Perform surgery on the speaker protection circuit

At present, the speaker protection circuit using discrete components basically adopts two circuits as shown in the above figure and the middle figure. The main working principle of this circuit is to delay the speaker when powered on (the delay time is determined by R5 and C5) and detect the output midpoint voltage. When the output midpoint voltage deviates from the zero point by a certain amplitude, the relay is turned off to protect the speaker. The sensitivity is determined by R1, R2, and R3.

The implementation principle of the above figure and the middle figure is the same. The middle figure uses transistors T1 and T2 to replace the rectifier bridge and transistor T2 in the above figure. Here we only analyze the middle figure. When L-IN is connected to a DC positive voltage (greater than 0.6V), transistor T1 is cut off, T2 is turned on, the Darlington tube composed of T7 and T8 is cut off, and the relay is disconnected; when L-IN is connected to a DC negative voltage (less than -0.6V), transistor T1 is turned on, T2 is cut off, the Darlington tube is cut off, and the relay is disconnected; when the voltage connected to L-IN is close to zero (-0.6~0.6V) after filtering by R1, C1, and C2, T1 and T2 are both cut off, the Darlington tube is turned on, and the relay is energized; the R-IN input is the same as L-IN, so it is not analyzed here. At first glance, there is no problem with these two circuits. After testing, it is found that when a 1.5V battery is connected to L-IN, the relay is immediately disconnected, but when the battery is connected in reverse, the relay will not be disconnected. The relay will only be disconnected when the negative voltage connected to L-lN is lower than -4V. A careful analysis will reveal that when a positive voltage is added to L-lN, the current only returns to the ground through R4 and T2; when a negative voltage is added, the current returns to the added negative voltage source through R4, T1, and R1. Therefore, when the input is a negative voltage, due to the existence of R1, when the input negative voltage is not large enough, T1 cannot be saturated. Only when the negative voltage is relatively large can T1 be saturated and the protection circuit can work normally.

To solve the above problems, the following figure is an improved circuit based on the middle figure, mainly adding transistors T5 and T6, and the circuit also retains a simple style. In the application, the power supply voltage is relatively high (dual 28V AC), so the voltage is also changed to 24V. The improved circuit (midpoint voltage detection sensitivity is ±1.2V) has been verified to disconnect the relay when a 1.5V battery is connected in the forward or reverse direction. In addition, the left and right channel detection circuits are separated in this circuit to avoid crosstalk and the protection circuit cannot detect when the output voltages of the two channels are equal and the polarity is opposite.

When L-lN is connected to a positive voltage (>0.6V), T2 and T5 are turned on, and the current flows to the ground through R4 and T5; similarly, when L-lN is connected to a negative voltage (<0.6V), T1 and T5 are turned on, and the current flows to the ground through R4 and T5; at this time, the current flowing through R4 will not flow through R1, but will directly flow into the ground, so it will not be affected by the existence of R1. This circuit has high sensitivity and can make the relay turn off quickly.

T6 is used to discharge capacitor C5 when the power is turned off. During normal operation, T6 is cut off. When the power is turned off, the 24V voltage drops quickly, while the -35V voltage drops slowly, so that T6 is turned on and C5 is discharged. If the power is turned on immediately after the power is turned off, the delay circuit can also delay normally. In the above and middle figures, the capacitor discharge current is very small after the power is turned off, and the discharge time is long. If the power is turned on immediately after the power is turned off, the delay time will be significantly shortened. If this function is not needed, R6, R7, T6, D3, and C8 can be removed.

From the two examples of wrong design of speaker protectors above, we can know that we should be targeted for some ready-made circuits. We cannot just use them casually. We should be skeptical, carefully study and analyze, and finally conduct experiments to confirm its correctness. When it is confirmed to be correct, we can use it, and we cannot consider the problem from the "take it for granted" way of thinking. It is worth mentioning that in addition to conducting experiments for confirmation, you can also use simulation software for circuit analysis, such as EWB. This is mainly because the use of simulation software can improve work efficiency and help designers to analyze, thus achieving twice the result with half the effort.