GJD020 Passive Infra-red Motion Sensor Review

The GJD GJD020 'Opal XL' PIR sensor is a high-quality, external motion sensor. It can be configured to work as a long-range (35m) sensor, support pet immunity, or as a short-range 'curtain sensor'.

We purchased this one for review for about £45 including delivery from Alert Electrical. This is not cheap but, you need good quality, outdoor PIR sensors, if they are to work well in smart home applications.

Note: This is not a 'smart' PIR/motion sensor. I usually install wired 'dumb' sensors wherever possible and then connect them to my smart home. They don't have the limitations associated 'smart' sensor such as batteries, poor range, intermittent connectivity, high latency or the slow re-triggering speeds associated with their battery saving techniques. Put simply, 'dumb' wired sensors give much better performance, resulting in a much better user experience.

The objective of this review is to test the sensor with a wide range of smart home use cases, to see how well it performs. When connected to my contextual smart home it will provide useful occupancy information and other context such as zone activity levels, to allow my smart home to intelligently control lighting, water features, a robotic lawn mower, irrigation, etc. It will also be used for security applications, including the triggering IP cameras and infra-red security lights.

Specification

This 'Opal XL' PIR has an ABS housing (approximately 10cm in diameter) with a white polythene UV stabilised lens cover, providing an IP55 rating. The internal electronics module can be adjusted both horizontally and vertically and it features a digital ASIC / microprocessor.

Operating voltage: 9 to 15V dc
Operating current: 6mA typical
Detection range: 10m to 35m (configurable)

The sensor has two outputs, both rated at 25mA max:

'A' is an open collector (-ve switching) output that triggers for 400mS every time the sensor is activated and is not affected by light levels.
'S' is an open collector (-ve switching) output with a 60s activation (re-triggerable) and can be limited by an adjustable light-level sensor. This would typically be used to directly drive security lighting.

The 'A' output is the one I'm interested in. It can be used with an Arduino, Raspberry Pi, etc. to detect motion in a timely manner. The ability to send pulses in quick succession means it can be used by my contextual smart home to monitor the activity level and work better for occupancy detection.

Connections

The installation and configuration process is clearly explained in the the GJD020 manual.

To gain access to the sensor module, the lens cover needs to be removed. This has a subtle logo and a tiny drainage hole, which need to be at the bottom.

Rotating the PIR module, reveals the screw terminal connections at the rear. The manual describes how to connect them.

Popping out the Fresnel lens from the front, allows access to the module's potentiometers, to adjust the range and light level sensitivity.

Testing

The manual shows how to connect a 12V sounder/buzzer for testing coverage. I planned to test this PIR sensor in a number of locations and configurations, so I fitted a 5m length of 4-core alarm cable to it:
+ = red, - = black, A = blue, S = yellow

For testing, I used a 12V battery and a 12V buzzer with a 150Ω resistor is series, to limit the current to 18mA. This means I could easily test it in lots of different locations and hear when the PIR is triggering. There is a red LED that flashes for a few seconds on applying power but, it is not on when the PIR is in normal use.

Out of the box, the sensor supports long-range, multibeam coverage. I fitted it above my garage door, to test the range over my drive.

The sensor can be configured to provide curtain coverage by masking some of the lens with tape (supplied).

Orientation

The sensor has to mounted the right way up and pointing in the right direction horizontally. It can be adjusted vertically internally. Although it claims to be adjustable horizontally, this isn't really the case. The 'forked' mount for the sensor module can only rotate at its base and this limits the range of movement.

Mounting

Note: As with all sensors like this, they need to be rigidly mounted and must not move about in the wind, etc. PIR sensors work best when people walk across the 'beams' and not towards them.

The shape of the housing makes mounting a little challenging in some cases because internal module mount is fixed to the top face. It is easy to mount it on the underside of a flat surface and also fairly easy to mount on a vertical surface. It would also fit quite well in a 90° vertical corner. Everything else needs a bit more thought. The cable exits on the 'bottom' face so unless it runs through the mounting surface as well, it will be in the way.

The internal mount is easily removed and is held in place by one screw.

This housing shape lends itself perfectly to a 3D printed module mount that enables the housing to be mounted at 90° into a vertical corner, so I designed a new sensor module mount using Autodesk Fusion 360. I measured the existing bracket and then added the alternative mounting point. This now gives me more mounting options but, this new mount doesn't have quite the same clearance for the sensor module and thus has more limited vertical adjustment.

This is my new 3D printed mount, printed on my Ender 3 Pro 3D printer. It worked really well during my testing. You can download the STL file.

Field Of View

The field of view horizontally is about 90°. Vertically, it depends on the configuration and installation.

Range & Sensitivity

The range and sensitivity are adjustable by using potentiometer in the module and by using the supplied tape to 'configure' the sensor, essentially masking off areas of the sensor. The sensitivity and range can also be adjusted by tilting the sensor accordingly. In this respect, there is quite a lot of set up required to optimise it for each installation. This is good though, as it will improve performance and minimise false triggering.

I started my testing with the range adjusted to its minimum setting but this didn't even cover my patio, so I ended up turning it up more.

One problem with some occupancy sensors is false triggering. A lot of this can be fixed with good positioning but, some sensors are simply susceptible to it. My testing shows this sensor to be very good in this respect. I have yet to experience any false triggers.

Use Case Testing

I have a number of use cases that I will be testing against:

Drive

With a P