary is out of phase with the primary, as is evident in Fig. 2 (the dots indicate
polarity).

When the MOSFET of IC1 is closed, the current flows through L1. Point A on L1 goes positive and
by transformer action, and considering the polarity of dots, point C on L2 goes negative. This
reverse biases diode D4, and no current flows in the secondary winding. Similarly, when the
MOSFET is open, the current flow through L1 is interrupted and, by Lenzs Law, a voltage of
polarity opposite to the applied voltage appears on L1 and L2. Thus, point A on L1 goes negative
and point C on L2 goes positive. This situation forward biases diode D4. The energy stored in the
core causes the current to flow through winding L2. This charges capacitor C2 and also powers
the load. The charge on C2 is used in the next half of the cycle to keep the current through the
load somewhat constant. The cycle repeats endlessly. The MOSFET is switched on/off
continuously at a frequency of around 120 kHz to keep this process running.
The design data for the transformer is as follows:
1. Duty cycle = 0.45 (max. duty cycle for DCM fly back = 0.5; less 10% safety margin)
2. Core saturation magnetization Bsat = 0.24T
3. Core area of EE20 = 25mm2
Winding details computed for the SMPS are shown in Table I.

The feedback circuit. Regulated output needs feedback to control the pulse width modulation (PWM) of the
MOSFET. TNY266 has a fabulous control feature; it stops the switching cycle as soon as any current is taken
out of pin 4 of the device. If SMPS output exceeds the zener break-down voltage then ZD1 conducts. This lights
the opto-LED and the opto-transistor grounds pin 4 of IC1, resulting in immediate stoppage of the switching
cycle. Also, when the primary is conducting, diode D4 on the secondary side is reverse biased. At this time, if
the voltage across D4 exceeds its reverse breakdown voltage, the SMPS will fail. Here we have used an SB160
Schottky diode with breakdown voltage= 60V.
Connector CON2 provides 12V regulated DC supply.

Construction and testing

A general core-selection rule of thumb for SMPS below 50 watts is 2-3 mm2 core area per watt. For a primary
input power of 16W, a core with a core area of 32-48 mm2 is needed. EE20 core will work well for this design.
Transformer wire.
Any wire that can carry the required current can be used. To know how much current a given wire can carry,
SMPS designers use a number called current density [J] for this calculation. Empirically, a good starting point is
J = 5 amps/mm2. The primary carries 0.3 amps, so it can be wound using SWG 38. The secondary carries 5
amps, so it can be wound with SWG 26. A good practice would be to wind the secondary using two parallel
strands of SWG 28. This reduces the skin effect. The key problem in flyback transformers is leakage
inductance, which is caused by poor coupling between the primary and secondary windings. So wind them
tight, with full overlap.
The air gap.
Flybacks made from power ferrites must have an air gap. The energy stored in a flyback primary is E = L.I2.
Peak primary current depends inversely on primary inductance. An air gap increases the energy storage capacity
of a flyback transformer. It is calculated as follows:
Air gap = (0 x N2 x Ac)/Lprimary
where:
0 = permeability of free space, 4 x 10-7
Ac = Core area (m2)
N = Primary turns
Lprimary = (Vprimary.pk Ton.primary)/ Iprimary.pk

The primary inductance computes to ~3mH, so the air gap is approximately 0.2mm. However, this value is not
critical, as was experienced by the authors during their experiments. Any air gap in the neighbourhood of the
calculated value works well. A thin sheet of plastic or paper works fine.
An actual-size, single-side PCB for the simple 12V, 1A SMPS is shown in Fig. 3 and its component layout is
shown in Fig. 4. Assemble the circuit on the recommended PCB to minimise assembly errors. Use IC base for
IC1.
To test if the circuit is functioning properly, first check the regulated output at TP1 with respect to TP0. The
voltage should be stable with or without a load.