Wearing specs and paying attention but I didn't see the light until Dippy "teeheed" me and made me dig a little deeper between the lines.
<A href='http://www.glolab.com/focusdevices/focus.html' Target=_Blank>External Web Link</a>
gives the hint I was looking for:
<i>A commonly used pyroelectric infrared sensor has two sensing elements internally connected in a voltage bucking configuration. </i>
So when exposed to a warm IR source, one of the two elements generates a positive polarity output signal while the other generates a negative going signal.
The module output signal is the sum of these two internal out of phase signals .
This is useful for the background radiation is detected equally on both elements and the two opposite signals cancel out. If one element is exposed to a PIR signal while the other element is not, it generates a signal which is not canceled and which appears on the module output pin.
As shown in the 3rd graphic, the module output signal reacts to a warm object entering the field of view and being detected first on the right element, by generating a positive pulse first while an object entering the field of view detected first on the left element generates a negative pulse first.
The polarity of the first pulse could therefore provide direction information.
For a warm object leaving the field the polarity of the last pulse could also be used for direction info.
Note that the polarity of the first and last pulse from each element are reversed.
Normally, a multi-facetted Fresnel lens is used because this lens causes many tiny moving images of warm objects in the field of view to be projected on on off the two PIR sensor elements. This results in a stream of positive and negative pulses from the module output making it easy to detect movement in the field of view.
This +/- pulse train of an object moving through middle of the field of view would be difficult to analyze.
A simple lens, pinhole, mask or baffle makes the output signals of the dual elements easier to analyze but reduces sensitivity.
When dealing with a pulse train, features such as differences in time between pulses or relative amplitudes might be used to determine which pulse belongs to which element. Another method might visually chop or superimpose a carrier signal or diffuse the aperture of one of the two elements to differentiate its output signal from the other element.
So it is possible but more difficult to use the summed output of a single PIR module with dual elements to determine direction of motion because it is not easy to determine which element is active at any time especially if signals overlap.
The analog signals from two separate PIR module outputs could be processed much more easily, for example, by measuring the phase relationship of the two pulse signals to determine direction and speed.
Thanks Dippy, next time I'll dig first and comment later to spare you another bout of the teehees.
On that note, to Dippy and everyone else on this wonderful forum: Enjoy! and not just now but always, goodwill to one and all.
wilf
Edited by - wilf_nv on 24/12/2006 03:17:23