I talked about the POP sound generated during the plugging and unplugging process of the earphones. Today I will tell you about the POP sound generated by the audio PA.
The POP sound generated by audio PA should be a problem encountered by many projects. It is almost impossible to solve this kind of problem completely. Due to the characteristics of the chip itself, almost most of the digital PA chips or analog PA chips in the world will have a peak voltage in a short time after the audio source is played when measured with professional PA equipment. However, due to the small peak voltage, the generated POP sound is very slight, and the user can hardly hear it at all.
So how much of this peak voltage is acceptable, and the human ear can barely detect it? There is such a parameter VOS in the PA specification, if the peak voltage tested is less than the VOS in the specification, this problem can be almost completely ignored. Most VOS values are ±30 millivolts, which means that if you measure spikes on your PA equipment that are less than this, the situation is arguably not audible to the human ear.
If the peak voltage value exceeds this value, such as hundreds of mV, it is easy for the human ear to hear. This kind of problem often needs to be solved, because it has already affected the user experience.
Causes
When the audio circuit is turned on and off, the moment the output signal with DC bias is loaded on the DC blocking capacitor, the voltage on the capacitor will form an instantaneous charge and discharge current, and then form a peak voltage, which is amplified again by the PA. Finally, it is output to an audio device such as a receiver or a speaker to produce a POP sound that can be heard by the human ear.
Solution
Regarding the POP sound problem caused by audio PA, there are actually many ways.
Take the simulated PA as an example:
From the above schematic diagram, it can be seen that the AC signal output by the platform Codec passes through the series resistance of R4 and R5, and then passes through the 100NF of C7 and C8 and finally enters the PA. When the Codec is turned on or off, it will generate an instantaneous voltage on C7 and C8, and finally it will be amplified eight times by the PA to form a POP sound that can be heard by the human ear.
Method 1: Change the Codec output capacity
In high school physics, we have learned the capacitance formula, Q=C*V. I=dq/dt. Therefore, since the POP is caused by the charging and discharging of the capacitor, the charging and discharging current value can be reduced by improving the capacitor, reducing the value of the peak voltage, and finally reducing the POP sound. The reason is very simple, but the modification of the capacitance value of the DC blocking capacitor is actually the least recommended. Because the recommended value of the PA chip manufacturer is 100Nf. Once adjusted, it may have a chain reaction on the debugging of related effects in the audio field.
Method 2: Change the series resistance value between Codec and PA
According to the formula U=I*R, reducing the resistance value on the path can also reduce the sudden voltage value. However, in some irregularly designed circuits, some people like to choose R4 and R5 as K-level resistors. It is also part of the reason for forming the POP sound.
Method 3: Modify the timing on the software
Earlier, Xiaobai had such a description: when the Codec is turned on or off, it will generate an instantaneous voltage on C7 and C8, and finally it will be amplified and output by eight times the PA, forming a POP sound that can be heard by the human ear . If you think about it carefully, you will find that there are two conditions for the existence of POP: (1) Codec is turned on or off first. (2) PA is working at this time
So, we can try to modify the order to solve this problem:
A When the audio is turned on: Let the Codec be turned on first, that is, first let the capacitor charge for a while, and then turn it on after the capacitor is fully charged PA (generally make the enable of PA pull high), output the audio source. It can perfectly avoid the generation of POP sound. Among them, the time interval between Codec opening and PA opening is usually set to 40ms.
B When the audio is off: first turn off the PA (usually make the enable of the PA low), that is, the PA does not work, and the audio source cannot be output through the PA. Then turn off the Codec. At this time, no matter how the capacitor is discharged, it will not be passed to the load through the non-working PA. Among them, the closing time of PA can be 1ms before Codec is closed.
In short, a simple description, that is, when the capacitor is charged and discharged, the PA is not in working condition.
Of the three methods, the third one is often used the most and is the most recommended.