[USB microphone]3D printing | PCM2912 | 26mm electret

# Preface I have been busy with autumn recruitment recently. In order to leave a good impression on the interviewer, I bought a USB desktop microphone from Lulian. However, after using it for a period of time, I found that although it looks very good, the sound quality is a little bit worse. Something is missing and the vocal sounds like a phone call. So I wanted to make a USB microphone myself, hoping that the sound quality would be better than that of Lulian, and the appearance would be in line with my own aesthetics. ### Production process and sound quality comparison: [[Open Source] Home-made USB microphone for 55 yuan, the sound quality is better than what Greenlink is selling? Contains production process, recitation scene and singing scene sound comparison](https://www.bilibili.com/video/BV1cw411i7RC/?share_source=copy_web&vd_source=88acd14672de527abba1242683452b9d) # 1 Design ## 1.1 Appearance design **Use software** - Rhino - SolidWorks - Keyshot Rhino is used for 3D design expression; SW is used for modeling parts that require more precision (such as screws and nuts); Keyshot is used for rendering renderings and animations. **Rhino modeling effect display:**  **Black microphone rendering rendering:**  **White microphone rendering rendering:**  ## 1.2 PCB design PCB designed using JLC EDA. Mainly referring to the schematic diagram in the[PCM2912A EVM user manual](https://www.ti.com.cn/cn/lit/ug/sbau141/sbau141.pdf?ts=1694512611605), only some filter capacitor parameters were changed. At the same time, the resistance value of the bias resistor was changed according to the characteristics of the purchased microphone. **PCB physical picture:**  # 2 Model 3D printing The 3D printing filament uses Sanlu's black matte PLA. I was worried at first that it would not be enough for a strong team, but in actual use, there is no problem at all. Friends who need to make their own products can rest assured to use PLA material for printing. **Tip: All models do not need to be supported when printing.** # 3 Finished product pictures      # 4! ! ! Known issue! ! ! Record the currently known problems, and friends who want to do it themselves should pay special attention to it. I am also looking forward to the passing experts who can give me some advice. Thank you very much! ## 4.1 Noise floor The sound recorded by the finished microphone has an obvious noise floor after amplification. The actual measurement is around -50dB, which I personally think is acceptable. After replacing the small-sized electret capsule, the noise floor disappeared. Therefore, it is speculated that the large-sized electret capsule is more susceptible to interference. Perhaps the problem can be solved by adding a metal mesh shield. ## 4.2 The mute button needs to be changed to self-locking The mute function in the PCM2912A chip manual should be triggered by a non-self-locking button, but the microphone I made currently requires the button to be pressed for the mute function to take effect, which means that the mute button needs to use a self-locking switch. But this is inconsistent with the actual logic of use, and the problem is currently being investigated. Referring to the evaluation board, the mute button inside is also micro-moving. Therefore, I suspect that it is caused by buying a fake chip or other unknown reasons. ## 4.3 Teflon wire is relatively hard The signal wire insulation material I used in the finished product is Teflon, which is relatively hard and affects the adjustment of the sound head angle. It is recommended to replace it with silicone material or enameled wire.