How an outdoor motion PIR sensor switch works with schematic

In this article I will be explaining how a pyroelectric (PIR) sensor works and show the reverse engineered schematic simulated in LTspice. The schematic is for NV-1111.35 outdoor PIR sensor used to switch a mains light and has 3 potentiometers for setting SENSITIVITY, LUX and TIMEOUT.

This sensor is based on a popular three stage op-amp topology. Understanding this can help you understand other pyroelectric sensors as well and also something about active filters using LM324 and removing and setting a DC voltage bias.

 

How a PIR sensor work

This particular sensor board is using the D203S PIR sensor. Pyroelectric passive infrared (PIR) sensors detect infrared (IR) radiation. A warm object emits infrared energy that is not visible with a human eye. Since the purpose of the sensor is to detect motion not only heat, there are two parts sensitive to IR that look like two little windows or slots. 

The sensor window is actually split into two parts

This two elements are used in series with opposite polarity so the average radiation is cancelled out. When the sensor is idle both slots detect the same amount of IR so the ambient temperature won't trigger the sensor. If a warm body like a human or animal passes by the infrared energy it is first detected by an element then by the second one and so generating an AC signal. A mosfet is also included to buffer the weak signal produced by the PIR sensor.

PIR sensor diagram

 

Image from AN4368 - Signal conditioning for pyroelectric passive infrared (PIR) sensors

The signal produced by the PIR sensor is around 1mVpp and it has a DC offset voltage that can vary from 0.3V to 1.2V. 

Fresnel lenses

In order for the sensor to cover a wide angle, Fresnel lenses are mounted in front of it. They are cleverly designed with many areas of Fresnel lenses arranged in a way so they can split the detection area into multiple sections. This for the sensor looks like it has not only two detection elements but many pairs of them. So even a small movement can trigger both elements.

Image from D203S datasheet

PIR sensor circuit explanation and schematic using the LM324N op-amp

The board can be powered from 5V to 12V however the value of the components are designed for 8V to 12V. It can work on 5V but the voltage division created by R17 and R18 must be changed.

Power consumption is very low: 1.5-2mA.

Parts of solder mask were burned due to the power board short-circuit but only the op-amp was damaged on the main board

Outdoor PIR sensor board schematic (click to enlarge)

This op-amp architecture consists of three stages.

Stage 1

The first architectural stage amplifies the signal. It cancels the DC part of the signal and filters the high frequency noise that could lead to false detections.

 

The high frequencies are filtered by C4 and R4 and the cut-off frequency is 2.2 Hz (fhigh1 = 1 / (2 x pi x R4 x C4)).

The second filter is used to reject the DC part of the signal. C1 and R3 perform a high pass filter that has a cut-off frequency of 0.34 Hz (flow1 = 1 / (2 x pi x R3 x C1)).

The stage gain is 221 (Gain = 1 + (R4 / R3)). The gain must be high enough to amplify the sensor signal above the noise level but not too high to drive the op-amp into saturation. The amplification is made around a common mode voltage set by the sensor and is not VCC / 2. 

 

Stage 2 - Setting PIR sensor sensitivity

 

Stage 2 is very similar to stage 1. It is used to filter and amplify the AC signal except that this time the signal is inverted.

The signal Vout1 from stage 1 enters stage 2 through a 10k potentiometer that is used to set the sensitivity of the PIR sensor so it does not trigger every time a dog passes by.

Regarding the filtering function, the low and high cut-off frequencies respectively are 0.72 Hz (flow2 = 1 / (2 x π x R6 x C5)) and 4.8 Hz (fhigh2 = 1 / (2 x π x R10 x C6)). The flow2 also depends on the potentiometer position.

The gain of this stage is -100 (minus represents the inverted signal) (Gain2 = -R10/R6). This gain means that after stage 2 the signal between 0.7 Hz and 4.8 Hz will have been amplified 22100 times (87 dB).

In this stage the op-amp common mode voltage is set by the resistors R12, R9 and R8 to 37% of the power supply (Ratio = R8 / (R12 + R9 + R8)). The gain is intentionally set high so the signal is clipping looking more like a square wave now.

Stage 3 - Setting PIR sensor timeout and light threshold

In this last stage 3 the signal from stage 2 is again inverted and filtered by the R11 and C7 with a 3.38 Hz cut-off frequency but this time the last two op-amp are used mostly as comparators. The non-inverting input is set by R12, R9 and R8 at 44% of VCC, one diode voltage drop above the ratio in stage 2.

The R16 potentiometer together with the LDR controls at what light intensity the sensor will trigger. The more light the lower the resistance of the LDR will be and if the pot is lower the base of Q1 will be pulled low turning off the transistor and R13 will let the diode D1 conducting the sensor signal to ground. If the potentiometer is on the other side even if the LDR would have 0 resistance the 150k of the pot would make the Q1 turn on then R13 will develop a voltage drop of VCC / 2 which is higher than the output voltage from stage 2 and the sensor will trigger. 

 

The last op-amp has the non-inverting input set at 83% of VCC. When no movement is detected Vout3 is always high and so the inverting pin of U1.4 is higher than (+) pin making the output to relay low. When Vout3 is low the negative plate of C8 is pulled to ground and so (-) pin is lower than (+) pin making the output high turning on the transistor that controls the power relay. Now even if Vout3 will be a very short pulse or the sensor will detect movement multiple times, the output will still remain high. If Vout 3 will keep going low to to motion detection the timeout will be reset but the relay will remain on. So when Vout3 remains high the capacitor C8 will be charged through R20 and the pot R19 that sets the timeout.

This method of using a capacitor was new to me and it took me a while to understand it. after the C8 charges to almost VCC the (-) pin will be higher than (+) and the output will go low turning off the relay. 

 

PIR sensor signal conditioning ploted in LTspice

The schematic was simulated in LTspice. In the first plot there is Vpir that represents the 1mV signal from the sensor around 500mV offset voltage.

In the second plot Vout1 is the amplified signal after stage 1.

Vout2 is the output from stage 2. Here the signal is inverted, amplified to the maximum op-amp output and the offset voltage is set to 2.96 V at 8V power supply.

Vout3 is the output from stage 3 and is normally high. When motion is detected and signal Vout2 is higher than 3.55V then Vout3 will go low. Notice that it only uses the second half of the PIR signal pulse.

Finally Vrelay goes high when Vout3 is low. It will remain high even if Vout3 will change states again. If Vout3 remains stable at a high level then Vrelay will remain high depending on the RC constant set by C8, R20 and potentiometer R19. 

 

PIR sensor simulated in LTspice

In the next tutorial I will be presenting the capacitive power supply.

Download

PIR sensor LTSpice simulation

PIR sensor schematic and PCB in DipTrace