Next-generation microphones capable of increased performance are required to address the needs of mobile phones, which are getting progressively smaller while incorporating an ever larger number of features. A microphone preamplifier implemented in modern submicron CMOS processes - compared to a traditional JFET solution - enables the use of the digital-output microphone for the mobile microphone market.
Microphones used in telecom applications have for many years been of the electret condenser microphone (ECM) type, and composed of a membrane, a back plate and an electret layer. The membrane and the back plate form a capacitor, where the membrane is movable and the back plate is fixed.
The electret layer stores a fixed charge corresponding to a capacitor voltage of approximately 100 V. Sound pressure causes the membrane to move and, consequently, change the capacitance of the microphone. The charge on the capacitor is constant, so the voltage across the capacitor will vary with the incoming sound pressure level according to the following formula:
Q = CV
Where Q is the capacitor charge, C is the capacitor value and V is the capacitor voltage.
Microphones for mobile applications are quite small, typically 3 to 4 mm in diameter and 1 to 1.5 mm thick. Consequently, the microphone capacitance is relatively small-typical values are on the order of 3 to 5 pF, and, in some cases, as small as 1 pF.
Figure 1. A cross section of a traditional electret condenser microphone shows the JFET that was typically used as a microphone preamplifier.
(Click on image to enlarge)
Figure 1 shows a cross-section of a JFET-based ECM. The signal emerging from a capacitive microphone has no drive strength and, thus, needs a buffer/amplifier prior to further processing. For many years this microphone preamplifier has been implemented using a simple JFET.
As micromachining of electret microphones has improved over the years, microphones have become smaller, and their element capacitance has decreased. Standard JFETs, with their relatively large input capacitance, no longer suffice as microphone preamplifiers because the emerging signal will suffer significant attenuation as a result of the capacitive split between the JFET and microphone cartridge element.
Further motivation comes from the quantum leap that CMOS process technologies have experienced over the past 30 years. From a technical perspective, there is a lot to be gained by replacing JFET-based amplifiers with CMOS digital counterparts. Preamplifiers implemented in modern submicron CMOS processes enable a wide range of improvements over traditional JFETS:
- Low harmonic distortion
- Gain setting
- Different functional modes, such as sleep mode for low power consumption
- A/D conversion, enabling digital-output microphones
- Drastically enhanced sound quality
- Noise immunity
Digital-output microphone preamplifiers
JFET-based amplifiers have inherently low power consumption due to their simplicity, but suffer from poor linearity and accuracy. Thus, the main goal of a digital preamplifier is to retain very low power consumption while significantly increasing dynamic range through improved linearity and lower noise.
Another drawback of the traditional JFET solution is that mobile phones constitute a very noisy environment. Analog microphone output signals can easily be corrupted by interference before A/D conversion takes place. Incorporating A/D conversion into the microphone itself provides a digital output that is inherently less prone to such corruption.