12
\$\begingroup\$

I would like to trigger a PIR sensor using an infrared LED. Is that possible? My goal is to simulate motion in front of the PIR, when there otherwise is none. I tried using a handheld remote control, but that didn't seem to work.

\$\endgroup\$
  • 1
    \$\begingroup\$ Have you tried googling how they work and how big a signal might be expected from someone moving about (compared to the signal from a remote control)? \$\endgroup\$ – Andy aka Apr 12 '13 at 20:56
  • 1
    \$\begingroup\$ Add what @Andyaka and rawbrawb say. Acting on Andy's assistive adumbration Google for Fresnel lenses and zones and PIRs and you'll see what he is referring to. Then you'll need a source that matches what the detector sees as IR as rawbrawb says. Many of these look useful I'd guess you'll need two - or a beam splitter and interupter. \$\endgroup\$ – Russell McMahon Apr 13 '13 at 4:48
  • \$\begingroup\$ Do you have to trigger it wirelessly? And can you modify the PIR sensor? If you can, you could simply tap into the PIR sensor's output pin. \$\endgroup\$ – Passerby Apr 13 '13 at 5:31
  • \$\begingroup\$ In response to the @Passerby suggestion to tap the PIR output pin, I was hoping to trip an off the shelf, in-wall, motion controlled light switch that is powered at 120VAC. As a safety precaution, I had ruled out the option of opening up the switch. \$\endgroup\$ – user1574082 Apr 13 '13 at 13:02
20
\$\begingroup\$

People often use imprecise terms to describe InfraRed and it is further complicated in that with respect to Infrared "that word infrared ... I don't think it means what you think it means" in a lot of cases.

PIR - AKA Passive infrared sensors are pryoelectric devices that are optimized to detect Mammalian body temperatures (around 300 K), these warm bodies emit light at around the 10µm to 14µm wavelength range. Some people call this the "mid Infrared" but the trend is towards using the term "Thermal InfraRed" or TIR.

The IR LED you are using emits probably around 900 nm -> 750 nm - so close to 1µm in wavelength. Some people call this "Near IR" but those people also tend to call 2 - 4 um wavelngth range the "Mid Infrared also". Confusing? yep. It comes from a different historical use. One from the military one from chemistry/astronomy.

So you are at least a factor of 10X away in wavelength terms. And a LED emits light in a very, very narrow band of energies (it is an electronic effect after all)

Also PIR's are "designed" to detect rather largish bodies, which means a fair amount of energy or photon flux.

A black body emitter will increase energy in all wavelengths with increasing temperature. So an emitter at 27 C (300 K) will emit less light at 10µm than a emitter that is at 100 C (373 K) in the same band of energies. So if you want to have an emitter that will trigger the PIR, make a temperature controlled emitter, run it at 100 C to be safe, and it will emit a lot more light in the 10 - 14µm band than a body temperature device.

On second thought make it 70 C just to be safe. 100 C is a little too hot. Read up about blackbody emitters for fun.

\$\endgroup\$
  • \$\begingroup\$ Thanks for the clear and detailed explanation. Very helpful! \$\endgroup\$ – user1574082 Apr 13 '13 at 13:07
  • 3
    \$\begingroup\$ You may also need to turn it on and off at the right frequency. I've seen a design with a bandpass around 2Hz (0.4 to 10 Hz IIRC). And you need to position the emitter so only one of the two sensor halves in the PIR sees it, as they cancel each other. \$\endgroup\$ – starblue Apr 14 '13 at 13:13
  • 1
    \$\begingroup\$ You will get much more reliable results if you have two emitters. \$\endgroup\$ – KalleMP Oct 3 '15 at 1:01
  • \$\begingroup\$ Could someone provide a link to a said emitter \$\endgroup\$ – uhfocuz Feb 5 '16 at 22:59
  • \$\begingroup\$ @uhfocuz A longwave IR emitter is a filament lamp that is barely glowing. \$\endgroup\$ – KalleMP Jan 13 at 0:18
-1
\$\begingroup\$

I remember running a mains filament lamp using 12 volts dc ... the filament gave out an attractive low glow with infra red being emitted and not so much light. The current used was quite high so a bench power pack was needed. ... I guess about 3 amps was used. If you use different mains rated wattage lamps you should soon find the right one. This should trigger your PIR device at the flick of the switch on and off. I did find that the PIR detector devices tend to trigger when the infra red producing source is removed and that movement was not needed if the location of the lamp was right. Take care not to burn yourself on the lamp as you may not be use to using a filament lamp. You could also try using a filament lamp at full operating voltage but the light emitted might be annoying. Carbon filament lamps are making a come back ... these could well work better for Infra red emission than the tungsten filament lamps.

\$\endgroup\$
5
\$\begingroup\$

You may be able to cause it to trigger if you place two filament bulbs operated under-voltage at a dim glow in front of it, you may have to tune the positioning a bit to get it right. Then switch them on alternately hoping that they will illuminate the alternate parts of the sensor. The spacing of the bulbs should be related to the detection 'beam' angle which can be different in different directions and zones and obviously will vary by model and manufacturer as well. The alternating frequency will need to be made to match the expected signal it would get if a person was crossing the detection 'beams' and getting it to match walking speed.

EDIT:
A domestic PIR sensor is designed to detect moving IR sources. They generally have a two active area detectors that have to have the areas alternately illuminated to create an alternating differential (pyroelectrically generated voltage) so they can detect a warm body moving but not a stationary heater. The alternating signal is normally achieved with the multi-zone Fresnel lens (or reflector) found on this type of sensor. To spoof it you need to place two sources that illuminate the two zones unevenly and then alternate their signal to mimic the flickering pattern that a moving person would create. A single source may not cause the proper alternating signal to be generated and be filtered out as spurious environmental changes.

\$\endgroup\$
  • \$\begingroup\$ This could be a good solution provided that he wasn't trying to switch a light in the first place! \$\endgroup\$ – Transistor Sep 30 '15 at 18:01
  • \$\begingroup\$ I understood OP was switching on say a room light. My method suggests alternately illuminating tiny pilot lamps that can be used to mimic a moving heat source. \$\endgroup\$ – KalleMP Oct 3 '15 at 0:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.