Design of a low power consumption toxic gas detector

Safety first! Many industrial processes involve toxic compounds, such as chlorine in the manufacture of plastics, agrochemicals and pharmaceuticals; phosphine and arsine in the production of semiconductors; and hydrogen cyanide released when consumer packaging materials are burned. Therefore, it is important to know whether toxic gas concentrations are dangerous.

In the United States, the National Institute for Occupational Safety and Health (NIOSH) and the American Conference of Governmental Industrial Hygienists (ACGIH) have established short-term and long-term exposure limits for many toxic industrial gases. "Threshold Limit Value-Time Weighted Average" (TLV-TWA) refers to the time-weighted average concentration to which most workers can be repeatedly exposed during a normal 8-hour workday without suffering harmful effects. "Threshold Limit Value-Short-term Exposure Limit" (TLV-STEL) refers to the concentration to which most workers can be exposed for a short period of time without suffering irritation or injury. "Immediately Threatening Life or Health Concentration" (IDLHC) is a limiting concentration that poses an immediate or slow threat to life, causes irreversible health damage, or affects the worker's ability to escape independently. Table 1 lists the limits for several common gases.

Table 1: Exposure limits for some common industrial toxic gases

Electrochemical sensors offer several advantages for instruments that detect or measure toxic gas concentrations. Most sensors are designed for specific gases, have a usable resolution of less than one part per million (1 PPM) of gas concentration, and require very low operating current, making them ideal for portable, battery-powered instruments. An important characteristic of electrochemical sensors is their slow response: after first powering up, it may take several minutes for the sensor to settle to its final output value; when exposed to mid-range gas concentrations, it may take 25 to 40 seconds for the sensor to reach 90% of its final output value.

This article describes a portable carbon monoxide (CO) detector that uses an electrochemical sensor. Carbon monoxide has a much higher IDLH concentration than most other toxic gases, making it relatively safer to handle. However, carbon monoxide is still a deadly gas, and extreme caution and proper ventilation should be used when testing the circuit described in this article.

Figure 1 shows the CO-AX sensor from Alphasense. Table 2 is a summary of the technical specifications of the CO-AX sensor.

Figure 1. CO-AX carbon monoxide sensor

Table 2. CO-AX sensor specifications

For portable instruments in this application, achieving maximum battery life is the most important goal, so minimizing power consumption is critical. In a typical low-power system, the measurement circuit is powered up, performs a single measurement, and then shuts down to a long standby state. However, in this application, the measurement circuit must remain powered up at all times due to the long time constants of the electrochemical sensors. Fortunately, because of the slow response, we can use micropower amplifiers, high-value resistors, and low-frequency filters to minimize Johnson noise and 1/f noise. In addition, single-supply operation avoids the power waste of bipolar supplies.

Figure 2 shows the circuit of this portable gas detector. The dual-channel micropower amplifier ADA4505-2 is used in the constant potential configuration (U2-A) and the transconductance configuration (U2-B). The power consumption and input bias current of this amplifier are very low, which is a good choice for both the constant potential section and the transconductance section. Each amplifier consumes only 10 μA, so the battery life is very long.

Figure 2: Portable gas detector using electrochemical sensor

In a three-electrode electrochemical sensor, the target gas diffuses into the sensor, passes through a thin film, and acts on the working electrode (WE). A constant potential circuit detects the voltage of the reference electrode (RE) and provides current to the auxiliary electrode (CE) to keep the voltage between the RE and WE terminals constant. No current flows into or out of the RE terminal, so the current flowing out of the CE terminal flows into the WE terminal, and this current is proportional to the concentration of the target gas. The current flowing through the WE terminal can be positive or negative, depending on whether a reduction or oxidation reaction occurs in the sensor. For carbon monoxide, when oxidation occurs, the current at the CE terminal is negative (the current flows into the output of the constant potential op amp). Resistor R4 is usually very small, so the voltage at the WE terminal is approximately equal to VREF.

The current flowing into the WE terminal causes a negative voltage to be generated at the output of U2-A with respect to the WE terminal. For a carbon monoxide sensor, this voltage is typically several hundred millivolts, but for other types of sensors, this voltage can be as high as 1 V. To operate from a single supply, a micropower ADR291 reference voltage source (U1) boosts the entire circuit to 2.5 V above ground. The ADR291 consumes only 12 μA of current; it also provides the reference voltage that allows an analog-to-digital converter to digitize the output of this circuit.

The output voltage of the transconductance amplifier is:

in:

I _WE is the current flowing into the WE terminal.

Rf is _the transconductance resistor (shown as U4 in Figure 2).

The maximum response of the sensor is 90 nA/ppm, as shown in Table 2, and its maximum input range is 2,000 ppm. Therefore, the maximum output current is 180 μA, and the maximum output voltage is determined by the transconductance resistance, as shown in Equation 2.

Sensors for different gases or from different manufacturers have different current output ranges. If U4 uses the AD5271 programmable variable resistor instead of a fixed resistor, the same structure and materials can be used for different gas sensors. In addition, such a product also supports sensor replacement because the microcontroller can set the AD5271 to the appropriate resistance value for different gas sensors. The temperature coefficient of the AD5271 is 5 ppm/°C, which is better than most discrete resistors; its power supply current is 1 μA, which has little impact on system power consumption.

When operating from a single 5 V supply, Equation 1 shows that the output range of transconductance amplifier U2-B is 2.5 V. Setting the AD5271 to 12.5 kΩ allows the worst-case sensitivity range of the sensor to be utilized, while providing approximately 10% overrange capability.

Using a typical sensor response of 65 nA/ppm, the output voltage can be converted to ppm of carbon monoxide using the following equation:

When using a differential input ADC, simply connect the 2.5 V reference output to the ADC's AIN- terminal, eliminating the 2.5 V term in Equation 3.

Resistor R4 keeps the noise gain of the transconductance amplifier at a reasonable level. The value of R4 is a trade-off between the magnitude of the noise gain and the settling time error of the sensor when exposed to high gas concentrations. For this circuit, R4 = 33 Ω, which allows the noise gain to be calculated to be 380, as shown in Equation 4.

The input noise of the transimpedance amplifier should be multiplied by this gain. The 0.1 Hz to 10 Hz input voltage noise of the ADA4505-2 is 2.95 μV _pp , so the noise at the output is:

This output noise is equivalent to a gas concentration above 1.3 ppm pp, and this low-frequency noise is difficult to filter out. Fortunately, the sensor response is very slow, so the low-pass filter formed by R5 and C6 can have a cutoff frequency of 0.16 Hz. The time constant of this filter is 1 second, which is negligible compared to the 30-second response time of the sensor.

Q1 is a P-channel JFET. When the circuit starts, the gate voltage is VCC and the transistor is turned off. When the system shuts down, the gate voltage drops to 0 V and the JFET turns on, keeping the RE and WE terminals at the same potential. This can greatly improve the sensor's turn-on settling time when the circuit starts again.

The circuit is powered by two AAA batteries. Using diodes for reverse voltage protection wastes valuable power, so this circuit uses a P-channel MOSFET (Q2). The MOSFET protects the circuit by blocking reverse voltage and turns on when a positive voltage is applied. The on-resistance of the MOSFET is less than 100 mΩ, so it causes much less voltage drop than a diode. In addition to the AAA batteries, the ADP2503 buck-boost regulator allows the use of an external power supply up to 5.5 V. When operating in power-save mode, the ADP2503 consumes only 38 μA. The filter formed by L2, C12, and C13 removes any switching noise from the analog power rail. When the external power supply is connected, instead of using a circuit to disconnect the battery, the instrument uses a jack to mechanically disconnect the battery, thereby avoiding power waste.

Using AAA batteries, total power consumption is approximately 100 μA in normal conditions (no gas detected) and 428 μA in worst-case conditions (2,000 ppm CO detected). If the meter is connected to a microcontroller and can be put into a low-power standby mode when not taking measurements, battery life can reach over a year.