Electret microphones are the most commonly used microphones today. Every cellphone and laptop has one embedded into it, and many studio microphones are also electrets. They can have an extremely wide frequency response (from 10Hz to 30kHz ), and typically cost less than a dollar. They are also very small and quite sensitive. Despite these good characteristics, they can also have a few drawbacks, such as a high noise floor, high distortion, and uneven frequency response. We will dissect an electret microphone, explain how it works, and talk about the reasons for its various attributes.
Figure 1 – A common electret microphone capsule.
Figure 2 – Front view.
Figure 3 – Back view.
In Figures 1 – 3, a common through-hole electret microphone capsule is shown. This capsule is on the larger side, and has two leads coming out for mounting to a PCB. A surface-mount (SMT) capsule would just have 2 bumps of solder on the bottom. 3 terminal capsules are also made, but are not very common. We will explain why later.
The top of an electret capsule is often covered with a porous material which is attached with glue. This is the black circle in Figures 2 and 4. This material keeps dust and other debris away from the sensitive electret material, and also gives some protection from wind noise on the microphone.
Figure 4 – Capsule with dust cover removed.
Underneath this dust cover is a small hole in the aluminum capsule. This is where the sound enters the microphone. On a directional microphone there are also holes in the back of the capsule (through the PCB) to help cancel out sounds from the sides.
The aluminum capsule contains the electret material itself, and a small amplifier, which can be seen in Figure 5.
Figure 5 – Contents of microphone capsule (dust cover, capsule, electret diaphragm, amplifier module).
The electret material is the shiny silver circle shown in the middle of Figure 5. It is made of a metalized mylar film which is adhered to a metal washer. There is also a small red plastic spacer to keep the film a fixed distance from the amplifier module. Both the spacer and electret are extremely thin (.001″ or less).
Figure 6 – Close up of plastic spacer and electret diaphragm. Note that the electret diaphragm is mounted to a small metal washer.
The electret material is capable of holding a fixed electric charge, which does not decay with time. This is different from a conventional condenser microphone which needs to have a charge placed on it (i.e. phantom power). When air impinges upon the diaphragm, it moves back and forth, changing the distance to the amplifier module pick-up plate, which in turn creates a voltage difference. How exactly this works will be explained later, but the plastic spacer keeps the diaphragm from touching the pick-up plate of the amplifier module. An exploded view of the amplifier module is shown below.
Figure 7 – Exploded view of the amplifier module (pick-up plate, transistor, plastic housing, PCB).
The amplifier module consists of the pick-up plate, a plastic housing, a single transistor, and the PCB. The pick-up plate has holes in it for the displaced air to move through, and is connected to one lead of the transistor (usually by a small tack weld, but sometimes by spring force of the transistor lead itself). The other two leads of the transistor are soldered to the PCB. So, on a through-hole electret capsule, the two leads sticking out the bottom are just the leads of the transistor itself. The plastic housing keeps all of these elements rigidly fixed within the aluminum housing, and insulates the pick-up plate from shorting to the housing.
The amplifier consists of a single JFET transistor, with the gate connected to the pick-up plate, the source connected to ground, and the signal appearing on the drain. This is called a common-source configuration, as the source is connected to ground, which is common to all signals. The JFET in this electret microphone is a 2SK596, which is designed for low-noise applications. A datasheet for it can be found here.
Figure 8 – Amplifier transistor (2SK596).
The transistor is connected to the PCB, which has two conductive pads, and a conductive ring around the outside. One of the pads is connected to this ring, and acts as a ground trace. When the aluminum housing is placed on, it is bent into contact with this ring, grounding the whole case.
Figure 9 – Electret module PCB.
Since the electret material has a conductive film on the outside, and is connected to a metal washer which touches the aluminum capsule, the entire assembly is essentially sealed in a grounded case. A cross section of the entire electret microphone module is shown below.
Figure 10 – Cross-sectional view of electret microphone module.
As can be seen above, the charged electret material and the amplifier pick-up plate are very closely spaced and have a lot of area facing one another, and therefore create a capacitor. In the olden days, capacitors were called condensers, hence the name condenser microphone. The electrical schematic of the full electret microphone is shown below.
Figure 11 – Equivalent schematic of electret microphone.
The JFET has three pins: the gate (G), drain (D), and source (S). The gate is the control pin, and goes to one plate of the microphone “capacitor” (with the electret material forming the other plate). The source is connected to ground, and the drain is connected to a resistor and power supply in your external circuit. In a 3 terminal microphone element, both the source and the drain would be pinned out, and a third pin for ground would be used. This allows for using a different amplifier configuration.
So, how does this all work?
The electret maintains a fixed charge, and therefore maintains a voltage across the capacitor. The mathematical equation for the voltage on a capacitor is V=Q/C, where Q is the charge on the capacitor, and C is the capacitance. In the case of the microphone, since the diaphragm is moving back and forth, the shape of the capacitor is changing, and its capacitance changes accordingly. The equation for a parallel plate capacitor is C=e*A/t, where e is a material constant representing the properties of the material between the plates, A is the area of the plates, and t is the separation between the plates. As the electret material moves due to sound pressure variations, t becomes larger and smaller, and the voltage varies linearly with this distance since V=Q/C=Q/(e*A/t)=Q*t/e*A.
As the voltage at the gate varies, the gate to source voltage (Vgs) varies since the source is grounded. This variation in Vgs causes the JFET to conduct more and less, and the current through the drain (Id) changes, producing a signal across the drain resistor (R). The output is taken from the drain.
A JFET is used as the amplifier because it has a really high input resistance (30Mohms or more). This means that almost no current is pulled off the electret capacitor. If the amplifier had a lower input resistance, the low frequency response of the microphone would suffer. This is because the input stage acts like a high-pass filter, with the electret being the capacitor, and the input of the amplifier being the resistor, and larger values of R and C give lower cut-off frequencies.
Where is the noise?
The main noise sources in this microphone are pick-up noise and transistor noise. Since the entire capsule is sealed and grounded, the pick-up noise is very low and usually not noticeable. The transistor noise, on the other hand, can be quite high, due to the high input resistance on the JFET. Typical values are around -120dB to -110dB, which may sound rather low, but the audio signal level is usually less than -40dB, so its only an 80dB signal to noise ratio (SNR). This is a common issue with condenser microphones due to the high input resistances required. Fortunately, this noise floor does not increase appreciably with signal level, so the SNR can improve greatly for high audio levels.
Where is the distortion?
The drawback to driving a microphone hard, is that the distortion will increase. The common 2 terminal electret microphone is particularly prone to this since it is setup as a common-source amplifier. The input to the JFET is basically a diode, which means it has the same non-linear behaviour as a diode. For input voltage swings greater than 10mV or so, you start seeing pretty heavy distortion. And, to make things worse (or better if you like distortion), the effects are not symmetric, as the diode conducts in one direction, but not in the other.
The 3 terminal electret microphone gets past some of this JFET distortion by using a source-follower configuration, which connects the resistor between the source and ground, so the source can follow the gate signal, and the voltage drop across the internal diode can stay relatively constant. But, this configuration isn’t as common, as the large production applications (e.g. cellphones) usually operate at low input volume levels, and distortion isn’t as much of a concern (in some ways the compression could be seen as a feature). The 2 terminal microphones also require a less complicated amplifier circuit, and only 2 contacts on a jack if an external microphone is used.
Another source of distortion is the diaphragm movement itself. The diaphragm is not moving perfectly linearly to the fixed amplifier pick-up plate. It bows in and out since its edges are fixed, so the voltage doesn’t vary perfectly linearly as well. The greater this bowing, the less linear it will respond. For this reasons, a smaller electret diaphragm (and smaller capsule) will give less distortion. It will have less surface area and therefore bow less for a given sound pressure. This will keep the JFET voltages lower and the diaphragm in a more linear range.
What accounts for its frequency response?
As stated earlier, the high input impedance of the JFET determines the low end of the microphone’s response. The high end, on the other hand, is a function of how fast the diaphragm can move back and forth. This is where the electret shines, as the material is so thin and small, that it can move very quickly. It has little mass, and can usually vibrate up to 30KHz. The smaller the electret capsule, the higher the frequency response will be.
But, because the diaphragm is so light and flexible, it can bend in a number of ways, giving uneven frequency response. And, just as a speaker cone will move differently depending upon volume and frequency (exciting different vibrational modes), so will an electret diaphragm. In this respect, the smaller diameter diaphragms are better, as they are relatively more stiff in the radial direction.
What determines its sensitivity?
The larger the electret surface area, the more it will move for a given sound pressure. So there is a direct tradeoff between the other parameters discussed and sensitivity. A smaller diaphragm will tend to give better high frequency and distortion characteristics, but will not be as loud, and therefore have worse SNR. A smaller diaphragm will also have a smaller capacitance, so its low frequency response won’t be as good. So be careful when picking a microphone for your application. Buy a few different kinds and try them out. We’ve found that they vary wildly between manufacturers, and even between the same part in a production run. Luckily they are inexpensive, so you can afford to try a few dozen and see what works for you.