Maker Pro
Maker Pro

Could you attach SLR Lens to a PIR motion sensor?

I'm currantly working on an electronics project employing an arduno and pir sensor to trigger a high speed camera. Could you attach a SLR telephoto lens to the sensor for coverage of a select location? Would it work or not?

Thanks in advance,

James LePage
 
I am not an optical specialist, but I can tell you that PIR sensors use infrared radiation rather than the visible radiation that the lense is designed for. The infrared light has longer wavelengths than visible light, which I think means that the lens will focus the light at a different distance.

I skimmed this, it looks as though this has some information on infrared focusing.

An alternative solution is to purchase a PIR with a smaller field of vision, Darnell provides a filter for PIR sensors with a 30 degree field of vision.

Another solution is to make a cone shape, though parabolas and other assorted bolas may become a problem when the cone is a very accute angle.
 

hevans1944

Hop - AC8NS
... Could you attach a SLR telephoto lens to the sensor for coverage of a select location? Would it work or not?
James, you may need to think outside the box on this one. If the goal is to trigger your camera when there is movement in a certain area within the field-of-view of the camera, a passive infrared (PIR) sensor is just one solution. True, it does have the advantage of responding only to movement of objects that are hotter (or colder) than the background ambient temperature, and that may be an unstated requirement, but it is possible to sense motion by photographing and storing successive frames of a video camera "looking" at the same FOV as the camera you want to trigger. There is hardware and software available to accomplish this task, comparing successive frames of video and triggering an alarm when they differ. This technology is mainly directed at the security industry, so it may cost more than you are willing to pony up just to take a few candid snapshots from afar.

If your scene is static (no background movement), it is fairly easy to process a raw NTSC video stream to recognize when it changes. This processing is typically analog and can be performed on a frame-by-frame basis or for a particular line, or lines, of video. That's a lot easier than digitizing a whole frame of video and then comparing the pixels in the next frame. However, being simple 20th Century technology, it is more susceptible to false positives.

Some sophisticated software algorithms and powerful pixel processors are needed for high quality motion detection (hence the expensive part) because there are always minute differences between time-adjacent pixels in the same spatial location. The usual solution is to define "region of comparison" surrounding each pixel in the two successive frames to be processed. There will be hundreds or even thousands of these regions of comparison to be processed by the motion detection algorithm, and the processing must be done on a frame-by-frame basis in real time. Piece of cake with 21st Century technology. That's why inexpensive systems opt for the PIR sensor approach (which is also late 20th Century technology).

A PIR sensor typically has a small array of Fresnel plastic lenses located in front of a pair of pyroelectric detectors. Each lens focuses a different point-of-view, within the overall field-of-view, onto the pair of sensors. Because the sensor elements are wired in a "series opposing" manner, their outputs cancel unless one sensor receives more (or less) radiation than the other. The fresnel lenses help with this by creating two (or more) slightly separated spatial images. An object entering from the side will change the energy impinging on one pyroelectric detector first without affecting the other detector and therefore an output signal that is not subtracted by the other detector occurs. When the warm object illuminates both detectors no output occurs. As it exits the FOV another pulse of opposite polarity occurs from the second detector. There is a pretty good explanation and an example circuit to be found here.

So, to answer your question: it is probably feasible to use an SLR camera telephoto lens to focus a distant area onto the Fresnel lens in front of the pair of detectors. It won't be very efficient because the telephoto lens elements are designed and corrected for visible light. They may also have infra-red absorbing coatings. Still, enough radiation will probably get through the lens to be useful. The problem will be getting the focus right and knowing where the target area is. I suppose a pair of walky-talkies and a helper to walk through the target scene could help with that. If you open the camera back and temporarily install the PIR sensor at the film plane, that might work. It's simple and worth a try, plus you get some idea of what you are looking at while the SLR camera mirror is down. Need the mirror up when in use of course.
 
Last edited:
Top