Sharing: Introduction to the RCC circuit principle and discussion of problems from basic to complete

Working waveform of each pin of the switch tube

Assuming that the amplification factor of Q1 is A and the base drive current is B, when the current of L T2A (that is, the collector of Q1) increases to A*B, it stops increasing, the magnetic flux of L T2A stops changing, the induced voltage of L T2B decreases, and the positive feedback current from L T2B to the base of Q1 decreases. In this way, the base current of Q1 begins to decrease, the collector current also decreases, and the voltage of L T2C reverses. Secondary output voltage. Is the operating frequency of the RCC converter determined by the R and C at the feedback (bias winding) end?

Many people think that the oscillation frequency of RCC is related to this capacitor and resistor. They use the RCC converter composed of three-pole tubes to illustrate this. According to the analysis of general circuits, after power is applied, the switch tube starts to conduct, the feedback injection base IC increases, VCE decreases, the primary coil voltage increases, the feedback winding voltage increases, and the current injected into the base increases. This feedback is a positive feedback, and the switch tube is saturated and turned on soon. The current in the primary inductor increases linearly. As the feedback capacitor is charged, the voltage established above will reduce the current injected into the base. When it is reduced to the point where the switch tube is pushed out of saturation and enters the amplification area, VCE increases, the primary coil voltage decreases, and the feedback winding voltage also decreases, so the current injected into the base decreases. Soon, the switch tube is cut off, and then the energy is transferred to the load through the secondary side. Therefore, it is believed that the oscillation frequency is related to the value of RC. However, according to the law of conservation of energy, Ef=VO*IO;1/2 *Li2=VO*IO/f; after more detailed calculations, it is found that there are no two terms R and C in it. Since the formula has been proved, it is obvious that there is no relationship. The explanation is as follows: In the first cycle of power-on, it is obvious that the oscillation frequency of RCC is completely determined by this RC value. Because the voltage has not yet been established on the output capacitor, the control circuit does not work. When the output voltage is established, the control circuit will detect the output voltage and control it. Generally, it is through an optocoupler and then divides part of the current injected into the switch tube. Therefore, the current injected into the switch tube should be the sum of the feedback branch and the current divided by the control. When the circuit reaches a steady state, as long as the input voltage remains unchanged and the load current does not change, the peak current (0.5L*i2) flowing through once remains unchanged. The circuit must have only one stable state corresponding to it, otherwise, such a topology is unreliable. Since the peak current flowing through once is fixed (RCC works in critical mode), even if you increase the feedback capacitance a little, the control circuit will inevitably correct it to the corresponding peak current corresponding to the load and the input voltage, and Dt=L*di/vin; this time (conduction time) is a fixed value, so even if the capacitance is increased, the control circuit will increase the divided current and still ensure that the current value injected into the base can satisfy the unchanged conduction time. As mentioned earlier, as long as the load current and input voltage remain unchanged, the conduction time of the switch must be unchanged; then what does the value of this RC determine? They determine the maximum conduction time of the switch, that is, the control circuit does not work and does not divide the base current injected into the switch. This RC value determines the conduction time of the switch. The oscillation frequency of this state is determined by RC, but when the power supply is working normally, a part of the current injected into the switch must be divided, and there is an adjustable positive and negative range, so the current divided by the control and the current injected by the feedback jointly determine the conduction time of the switch, but this conduction time must be determined by the load current and input voltage. The resistor R determines the maximum current injected into the base of the switch tube. C cannot be too large or too small. I have tested that by modifying the capacitor, as long as it is not too large or too small, the circuit still works stably, because the control circuit is still capable of making corrections, that is, ensuring that it is within the control range of the control circuit. Of course, it is necessary to set a suitable RC value to ensure that the current driving the switch tube is appropriate. If it is too large or too small, the circuit will be unstable. Therefore, the capacitor and resistor do not need to be very precise. It is not a timing capacitor as many analyses say. As the name suggests, timing must be precise, but this is not the case. Therefore, the RCC oscillation frequency is related to the load current and input voltage, rather than being determined by the feedback RC. The formula derived from RCC first explains the problem.

The main advantages and disadvantages of the RCC circuit are as follows:

1. The circuit structure is simple and the price cost is low.

2. Self-excited oscillation, no need to design auxiliary power supply.

3. As the output voltage or current changes, the frequency period changes greatly after startup.

4. The conversion efficiency is low and cannot be made into a high-power power supply.

5. Noise is mainly concentrated in the low frequency band.

Let’s talk about the RCC principle again:

The power part of RCC operates like a normal flyback converter. The signal and control part works as follows:

1. When input voltage Vin is added (resistor RG is connected to the base of Tr1), current Ib flows through Rb, Tr1 is turned on, and this Ib is the starting current. The waveform of collector current Ic of Tr1 is shown in the figure, which usually starts from 0.

2. Once Tr1 enters the ON state, the P1 coil of the transformer has been added with the input voltage Vin, so the voltage formed by the P2 coil provides the base current for Tr1, so that Tr1 can remain turned on.

3. The collector current of Tr1 rises in a ramp shape until the current is βIb. At this time, the base current cannot maintain the saturation conduction of the Tr1 transistor, and the voltage between the collector and emitter of the transistor rises. The voltage rise here causes the input voltage on the transformer Np to drop, which further causes Ib to drop. Thus, positive feedback is formed, causing Tr1 to eventually turn off.

4. After Tr1 is turned off, like other flyback converters, the energy stored in the transformer flows to the secondary capacitor to power the load. When the energy inside the transformer is not fully released, the base is always pulled down by the negative voltage reflected from the secondary, and the transistor remains turned off. After the energy inside the transformer is fully released, the circuit working state enters step 1, forming a periodic cycle.

5. If other methods are used to cause insufficient base current when the collector has a large current, the positive feedback mechanism can also be triggered to turn off the transistor Tr1. This feature is often used to achieve current limiting and voltage regulation. (That is, reduce the duty cycle or prohibit the transistor from turning on when the current or voltage is too large) RCC should pay attention to:

Instability of RCC Circuit

It is generally recognized that RCC circuits have very high requirements for components, wiring, and production processes. Using inferior components, low-level layout, and improper transformer winding may cause the RCC circuit to fail to work, or fail after working normally for a period of time. Common failure modes include but are not limited to:

Secondary breakdown due to leakage inductance

The most common and typical failure phenomenon of RCC is the burning of the main switch tube. Most of these failures are caused by the leakage inductance of the transformer base coil. The leakage inductance of the transformer base coil and the resistor in series with the base form an LR low-pass filter circuit, which has a delay effect on the current signal, resulting in a delay in the positive feedback of the base current reduction when the collector voltage rises. Such a delay is fatal to most bipolar switch tubes, causing the transistor to exceed the safe working area and generate excessive heat, ultimately leading to irreversible secondary breakdown.

This type of failure rarely occurs in RCCs made with power MOSFETs , because the safe operating area of ​​power MOSFETs is much larger than that of bipolar transistors. In addition, power MOSFETs are voltage-controlled, with a narrow on/off threshold range. MOSFETs are less likely to withstand large currents and high voltages at the same time, and even if they occasionally fail, they will not fail irreversibly. There was once a batch of MOSFET-based RCC power supplies that often failed due to damage to the switch tube. After investigation, it was found that the manufacturer's technical considerations were not thoughtful, and they mechanically imitated products in the 110V area, using a 500V MOSFET (model IRF840) on a 220V AC line (the voltage after rectification is as high as 311V).

The output voltage is unstable, which may damage electrical appliances.

Another common problem is that the output voltage significantly exceeds the designed output voltage, causing the load to overheat and burn. Especially when the load is a lithium-ion battery, the output voltage is too high and is extremely dangerous, which may cause gas and liquid leakage or even explosion inside the battery. The first reason is that the transformer windings are not fully coupled, and there is leakage inductance, resulting in poor mutual regulation. This problem is more serious when the converter is lightly loaded and the duty cycle is small. Second, compared with the operational amplifier contained in the integrated chip (with an amplification factor of hundreds or thousands of times), the open-loop gain of the voltage loop is too small, making it difficult to accurately regulate the voltage.

And it is almost impossible to properly solve these two shortcomings at the same time. Solving the secondary breakdown problem requires the base coil and the main coil to be wound close to maintain good coupling, while solving the problem of unstable output voltage requires the secondary coil and the base coil to be wound close to each other, and requires thousands of volts of electrical isolation between the primary and the secondary. It is very difficult to achieve these two contradictory goals under the transformer skeleton with limited winding positions.