Intermittent oscillation of switching power supplies

Thread Content

Intermittent oscillation in switching power supplies is a relatively common fault phenomenon; its occurrence rate is much higher than that of cases with no output voltage, or when the output voltage is too high or too low. The difficulty of diagnosing and fixing this fault, as well as its symptoms, are also more complex compared to those of the aforementioned three types of faults. It is manifested as a voltage output that is sometimes high and sometimes low, or present occasionally, accompanied by a slight \"tap, tap\" sound, or a \"hiss, hiss\" sound. If summarized briefly, the causes of failure can be categorized into three types: overload type, under-voltage type, and no-load type. 1. Overload type: This type of fault is relatively common among the four types, and its repair difficulty is rated at level two. It is usually caused by the breakdown of the rectifier diodes on the secondary side of the switching transformer, leakage in the filter capacitors, or damage to IC devices and cooling fans in the load circuit. When the corresponding load branch is removed, or the faulty component within the power supply itself is replaced, the power supply shifts from intermittent oscillation to a normal output state. ①In the case of voltage sampling, such as faults in the +5V load circuit or in the power supply itself (e.g., leakage in the filtering capacitor), the maximum increase in the output voltage of other circuits may slightly exceed the rated voltage; for example, a +15V output might rise to +17V. ② If the fault occurs in a power supply or load circuit other than those involved in voltage regulation, the maximum increase in the output voltage is generally limited to a small value. For example, the +15V output overshoots to 8V. Although it is intermittent oscillation, the output voltage has a certain amplitude. 2. Under-voltage type: This is usually caused by a problem with the power supply circuit at pin 7 of the 3844, or by leakage components in the reverse current absorption circuit connected in parallel with the switching transistors in the primary winding of the switch transformer; the difficulty level of troubleshooting is three or four stars. ①Due to a faulty 7-pin power supply circuit in the 3844, issues such as an inefficient rectifier diode (with normal values for forward and reverse resistance as well as forward voltage drop) or a failed filter capacitor (with its value changing from 100uF to 10uF) can occur. If the necessary operating energy for the oscillation chip (the energization energy for the switching transistor) cannot be supplied properly, the horse cannot be fed adequately and thus cannot pull the cart. In such faults, if it is confirmed that the voltage regulation circuit is functioning properly, applying an external DC18V power supply to pins 7 and 5 of the 3844 component will restore normal operation of the power supply. If the filter capacitor is found to be in good condition at this point, it is necessary to replace the rectifier diode without hesitation! ② There are leakage components in the reverse current absorption circuit connected in parallel with the primary winding of the switching transformer, which results in a significant reduction in the induced voltage of the self-powered winding; as a result, insufficient energy is available to power the oscillation chip and the switching transistor. After confirming that there are no abnormalities in the 7-pin power supply circuit of 3844, the focus of inspection should shift to this circuit. The most common cause of failures is leakage from high-speed, high-reverse-voltage diodes or voltage regulators; the problem is that when these diodes or regulators are tested, they seem to be functioning properly. Temporarily disconnecting the absorption circuit or replacing components for testing can often resolve the issue quickly. In the case of under-voltage faults, the voltages output by the secondary windings of the switching transformer are all low, fluctuating within the range of a few tenths of a volt to a few volts; it gives the impression that the power supply is not able to deliver sufficient power. Haha. 3. No-load type: This is a specific type of fault. This only occurs when servicing the switching power supply/drive board separately. I won’t mention how many stars the maintenance difficulty is rated at. As shown in the power supply diagram below, ① the output side of this feedback optocoupler is directly connected to pin 12 of the 3844; ② the filter capacitors of the secondary winding of the switching transformer, especially those associated with the sampling voltage winding, have a capacity on the order of millimicrofarads. When the load branch is unloaded, it is not possible to establish a minimum stable load current; as a result, the energy supply side loses its target for supply and does not know what to do, ending up supplying either too much or too little energy, which leads to intermittent oscillations. The schematic diagram of the Sain switching power supply circuit: This is not a fault per se, but rather results from the load branch being in a \"floating\" state. When it is connected directly to the MCU (with special attention paid to the connection of the display panel, as digital displays require a relatively high operating current), the power supply will function properly; in a sense, no repair is necessary. When maintenance can be carried out separately, it is necessary to activate the circuit slightly so that the switching power supply can function properly, thereby facilitating the maintenance process. The method is to connect a 2W, 30–100Ω load resistor in parallel across the C87 capacitor in the diagram. If it is considered to be a fault, there is a distinct characteristic: when measuring the output voltages of each circuit on the secondary side of the switching transformer, it is found that the voltage fluctuations are significant, and the peak values of these voltages are much higher than the rated voltage – for example, +5V may rise to 8V, or +15V may rise to over 20V. And the downward swing can reach 0V.