Design Ideas: March 14, 1996
A sensitive and reliable way to measure airflow is to exploit the predictable relationship between airspeed and the heat dissipated by a sensor exposed to the flow while being held at a constant temperature differential above ambient. The power required to maintain the elevated sensor temperature is proportional to the square root of the airspeed (King's law). The popular hot-wire anemometer uses this principle, but suffers from the disadvantage of using a relatively fragile airflow sensor, a thin metallic filament. The circuit in Figure 1 uses a robust and inexpensive transistor, Q1, as the heated airflow sensor.To understand the operation, first consider the case of zero airflow. You use the zero-adjust trimmer R2 to set the quiescent base-bias currents for Q1 and the reference transistor Q2. With the proper adjustment, Q1's temperature rise (~50°C) in still air, caused by collector power dissipation, reduces Q1's VBE (by ~2 mV/°C) to just slightly below Q2's VBE plus the drop across R1. The noninverting input of comparator IC1 is then slightly less positive than the inverting input. The output therefore switches low, holding C1 discharged and resetting multivibrator IC2, whose output goes high. This condition does two things: It produces a zero-frequency output and holds Q3 off.
Now imagine some airflow directed at Q1. The resulting increase in cooling rate tends to reduce Q1's temperature, causing its VBE to increase relative to that of Q2. The relation between IC1's inputs reverses, and the comparator releases the reset on C1. C1 then charges through R4 and turns on Q3. Then Q3 applies a ~700-µsec pulse to Q1's base through the full-scale trimmer R3. The resultant pulse of collector current forced in Q1 deposits a quantum of heat, which tends to return Q1's temperature to a value hot enough to restore the original zero-flow voltage balance with ambient-sensor Q2. Until Q1 achieves that temperature, IC2 continues to oscillate and cycle Q3 on.
Thus, a feedback loop exists that acts to maintain a constant temperature differential between Q1 and Q2. The average frequency appearing at Q2's output is therefore proportional to the extra power required to heat Q1 and, by King's law, to the square root of the airspeed. The maximum output frequency for the circuit values in Figure 1 is 1 kHz. Appropriate adjustment of R3 establishes almost any desired full-scale flow rate from <1 to >30 meters/sec. Note that the tracking between the Q1 and Q2 VBE voltages provides good compensation for changes in ambient temperature. Power dissipation is dependent on airflow rates, but is typically modest: <1W. (DI #1841)