Great test Brian, happy to see IPVM doing this. This motion should never be mounted on a wall unless it's looking down a narrow hallway. Corner mount is ideal. You mentioned a delay at further distances, this is intentional to help reduce false alarms, from my understanding. Motion detection should be redundant from perimeter alarms including door contacts and glass break detectors. We often install floor sensors as well.
Absolutely important, however we recently learned a lesson in bugs and this exact model.
Had one corner mounted in a restaurant kitchen and some cockroaches decided to make it their home. they entered through the wire entry point and started to short the alarm terminals and cause false alarms.
I had no idea they were in there at first. When I opened the unit literally dozens fell out of it in sizes ranging from a flax seed to half inch. I nearly fell off the ladder.
Needless to say I found the problem, which made me feel pretty good, because I know these units are essentially false alarm proof.
I did a lot of work in the projects. It got so where I would yank open a sensor or intercom or siren box or whatever and then automatically step back to give all our little insect friends time to find new homes.
If there's lots of little black spots on the wall around the sensor, like coffee grounds, you're about to have a real bad day.
You have missed quite a number of possible issues / tests.
1/ Humidity, from personal experience these sensors are crap in a high humidity environment, try 90% RH @ about 30 deg C, leave sensor for 1 hr then try a walk test, you will be horrified. The high humidity test should be performed for all sensors tested.
2/ RF immunity, for all sensors try mounting near home wifi router, handheld radios etc & see what happens. Also see what happens when cabling run in proximity of aircon compressors etc.
3/ Temperature - test detection range & consistency at differing temperatures. Please ensure that you allow the sensor to settle for at least 30 min between triggering to simulate real world conditions for intrusion detection.
4/ Mounting height, mount 3 sensors at max, min & recommended mounting height & see how much performance is changed at these heights.
As part of my job I have to evaluate numerous sensors annually & I for one will never 'pass' a sensor until I have performed the above tests. This means that NO new sensor ever makes it to our inventory until it has been tested through our hot & humid summer.
For false alarm testing it is so much environmental. I remember a sensor years ago that was great for detection etc in summer catch rate tests & behaved very well so was installed in a number of test sites. Worked well over winter & then at the start of the next summer about 1 in 10 started false alarming. Worked out that the auto temp compensation was going hyper in a certain temperature range & caused the false alarms.
Thanks for the thoughtful recommendations here. Much appreciated.
I have a few questions to make sure I understand these:
1. Re: Humidity - Is it a matter of variable humidity at the sensor, or through the detection field? Changing humidity levels at the sensor during test is easy, but I don't want to ignore the potential impact in the entire room/ detection area.
2. Re: RF Immunity - Is this primarily a concern of wireless sensors? If EM interference affects cables, it is hard to categorize the issue as a sensor problem. Does that sound right?
3: Re: Temperature - Good suggestion! Will be incorporating temp delta impact into tests.
4: Re: Mount Height - Another good criteria that will be added.
Humidity testing would have to be for the whole area, not just localised at the sensor. From personal experience many PIRs (& dual technology sensors that use a PIR component) are more affected by a humid environment than a hot one. Whilst the 2 go together much of the time (think summer in tropical / sub tropical areas) many sensors compensate well for a 'dry' heat but not humidity.
It is not easy to create artificial humidity unfortunately, this is why I always test through summer.
RF Immunity - this is a cause of many false alarms in the real world. Typically because the RF can raise the noise threshold in the amplifier circuit in the sensor. Also the induced spikes in cables run near the likes of compressors can cause issues with some sensors stability.
Honest question, since these are indoor sensors, where are you using them where it is 90%RH/30C indoors for any period of time? That environment seems like it would cause all kinds of problems in general, and I would think people would have some form of A/C or at least dehumidification running.
You must remember the intruders normally break in whilst the premises are unattended. Normally aircon etc is not left running in an unoccupied building. I am in the southern hemisphere & at the moment inside my home would probably be 28 C & at least 80 % humidity as we are both at work & it is summer here.
The mistake many installers make is to walk test the sensors only on installation when aircon is running & the rooms are cool. A proper test is to walk test at 3pm on a Sunday afternoon when the premises have been unoccupied for the day. This really shows the true catch rate of your motion sensors in a real world environment.
That is interesting, in the US heat/air is usually left running when a building is not occupied, but at lower/higher temps then normal. If heat is usually set to 70F in the winter, it might get kicked back to 60F or 55F, and for aircon, if it is typically at 74F, it might go up to 80F. For the heat/humidity example, I would not expect a building in the US to get to 90%RH/30C (from your example above) ever, even if unoccupied for a long time, the HVAC systems would run enough to keep humidity lower.
In either case, I do agree that the ideal walk-test scenario is in the environment that you expect it to be when you actually need the unit to work.
Another place you will find high humidity indoors is in building that use evaporative coolers or boilers for heating and cooling. Especially the mechanical rooms for these systems often have very high humidity. Just a thought.
Just this evening, working on a home with a Bosch Blue Line, installed about a year ago, I noticed it did not detect me at 25' but did at 15'. This home is in Northern California, it was about 62F, and I was wearing a heaving jacket. This poor catch, really no catch, performance at a longer distances is worrisome, and I'm considering changing it out. I've used them inside converted shipping containers and they work great in smaller areas.
No, this sensor is located in a dining room that opens into the kitchen and family room. It's a large home and the temps are kept low during the week as it is a vacation home for my customers. This Bosch replaced a different PIR, by another manufacturer, due to false alarms. I haven't had any false alarms since replacing with the Bosch. But, now I'm wondering if the reason there are no false alarms is because of the reduced detection range. The model number escapes me at the moment, but I think it claims a 40 x 40 pattern and pet immunity. I'll be going back to the job tomorrow and if I have time I'll check it.
Yesterday, I replaced a C&K DT435 dualtec with one of these. The problem I experienced had nothing to do with the performance of the detector. Rather, it was the installation of the detector onto the bracket. This was a residential install, corner mount. I thought this would be a quick job, no more than 30 minutes. I removed power, removed the old detector, installed the bracket of the new model, and then tried to get the detector part to slide in and lock onto the bracket.
I've used these detectors before, especially their commercial models, and sometimes struggle with the alignment and locking when installing this part. I'm careful not to over torque the fasteners to avoid distorting the bracket, and use countersunk screws to avoid obstructing the two pieces, but I've had difficulties with their twist lock. I took the detector and bracket off the wall and tried fitting them together, which did after some more work.
I tried to reinstall this detector and encountered the same problem. I removed it a second time, went to the truck grabbed another one and, after more struggling, got this one to slide in and lock. This job took a lot more time than I'd like to admit. I've been installing motion detectors over 30 years, so I don't think I did anything improper. I think this detector's mounting bracket is not very forgiving and tolerances between the bracket and the detector are too close. If everything doesn't line up precisely your not going to fit these two pieces together.
I'd be curious to know if anyone else experiences problems like this with the Bosch twist lock detectors.