A closer look at a bulkhead light with PIR sensor

In this video I look at a variant of a bulkhead light with passive infrared motion detect sensor that is completely sealed with the LED panel built-in. Apart from the novel way (for me) of connecting and installing, I took a good look at real and apparent power drawn when the light is on and off. It turns out that this light can get expensive if you being charged for apparent power and strangely that happens when it is NOT on.
P.S. If you hear me saying "power monitor", I do mean "power meter". I do have a power monitor as well but in this case I do mean the meter. Sorry for getting the terms mixed up.
The build-your-own power meter video is here: • Building your own powe...

[@pebblebin4587]

Surely the PIR is not active (using power) during the day? It is activated after dark by the light sensor.

[@TheHWcave]

It was on my reasonably bright bench and the PIR was active as I showed by moving my hand over the sensor. I did not measure the full daylight consumption but I think the chances are very slim that this would be any different. You are thinking that the light sensor would turn off power to the combination of PIR sensor+Light, but that would require 2 circuits that can control mains power independently, one controlled by the light sensor and the other by the PIR and I agree that might be better from an energy consumption, but I think manufacturers always go the cheapest way. One circuit controlling mains to the light, controlled by two sensors (light and PIR), wired so that both need to be "on" to turn on the light. This means, both sensors are permanently powered. Very likely the "light" sensor may even be the PIR sensor itself because PIRs contain an infra-red sensitive circuit. If I had a good way of creating "bright" infra-red light, it would be easy to test that by shining it (non-moving) onto the PIR and see if the sensor thinks its daylight.

[@pebblebin4587]

Thankyou for correcting my misunderstanding. I reciently purchased a solar PIR security light and am impressed with a good amount of light supplied for the short duration needed when activated. Obviously power while on standby is at a minimum and hugely more efficient. Seems irresponsible when manufacturers don't do the same for wired units.

[@homebrewham-m0omo977]

A nice and thorough review. Thanks. Schrödinger version may cost the same to run I think (ha ha)

[@randomfatguy1]

Is it possible to remove the clear blub housing? I have the same model and it filled with water somehow!

[@TheHWcave]

Sorry, I can't help there. Mine is installed and in use and I don't want to take it down and break it apart for an investigation. I am afraid some destruction would be needed as the unit sealed, but apparently not very good though. I am lucky as mine is protected from direct rain. One thing is to drill a drainage hole at a location that can safely discard any water that gets in. That is anyway good practice and I would surely do that for any new installation as a preventive measure. In your case, drilling a drainage hole might get rid of the water but it is likely some corrosion has happened or will follow, so how long it might work after that operation is hard to say.

[@anythingidontcareyoutube]

Would the sensor use more or less power if the sensitivity and lux settings are changed?

[@TheHWcave]

I don't think that would make any difference to the sensor power use. It would change the overall power use of course because the light would be more/less frequent on

[@alexwade9921]

Not all smart meters charge in VA. Mine is in kWH.

[@CGDreamsTutorials]

Is it possible to override the PIR so the light comes on at night and stays on all night? Thanks.

[@TheHWcave]

Unfortunately, no. You can override the PIR by turning it off-on 3x within a total time of 3 seconds, and then the light stays on regardless of movement or level of darkness. Once you turn it off and then back on, it is back in PIR mode, which only activates the light when movement is detected AND its dark. So it has the light sensor but they do not use it in override mode.

[@CGDreamsTutorials]

@@TheHWcave Thank you for your time to reply.

[@jenskaa4044]

I have been wondering if the current measurement could be improved by making 10 turns in the current transformer. That should give 10 times more sensitivity on the cost of only going up to a max of 10A. (100A / 10)

[@TheHWcave]

I agree it would improve the current sensitivity but of course you would now have to divide the watts readout by 10 as well. In fact when I designed this meter I considered adding a switch allowing the current to go through either a straight (1x) wire or a 10x loop. But then most of the low-watt stuff I tested so far (LED lights etc) were still well above the limit and the meter work just fine. This video was truly the 1st real-life case where it did not show current. Thanks for reminding me and I will definitely try it. But it will be a while because of other projects. BTW I did already another mod. In the original video you can see that the module's live wire goes around the transformer and then connects to the live wire at mains front plate socket. That was convenient but by doing that I inadvertently added the power consumption (and power factor) of the module to the load. I fixed that before shooting this video and the module's live wire connects now to the IEC inlet socket, i.e. gets its power before the transformer. This may have been the reason that I never saw the zero current earlier, because the module's current was always there.

[@TheHWcave]

Jens, I am preparing a video doing this modification. Do you mind if I use your comment in it?

[@jenskaa4044]

@@TheHWcave You are welcome, looking forward to see your experiments :-)

[@brianadams4125]

Henley how to heeley