Processing technology of high frequency transformer of switching power supply

- May 14, 2019-


1 the introduction
In switching power supply, high frequency transformer is an important component for energy storage and transmission. A high-frequency transformer should have the characteristics of small leakage inductance, small distributed capacitance of windings and small coupling capacitance between windings. This paper describes how to reduce the leakage inductance and the capacitance of the coil and improve the reliability of the switching power supply by improving the processing technology of the high-frequency transformer.
Primary winding leakage inductance and distributed capacitance

In the design of high-frequency transformers, leakage inductance and distributed capacitance of the transformers must be minimized, because high-frequency transformers transmit high-frequency pulse square wave signals in the switching power supply. In the transient process of transmission, leakage inductance and distributed capacitance will cause surge current and peak voltage, as well as top oscillation, resulting in increased loss. Although increasing the clamps and absorption circuits on the drain of the switching transistor can overcome the peak voltage, the excessive peak will lead to an increase in the loss of the clamps and absorption circuits, which will reduce the efficiency of the switching power supply and lead to the damage of the power switch tube in serious cases. Normally the leakage inductance control of the transformer is 1-3% of the primary inductance.

2.1 leakage inductance of primary windings
Leakage inductance of transformer is caused by incomplete coupling of magnetic flux between primary winding and secondary winding, layer to layer and turn to turn. The following measures can be taken in the winding process of transformers.

Minimize the number of turns of the winding, choose high saturation magnetic induction intensity, 

low loss of magnetic materials.

Increase the ratio of winding size height to width.
The insulation thickness between windings should be reduced as much as possible, but the transformer itself must have sufficient insulation strength.
The primary and secondary windings (sandwich method) are wound by layering and crossing.
When the ring type magnetic core transformer is used, no matter how many turns the primary and secondary windings have, the winding is evenly distributed along the circumference of the ring type. For the ring core transformer with large current, the multi-winding parallel winding is adopted, and the wire diameter is reduced as much as possible.
Improve the degree of coupling between windings.

In the case that the input voltage is not too high, the primary and secondary windings are processed by double winding.The ratio of decreasing the number of turns of primary winding and increasing the height and width of winding size is related to the selected core shape. If the size of the central pole of 

the winding is long enough for the primary winding to be wound into two layers or even one layer, the primary leakage inductance and distributed capacitance can be effectively reduced. The high frequency transformer is suitable to adopt the magnetic core with long central column, but not the one with short and fat shape. In the measures for transformer winding turns not too little, or when the input voltage is too high, or when the pulse is too wide, can lead to core saturation, lead to the transformer winding inductance sharply reduce, winding current limiting effect of alternating current is reduced, serious when the state of short circuit, in microseconds, dozens or even hundreds of ampere of electric current through a semiconductor device, failure.

2.2 distributed capacitance
Distributed capacitance between turns of transformer windings, between upper and lower layers of the same windings, between different windings, and between windings and shielding layer (or magnetic core). The distributed capacitance of switch transformer is mainly composed of the following parts.
The distributed capacitance between the windings and the shielding layer (or core).
The distributed capacitance between turns of each winding.
Distributed capacitance between windings.
Distributed capacitance between upper and lower layers of each winding.
During the on-off and on-off of the transistor of the switching power supply, the distributed capacitance of the winding is repeatedly charged and discharged, and its energy is consumed by the clamping and absorbing circuit, which reduces the efficiency of the switching power supply. In addition, the distributed capacitance of the windings also forms LC oscillation together with the leakage inductance of the windings, producing ringing noise. To reduce the distributed capacitance, the following measures can be taken:
The windings are wound in segments.
Correct arrangement of the polarity of the windings to reduce the potential difference between the windings.
Electrostatic shielding measures shall be added between primary and secondary windings.

Select the number of magnetic flux leakage potential groups M=4.

3 winding processing
3.1 primary winding
The primary winding should be wound in the innermost layer, so that the primary winding of the transformer can use the shortest wire length for each turn, so that the entire winding of the wire for the minimum, which effectively reduces the primary winding itself distributed capacitance. Under normal circumstances, the primary windings of the transformer are designed to have two or less layers of windings, so that the leakage inductance of the transformer is minimized.
The primary windings are wound in the innermost part, so that the primary windings are shielded by other windings, which helps to reduce the coupling of the primary windings with the electromagnetic noise of adjacent devices.

The primary winding is wound in the innermost part, so that the starting end of the primary winding can be used as the drain or collector driving end connecting the power transistor of the switching power supply, which can reduce the coupling of the transformer primary to the electromagnetic interference of other parts of the switching power supply.

3.2 secondary winding
After winding the primary winding, it is necessary to add winding (2-4) layer of insulating tape to rewind the secondary winding, so as to reduce the capacitance distributed between the primary winding and secondary winding, and increase the insulation strength between the primary and secondary winding, meeting the requirements of insulation voltage. Reducing the capacitance between the primary and secondary of the transformer is beneficial to reduce the common mode interference at the output end of the switching power supply.
If the secondary of the switching power supply has multiple outputs and the outputs are not common, in order to reduce the leakage inductance, make the secondary with the maximum power close to the primary windings of the transformer. If the secondary winding has only a relatively few turns, it should be covered with a complete layer in order to improve the coupling. If multiple enameled wires can be used in parallel, it will help improve the filling coefficient of the secondary winding. Other secondary windings are tightly wound around this one.
When the common ground technique is used for the multiplex output of the switching power supply, the processing method is simpler. The secondary can be output in the form of transformer tap, and there is no need to adopt insulation isolation between the secondary windings, so that the winding of the transformer is more compact, and the magnetic coupling of the transformer is strengthened, which can improve the voltage stability under light load.


3.3 bias winding

The bias winding is wound between the primary and secondary, or in the outermost layer, and the switch power supply adjustment is based on the secondary voltage or primary voltage. If the voltage adjustment is based on the secondary, the bias winding should be placed between the primary and secondary, which helps to reduce the transmission of conduction interference generated by the power supply. If the voltage adjustment is based on the primary, the bias windings shall be wound around the outermost layer of the transformer, which allows maximum coupling between the bias windings and the secondary windings and minimal coupling between the primary windings. The primary, bias winding had better be full of a complete layer, if the number of turns of the bias winding is very small, you can take the line diameter of the bias winding, or multiple enameled wire parallel winding, improve the filling coefficient of the bias winding. In fact, this improvement also improves the shielding ability of using secondary voltage to regulate the power supply, as well as the coupling of secondary windings to bias windings when using primary voltage to regulate the power supply.

The windings are thinly wound and tightly wound
What do we do when the winding can't wrap a full layer? It depends on the distribution of the windings. In general, the output voltage is more stable if the windings are scattered. Moreover, the output voltage is slightly higher and the carrying capacity is stronger. In the case of loose winding, the effective coupling coefficient is higher than that of tight winding, because in the case of tight winding, part of the magnetic energy is not coupled with each other, and the leakage inductance is larger. Now let's talk about 3 different cases of wire winding.
The primary winding is not covered by one layer
When the primary winding is not wide enough, it should be tightly wound, which is beneficial to stabilize the output voltage and enhance the carrying capacity.
B. The secondary winding is not covered by one layer
When the secondary winding is not enough width all want to thin winding and uniform, such as secondary as far as possible around or more than one or two circles when reducing the wire diameter around full, avoid more than one or two circles caused by the transformer difficult to produce and also affect the leakage.
C. The feedback winding is not satisfied with one layer
When the feedback winding is less than half of the width, please select the middle close winding. The density winding of the feedback winding affects the feedback voltage and EMI of the power supply. Because the feedback windings are sensitive to EMI, the winding process of feedback windings should be considered in the rectification of the switching noise EMI of power supply. If the feedback winding needs to be unwound with shielding, it is recommended that the power supply engineer wrap it with multiple strands.