The Motion Detector the Cat Kept Setting Off

Why this matters

A living room motion detector sent alarms on weekdays with nobody home, roughly once a week for five weeks. The first visit named the cause correctly, made a sensible change, and the alarms came back. The detector was the right model at the right setting for the household's cat. It was mounted beside a bookshelf, and a pet-immune detector only ignores an animal that stays where the detector expects animals to be. Two visits, three police responses and a customer close to cancelling, for a fault that was decided the day the detector went on the wall.

The first visit, and why it was right

The signal history was plain. Five alarms over five weeks, all on zone 9, the living room motion detector, all between 10:00 and 15:00 on weekdays, with the system armed away. No door or window zone opened before any of them. Two were dispatched after the station's calls went unanswered, three were cancelled by the homeowner from work.

The household had one cat, a little over 11 lb. The detector was a pet-immune passive infrared unit (PIR, which responds to a moving heat source crossing its zones) whose maker rated it for pets well above that weight. The tech checked the setting switches against the installation sheet and found them set for the heaviest pet the unit supported, which covered the cat with room to spare. The mounting height matched the height the instructions specified, and the detector was level and aimed out across the room as designed.

So the tech reasoned, correctly, that a pet-immune detector at the right setting alarms on a cat only when the cat gets somewhere the immunity does not cover. Standing in the room, the obvious candidate was a carpeted cat tree about 5 ft tall, 4 ft from the detector, with a platform at the top where the cat slept. A cat on that platform was at the height of the detector's upper zones and close to the lens.

The cat tree moved to the far side of the room, beside the sofa. The homeowner was told why. The ticket said "pet tree relocated out of detector near field." That was a correct diagnosis and a correct repair.

Nine days later

The sixth alarm arrived on a Thursday at 13:40, zone 9 again, nobody home, dispatched. The homeowner's message to the office said what most customers say at this point: the dealer had already fixed this once and they wanted the detector removed.

The second tech read the first ticket before going. The question was not whether the cat was the cause. Nothing else in the record pointed anywhere else. The question was where the cat was now.

Where the cat went

The detector was mounted in the corner beside a tall bookshelf, its top shelf at about 6 ft with a few inches of clear space above the books. The homeowner, asked where the cat spent its days now the tree was gone, said it had started sleeping on top of the bookshelf. It reached it by jumping from the arm of a reading chair to the middle shelf, then to the top.

 SIDE ELEVATION AT THE DETECTOR WALL

 detector at mounting height  [PIR]
                               |\
                               | \  upper zones look
                               |  \ across the room
                               |   \ and down to the
 top of bookshelf, about 6 ft  |##  \ far floor
   cat sleeps here, within 2 ft|##
   of the lens                 |##
 middle shelf, about 4 ft    --|##   <- jump 2
                               |##
 reading chair arm, about 2 ft-|##   <- jump 1
                               |##
 floor (pet zone, ignored)   --+--------------------

That is the whole fault. A pet-immune detector ignores pets mainly through geometry and signal size. Its lens pattern and processing are built so that a pet moving across the floor produces a small signal the detector discounts, while a person standing upright crosses the upper zones and produces a large one. A cat on a shelf near the detector is no longer a small, low target. It is in the upper zones, and it is so close to the lens that it fills as much of a zone as an adult across the room. When it stands, stretches or jumps down, the detector sees what it was designed to report. See related: Pet-Immune Motion Detectors and Where They Still Fail.

The first visit's change had been right and had made things slightly worse. With the cat tree gone, the cat looked for the next high perch, and the next high perch was directly beside the detector.

Before testing: the account on test

The tech called the central station, placed the account on test for the visit, and had the operator confirm it back. Walk-testing a motion detector with the system armed, or a cat jumping off a shelf during the visit, would otherwise have been the seventh alarm. See related: How to Put an Alarm Account on Test and Take It Off.

Proving it with the cat

A tech cannot make a cat climb on cue, but a homeowner with treats usually can. The detector was put in walk-test mode so its LED lit on each detection.

• Cat on the floor, walking across the room: LED dark on every pass. The immunity was working.
• Cat on the reading chair arm, about 2 ft up and 6 ft from the detector: LED dark.
• Cat on the bookshelf top, sitting: dark.
• Cat standing and turning on the bookshelf top: LED lit, twice in about a minute.
• Cat jumping down from the top shelf to the middle shelf: LED lit.

That was enough. The detector was doing exactly what it was built to do with a target it was never built to ignore.

Options that were not taken

Lower the sensitivity or raise the pulse count. Many detectors let you require more zone crossings before an alarm. That reduces pet alarms by reducing detection of everything, including a person moving slowly or only crossing part of the room. It treats the symptom with a weaker detector.

Remove the bookshelf or block the top. It is the customer's living room. A dealer who fixes a false alarm by asking someone to rearrange their furniture gets the alarm back the next time the furniture moves.

Change to a dual-technology detector. A unit that requires both its PIR and microwave sections to agree resists some false alarms well, but a cat moving within 2 ft of the lens can satisfy both. It does not fix placement.

Take the detector off the arming plan. The homeowner offered this. A living room with three ground-floor windows and the slider to the deck was the room an intruder was most likely to cross, and the windows had contacts but no glass-break coverage. Bypassing the only interior device there left a gap the customer did not understand.

Where the detector went

The fix was to move the detector to a wall where nothing climbable sat within the detector's near field, and aim it so the cat's likely perches were in its long-range zones near the floor, or outside the pattern entirely.

The tech walked the room with the coverage pattern from the installation sheet in hand, looking at every surface a cat could reach above about 3 ft.

• The bookshelf corner: out. Climbable to 6 ft.
• The wall above the sofa: out. The sofa back was low, but the sofa sat 3 ft from that wall and the cat slept on the back of it.
• The corner between the deck slider and the fireplace wall: chosen. The nearest furniture in the field was the reading chair at about 14 ft. The bookshelf was now on the far wall, where the detector's zones reached it low, near the floor, and its top was outside the pattern.

The detector went at the height its instructions specify, aimed across the room so a person entering from either the slider or the hall had to walk across the zones rather than straight at the lens. Walking across is what a PIR detects best, and coverage had to remain after the move, not just immunity. See related: Motion Detector Coverage and the Walk-Across Rule.

The wire was fished to the new corner through the basement ceiling, and the zone was re-terminated at the detector with its end-of-line resistor at the device. The old location was patched and painted to match.

The finish test

Still on test, the homeowner and the treats ran the same sequence.

• Cat on the bookshelf top, sitting, standing, turning: LED dark over about five minutes.
• Cat jumping down from the bookshelf: dark.
• Cat on the back of the sofa, now about 16 ft from the detector: dark.
• The tech walking across the room from the slider, then from the hall, at normal pace: LED lit within two steps each time.
• The tech walking slowly from the hall toward the slider: lit before the middle of the room.

He armed the system away, walked in from the hall, let the alarm report, and confirmed with the station that zone 9 was received under the test window. Then he took the account off test and confirmed it was back in normal service. The ticket recorded the cat's perches by height, the LED results at each, both walk-test passes, and the new location.

What the first ticket missed

The first tech's reasoning was sound and the record of it was clear, which is why the second visit took an hour instead of three. What it did not do was ask the second question: where does the cat go when the tree is gone? A pet-immunity fault is never fixed by moving one perch, because the animal looks for the next one. The question that closes it is whether any perch within reach of the cat sits in the detector's near field or upper zones, and the answer to that is about the detector's location, not the furniture.

When the room will not allow a clean location

Some rooms have no wall where nothing climbable sits in the near field: open shelving on every wall, a cat that uses the tops of kitchen cabinets, a dog that stands on the sofa back. In those rooms the honest options are a different detection method for that room (glass-break detectors covering the windows, for example) plus contacts on every opening, or an arming habit where the pet is confined to a room without a motion detector when the system is armed away. Both are conversations with the customer, and both are better than a detector everyone has learned to ignore.

References

• See related: Pet-Immune Motion Detectors and Where They Still Fail
• See related: Motion Detector Coverage and the Walk-Across Rule
• See related: How to Put an Alarm Account on Test and Take It Off
• Detector manufacturer installation instructions for mounting height, coverage pattern and pet rating