Solenoid Trip Circuit Works at Battery's End of Life

Texas Instruments LP339 TLE2426

This Design Idea involves a low-power motion-detector circuit that operates from battery power for extended times. Part of the design includes a solenoid-operated trip mechanism that triggers whenever it detects motion. The drive circuit for the solenoid in the original design worked fine as long as the battery was fresh but failed as the battery got into the middle of its life, even though sufficient energy remained in the battery to operate the solenoid. The culprit was the battery's internal resistance. The internal resistance of a standard alkaline cell increases as the cell's life accumulates, whereas the cell's open-circuit voltage hardly changes. This increased resistance causes a sharp drop in supply voltage whenever the drive circuit attempts to energize the solenoid. This drop upsets the drive circuit, preventing reliable operation.

The original design solved this problem by using a large electrolytic capacitor across the battery supply. This capacitor functioned as an energy reservoir and prevented the supply voltage from sagging so dramatically, allowing the device to continue functioning much further into the battery's life. However, the electrolytic was bulky and expensive, and presented an awkward packaging problem.

The circuit in Figure 1 solves the problem by incorporating feedback in such a way that any drop in supply voltage only turns on the drive circuit harder. In testing, this circuit functioned even when the nominally 6 V battery sags as low as 2 V. Q1 is a 3 A, high-beta, low-saturation-voltage pnp transistor that drives the solenoid. To energize the solenoid, Q1 has to turn on hard to minimize voltage drop and get the most from the battery's life. The circuit achieves the full turn-on by using three sections of an LP339 quad comparator in parallel. The LP339 is similar to the venerable LM339 but with lower power consumption, making it more suitable for battery-powered applications. Interestingly enough, it also has higher output drive. The three parallel devices provide approximately 200 mA to the base of Q1, sufficient for Q1 to supply 2 A to the solenoid and remain well-saturated. The design requires no current-limiting resistor in series with the base of Q1 because the outputs of the LP339 are naturally current-limited to approximately 60 or 70 mA each.

This motion-detector circuit works with 6 V battery voltages that sag as low as 2 V.
Figure 1. This motion-detector circuit works with 6 V battery voltages that sag as low as 2 V.

Once the trip mechanism triggers, S1 opens, removing all power from the circuit. The device remains in this state until you manually reset the trip mechanism. An entire trip event takes only about 10 msec, thus conserving the battery and preventing excessive power dissipation in Q1 and the LP339. The remaining section of the LP339 implements a single-section window comparator. A window comparator is necessary because motion detectors are commonly ac-coupled circuits. You must apply the detection threshold equally to positive or negative excursions. In other words, any excursion outside the window should trigger detection. In a quiescent state with no input signal, the network comprising R3, R5, and dual-diode D2 keeps the negative input of IC1A one diode drop above ground and the positive input one diode drop below ground. This level keeps IC1A’s output and, hence, the negative inputs of IC1B, IC1C, and IC1D, low. Because R1 and R6 bias the positive input of IC1B, IC1C, and IC1D at approximately 0.75 V, the comparators' outputs remain off.

The circuit trips if the input goes more than two diode drops above or below ground. For example, if the circuit's input goes below ground, the negative input of IC1A pulls down until it is lower than the positive input. If the input goes up, the positive input of IC1A pulls up until it is higher than the negative input. Either case results in the output of IC1A’s going high. This action in turn causes the IC1B, IC1C, and IC1D outputs to turn on, turning on Q1 and energizing the solenoid. R4 provides positive feedback. If Q1 even starts to turn on, the partial turn-on causes IC1B, IC1C, and IC1D to turn on more, precipitating a latch-up to ensure that Q1 turns on all the way. Thus, all possible battery energy is delivered to the solenoid. C1 provides further positive feedback when the supply sags. D3 protects the LP339's inputs from being driven more than a diode drop below ground when the power-supply sag is severe.

A beneficial side effect of the time constant formed by C1 and R6 is to prevent false triggers on start-up by holding off the IC1B, IC1C, and IC1D stage until the motion detector has had time to stabilize. IC2 is a “ground-generator” chip, used to create a circuit ground midway between the supply rails. In contrast to the original circuit, everything in this circuit is going in the proper direction to ensure positive actuation of the solenoid once it reaches a trip threshold. The circuit uses only common, low-cost, and small components. The circuit offers full usage of battery life and needs no bulky energy-reservoir capacitor.

Materials on the topic

  1. Datasheet Texas Instruments LP339
  2. Datasheet Texas Instruments TLE2426
  3. Datasheet ON Semiconductor 1N4935
  4. Datasheet Nexperia BAV99
  5. Datasheet Diodes FZT788B

EDN

EMS supplier