Weird Security Light Thingy

[The Medway Handyman]

Changed two security lights for a customer a few weeks ago, both exactly like for like. Fairly remote rural location.

Few days ago they had a power 'cut'. Seems like the power went off for a very short time, neighbours alarm went off briefly, video etc needed re setting.

Security light on the front (500w) stayed on all the time, it had switched to manual override. Light at the back (150w) stayed in auto mode.

Switching off & back on the front light put it back to auto mode.

Both lights are the same make, same PIR, same controls, same instructions.

Can't work out why the 500w light threw a wobbly, but the 150w didn't?

Problem sorted, but I'd like to know.......

[Andy Hall]

Sheerness?

Different switch-on load affecting the poorly designed controller?

[The Medway Handyman]

There is remote rural & wilderness matey.

Not sure Sheerness actually has electricity yet, running water is a recent inovation.

Both B&Q's finest so anything is possible. Whats that mean then?

[Bob Eager]

And then there's Milton Regis.

[Colin Wilson]

It might be a security "feature" - it's not unheard of for thieves to pull the main fuse of a property with an external meter cupboard.

[<me9]

Thought it had had running water for years, just not in pipes.

[Frank]

The front one was noticed so the customer was playing with the on/off switch before you got there? They might have done the same to the one at the back, but left it off longer.

[chrisj.doran]

What is the sequence for setting it to permanently on? Usually it's something like turn it off then back on within n seconds. But n won't be particularly accurate, so maybe the power glitch was just the right length of time to turn one permanently on but not enough to do the same for the other.

Chris

[The Medway Handyman]

Steady Bob. On the map it says 'here be dragons'.

[Richard Russell]

Exactly what I was going to say. Expecting two lights to behave the same way in these circumstances is unreasonable. The off-on delay for the manual override is probably set by a crude analogue CR time constant, likely to vary significantly between samples. It was just bad luck that the power failed for a time which was within the set delay for one light but not for the other. In my opinion the manual override facility is far too prone to leave the light on by mistake, and I would rather not have it at all.

Richard.

reply by email change 'news' to my forename.

[Ian White]

There are several different PIR controller chips around. The older ones have that one-blip feature that will leave them on permanently after a brief dropout. I recently bought one that requires tightly specified double-blip ritual, and even then, will turn itself off at daybreak.