Analog to Digital Conversion

Any decent microcontroller will be outfitted with analog to digital conversion modules. Analog to digital conversion (ADC) modules convert the information presented to them into a binary representation that the microcontroller can manipulate.

Our physical world is analog in nature. Take your phone’s recording system: you speak into your phone’s microphone, and the signal received is analog. The microphone must then convert this analog information from your voice into a digital signal representation that can be interpreted by your microcontroller. One application of this would be in voice recognition systems where a microcontroller in your device needs to listen for a “wake-up word.” A “wake-up word” is a word or phrase you can use to get the attention of your computing device. The microcontroller would need to convert your voice into a digital form so that it can search for the word, which would wake up the device from sleep. Most smartphones also incorporate features that filter out background noise or amplify voice signals. Using the ADC module of a microcontroller, you can easily filter out noise from the audio coming into the microphone. This type of filter though is not like the hardware filters we learned about in Chapter 2 but is in fact a digital one which operates in software. Their function however is similar to the filters you learned about previously.

As we learned in Chapter 2, analog circuits are known as continuous circuits; thus, the signals produced by them are continuous in nature. When we say a signal is continuous, we mean that it is characterized by smooth transitions between points in the signal with a variation in time. A digital signal, being produced by digital circuits, as we learned has a discrete nature. For our purpose, these can be thought of as a digital representation of a continuous signal.

Let’s first take a closer look at what’s meant by a signal. A signal can be thought of as energy that we can use to extract information. Within the context of microcontroller electronics, signals take the form of a voltage or current that is produced by a sensor. The signal that is produced by a sensor is usually not in a form we can directly get information from, as we lack the faculties to process sensor information. The microcontroller would need to read the sensor for us, then via a display of some sorts relay the information in a way we can understand. The microcontroller can also take this information and perform some action. For example, a digitally controlled oven may have a thermocouple sensor to read the temperature inside the oven. The microcontroller would need to switch off a heater when the temperature inside the oven reaches a preset temperature. This is just one application where an ADC module may be utilized.

ADC Hardware

There are several ways to perform analog to digital conversion, the most common of which is by using the successive approximation ADC circuit. A successive approximation ADC uses various circuit components. These components used by the successive approximation circuit include a digital to analog converter (DAC) that converts a digital signal into an analog one. It also uses a comparator, a sample and hold (S&H) circuit , and a control circuit, which we’ll go through in a moment. The successive approximation ADC circuit looks something like Figure 6-1.

The circuit operates by first taking an input from the microcontroller pin. Since there is only one ADC module on a typical microcontroller, using a multiplexer circuit, we can route the voltage from several pins to the ADC module within the microcontroller. This is because one module can only read one pin at a time, and many times, we may need our microcontroller to read several sensors.

To ensure that the ADC functions correctly, a snapshot of the voltage of the microcontroller pin is taken. This snapshot is needed to be taken since the sensor would keep continuously outputting data, and the ADC circuit would need time to perform the conversion. This snapshot is taken at that point in time the ADC module is invoked to perform the conversion by the microcontroller. To take the snapshot, a capacitor is used because, as we learned before, a capacitor can store voltage. The capacitor will be charged up to the same voltage value as the pin we want to read the voltage of. For example, if a sensor outputs 3.7 volts, then quickly outputs 3.9 volts, then 4.0 volts, we need some sequence to effectively digitize all these readings. This is because a lone ADC module lacks the ability to perform parallel conversion. The ADC module would first need to take the 3.7 volts, convert it to digital form, then the 3.9 volts, and so forth.

This voltage that the capacitor is charged to will be what we will use to perform the successive approximation procedure on the microcontroller. We call this front end on the ADC module with the capacitor a sample and hold (S&H) circuit.

Once there is a voltage on the S&H circuit, the comparator working together with the ADC circuit comes into action. The combination of these two circuits will keep cutting the voltage at the input in half. When we say the voltage at the input, we are really talking about the snapshot voltage. The process will continue until we reach as close to the input value as the combination of these circuits will allow.

When discussing ADC, a word that comes up frequently is resolution. Let’s take a closer look about what is meant by resolution. The ADC module has a range of values that it can use to represent the converted signal. This value is represented as a number of binary bits which is output by the module. Essentially, the smallest voltage which causes a bit change is what we refer to as the resolution. According to the resolution of the ADC, the ADC module will have a certain number of voltage steps that it can use to represent the voltage measured. The greater the number of voltage steps, the greater the resolution of the ADC module.

To output the information, the ADC circuit will manipulate the most significant bit (MSB) down to the least significant bit (LSB) of the binary number which is to be output by the module using the conversion value from the comparator and DAC. This is to try to match the voltage value read on the input as closely as possible and represent it in binary form. This output value is of course limited by the resolution of the DAC within the ADC module. This is because the amount of times the voltage can be cut is dependent on the DAC resolution. This value is then output from the ADC module so that it can be read and manipulated by the CPU.

When selecting a microcontroller for your application, sometimes the resolution of the ADC can be a deciding factor in selection of your device. The resolution required to read a temperature or light sensor, for example, may not be as important. However, if we need to build a circuit for voice recognition, you will need high resolution to have a functional circuit. ADC circuits onboard the microcontroller typically have a resolution of about 8 to 16 bits. For most sensor reading applications, this is sufficient, and if greater resolution is required, for example, in specialized medical applications, an external ADC IC is used.

Going Deeper into ADC

At this point, we will look at some additional terms you are likely to encounter when discussing ADC. The first thing we need to talk about when discussing ADC is sampling. As we explained in the previous section, in order for the ADC module to have time to process the read value, a snapshot of the voltage must be taken. Sampling is the name given to the number of snapshots we can take over a period of time. When we talk about sampling in analog to digital conversion, what we are talking about is the ability of the ADC module to replicate the signal it is analyzing and create a digital equivalent as close to the original as possible. The more samples a digital to analog converter can capture, the more accurate the digital representation of the original signal will be.

A dilemma you may arrive at is trying to determine the optimal amount of samples we need to take to have a decent representation of the signal we are sampling. If we take a lot of signals, then our module will be slow and may not have the response time we need for our application. On the other hand, if we take too few samples, we will have a fast conversion; however, we may not have an accurate representation of the original signal. A special sampling rate known as the Nyquist sampling rate is the benchmark that is used to determine the minimum sampling time.

The Nyquist sampling rate is stated as having the minimum sampling rate being at least twice the highest frequency of the signal we are trying to sample. For example, if a signal is expected to have a maximum frequency of 10 kHz, then we must have a minimum sampling rate of 20 kHz. Another way of saying it is that the sample and hold circuit must take 20,000 snapshots a second to adequately measure the signal.

ADC modules also have a data sampling rate that is usually measured in the number of samples per second (SPS) the ADC module is capable of measuring. Typically, we would see the manufacturer specifying the sampling rate in kilo samples per second (kSPS) or mega samples per second (MSPS).

Another facet we can look at within ADC conversion is that of quantization. Quantization is the name given to the process of mapping the value of input voltage to the ADC to an output value that the ADC is capable of producing. A 10-bit ADC module, for example, will have steps from 0 to 1023. We call this value from 0 to 1023 the quantization level of the ADC. An 8-bit ADC will have quantization levels from 0 to 255. This quantization is the name given to the rounding of values that are equal to the voltage we tried to measure on the pin. What this means is that in the strictest sense we can redefine resolution to mean the measure distance between two adjacent quantization levels.

The Potentiometer

The first device we will look at for using with the ADC module is the potentiometer. When we look at real-world embedded devices, you will realize that potentiometers are found in a lot of devices. For example, the volume knobs that are encountered on stereos and speaker systems use potentiometers in order to allow the user to make adjustments to the volume on these devices. Such devices all use potentiometer circuits to operate properly. In Figure 6-2, we see what a potentiometer looks like.

The potentiometer may come in varying form factors; however, if we look at its mechanical construction, we will observe that they are all three-pin devices with a turning mechanism at the top to allow for adjustment. Sometimes, you may see a device looking like a potentiometer with five or more pins protruding from it. This is not a potentiometer but is a rotary encoder and functions differently.

Of the three pins on the potentiometer, one pin is connected to the supply rail, the other pin is connected to the ground supply, and the third pin is connected to the input pin of the microcontroller. When you adjust the knob on the potentiometer, what you are doing is varying the voltage on the analog input from 0 volts to VDD. This can typically range from 0 volts to 3.3 or 5 volts.

The reason this occurs is because the potentiometer can be thought of as a type of voltage divider circuit. A voltage divider is a circuit that turns a larger voltage into a smaller one using the properties that exist when two resistors are placed in series. In Figure 6-3, we see what this voltage divider arrangement looks like.

The voltage divider as shown in Figure 6-3 will output approximately half of the voltage that is present on VCC. So if VCC is 5 volts, then the output of the voltage divider will be 2.5 volts. The output of the voltage divider is determined to be the voltage measured at the center point of both resistors. This is what voltage dividers do; they output a fraction of the input voltage that is present on the input. The output of the voltage divider is not arbitrary but is determined based on the values of the resistors that make up the divider circuit. Once we know the value of the resistors, we can use what is known as the voltage divider formula to calculate the output. This formula is useful when we need to get a specific voltage output using resistors.

To calculate the output of the voltage divider, we use the formula which is given as

Vout = Vin(R2/(R1 + R2)

We know that the divider circuit in Figure 6-3 would give half the input voltage; however, if we wanted to calculate the same mathematically, we can use our voltage divider formula. If we wanted to calculate our Vout, we could use the formula and input the values of our Vin, R1, and R2. When performing our calculation, we take R1 to be the top resistor and R2 to be the resistor on the bottom, and we get

Vout = 5v (1 / (1 + 1) = 5v x 0.5 = 2.5

The potentiometer functions as a resistor and exhibits resistance. However, it can be thought of as a different device entirely, and as such it is given its own schematic symbol. We see the schematic symbol for the potentiometer in Figure 6-4.

The schematic symbol for the potentiometer consists of an arrow pointing to a resistor. This is meant to indicate that the resistance is adjustable. Sometimes, you see a smaller version of the potentiometer as shown in Figure 6-5.

This version of the potentiometer is called a trimmer potentiometer or a trim pot. The function is the same as the regular potentiometer; however, it is designed to be hidden in the end product and adjusted by the manufacturer or service personnel, not by the end user of the product. These are easy to identify because whereas a regular potentiometer has the adjustment knob extended out from the device, these must be adjusted by a screwdriver in order to change their values.

Analog to Digital Conversion in CircuitPython

ADC with MCU Schematic

The potentiometer can be connected to our MCU as shown in Figure 6-6.

ADC Circuit Connection Tips

When you have finished connecting the circuit, it should look similar to Figure 6-7.

Once you have your circuit setup, we run the program as shown in Listing 6-1. Remember to check your connections and ensure they are in accordance with your schematic, or else you risk damaging your board circuit components. For that reason, I always recommend you connect your circuits with the power off, then power it on before writing your program to test it.

CircuitPython with Potentiometer Program

The first thing we do is import the CircuitPython libraries we need to control the board. These are our usual imports of our time and board libraries. At (1) we import the AnalogIn object from the analogio module. This is the library that allows us to perform analog functions. The next step in our program at (2) is to create a pin object that will allow us to read the specific analog pin we are working with. In this case, we are working with pin A0 on the board, and we name it analog_in.

Moving along the program at (3), we then have a function that allows us to read the voltage level from our potentiometer. This function get_voltage will take a pin as the input and then will perform calculations to return the correct potentiometer reading. At (4) in our super loop, we use the get_voltage function to read our analog input pin and continuously print this read voltage to our serial console. Once everything is running okay, in your serial console, you should see the output as in Figure 6-8.

When we adjust the input of the potentiometer, we see that the output goes from 0 volts to around 3.3 volts, and this value is printed to the serial terminal.

Photoresistor

One sensor that we will now be able to use thanks to this ability to read analog sensors is the photoresistor. A photoresistor is a type of resistor that changes resistance based on the amount of light that is falling on them. When there is no light falling on the photoresistor, it has an extremely high resistance in the order of hundreds of kiloohms. However, when light falls on the photoresistor, the resistance drops to a few hundred ohms. To create a program to read the photoresistor, we will need to create a voltage divider circuit. This voltage divider circuit will consist of the photoresistor with a regular resistor. The output of the circuit will then be read by the microcontroller to determine change in the light levels of the photoresistor.

Photoresistor with MCU Schematic

We connect the circuit as is shown in Figure 6-9. We have the photoresistor connected to pin A1 and an LED connected to pin A2.

Photoresistor Circuit Connection Tips

When you have finished connecting the circuit, it should look like Figure 6-10.

Now that we have our circuit setup, we can write some code!

Photoresistor with CircuitPython Program

This program does a lot of things. Let’s look at them to see what happens. At the top of the program, we import our board, time, analogio, and digitalio libraries. At (1) we create an adjustValue variable that is used to aid in the reading of the photoresistor. Our next step is to create an object that represents our real-world LED at (2) and set it to an output at (3). We then create an object to represent our photoresistor at (4). At (5) we then take a single reading from the photocell and store that value. This is the value of the ambient light that currently exists.

Note that after taking our reading, we must use the deinit method at (6) to allow our pin to be used in other parts of the program. If we don’t do this, then we won’t be able to use the pin in our super loop.

Inside our super loop at (7), we continually read the photoresistor value. There is a conditional that checks to see if there is darkness on the LED. If there is darkness, the LED will be turned on, and when there is light, we will turn the LED off.

Save the program and run it. You will observe that when you cover the photoresistor with your hand, the LED turns on; however, when the LED is at ambient brightness or brighter, the LED will not be lit.

Temperature Sensor

One sensor that we can use now that we have access to using analog inputs is a temperature sensor. The temperature sensor we will be using is the TMP36 temperature sensor that gives us temperature readings in Celsius using voltage as an output. The TMP36 is pictured in Figure 6-11.

The TMP36 temperature sensor device has three pins. One is the VCC pin which we connect to our positive supply. There is also a GND pin which we connect to our ground and an analog voltage out pin which is read by the microcontroller. Depending on the value of the voltage at VCC, the microcontroller can then do some calculations to determine the temperature we are reading.

The sensor works from 2.7 to 5.5v, making it a very versatile sensor as it works with all common microcontrollers equipped with an ADC module. It is capable of being used with CircuitPython-based MCUs that are 3.3-volt devices, which fall within the voltage range of the device. The temperature sensor can take measurements from –50 degrees Celsius to 125 degrees Celsius which is a good range for most projects you will undertake.

Temperature Sensor with MCU Schematic

We connect the circuit as is shown in Figure 6-12. Our TMP36 is connected to our A0 analog input pin. We need the capacitor C1 and resistor R1 to get accurate reading from the sensor when working with a CircuitPython MCU. This is due to the underlying configuration of the device which reads the sensor at a high speed.

Temperature Sensor Circuit Connection Tips

When you are finished connecting your circuit, it should look similar to Figure 6-13.

Temperature Sensor with CircuitPython Program

In the program, we do our usual imports to get the board up and running. At (1) we set up the temperature sensor connected to the analog input pin A0. At (2) we have a function for reading the temperature sensor and doing the conversions from voltage to temperature. At (3) we create an instance of the analog object for the temperature sensor. In the main program, we run a super loop at (4) where we read the sensor data and print it to the output consoles every second. If you look at your serial console, you should get an output similar to the output I got in Figure 6-14.

Conclusion

In this chapter, we covered the basics of analog to digital conversion. We looked at how microcontroller ADC circuits work in hardware, and along the way of learning how these circuis function we discovered how potentiometers, photoresistors and voltage dividers work. We also learned how to use CircuitPython with analog inputs. Using this information, we were able to read a temperature sensor and output the information to a serial terminal. With the knowledge gained here, hundreds of sensors are now at your disposal.