Strange things happen with two PIRs in parallel

[SquirrelSparks]

I needed to illuminate our dark side walk at the house. I have used two PIRs wired in parallel which switch four LED lights. The idea is no matter which side of the walk you start at, you'll always set one PIR off and have light.

This works, but randomly the MCB of the circuit triggers. I debugged the issue and found that it triggers even if I take out all lights and just have the two PRIs in parallel. (PIRs are two CLA SENS008.)

So my theory is that there's an issue wiring them in parallel. How can I solve this?

The wiring is 3C + earth with the switched core connected to the two PIRs and all LED lights. Hence adding a wire would be very problematic. I'm hoping this can be solved without touching the cable setup.

Thank you for your ideas.

 

[telectrix]

as it's configured, wired in parallel, triggering either PIR wil activate all the lights.

 

[mainline]

I needed to illuminate our dark side walk at the house. I have used two PIRs wired in parallel which switch four LED lights. The idea is no matter which side of the walk you start at, you'll always set one PIR off and have light.

This works, but randomly the MCB of the circuit triggers. I debugged the issue and found that it triggers even if I take out all lights and just have the two PRIs in parallel. (PIRs are two CLA SENS008.)

So my theory is that there's an issue wiring them in parallel. How can I solve this?

The wiring is 3C + earth with the switched core connected to the two PIRs and all LED lights. Hence adding a wire would be very problematic. I'm hoping this can be solved without touching the cable setup.

Thank you for your ideas.

They are probably not pirs that use a relay (can you hear a click) when switched.

You may need to use a relay /contactor to isolate the pirs from each other.

 

[SquirrelSparks]

They are probably not pirs that use a relay (can you hear a click) when switched.

You may need to use a relay /contactor to isolate the pirs from each other.

There is a relay in the IC (HRW-124LM) that switches main power, however, it's also connected to N via a 680K resistor. Here's a very simplified circuit diagram of my setup:


There's more IC logic between L (the upper wire) and the relay coil, obviously, but I'm thinking that the resistor might play a part? Do you know its function? What if I removed it?

 

[mainline]

There is a relay in the IC (HRW-124LM) that switches main power, however, it's also connected to N via a 680K resistor. Here's a very simplified circuit diagram of my setup:

View attachment 107400
There's more IC logic between L (the upper wire) and the relay coil, obviously, but I'm thinking that the resistor might play a part? Do you know its function? What if I removed it?

It’s defiantly tripping mcb ? Rather than rcd ?
What size mcb ?

A pair of parallel 680k resistors isn’t going to draw any significant current

Does it function when run one at at a time

 

[SquirrelSparks]

Hi @mainline,
I have temporarily disconnected one sensor and waited a few days, and so far it works fine.
Definitely the MCB that trips as there is no RCD connected to that circuit.
I agree with your current assessment. However, I'm still thinking something must go into the second sensor that it doesn't like.
What if I used the active out to switch another relay so that any switched lamp power is definitely not connected to any sensor once both are active?

 

[mainline]

Hi @mainline,
I have temporarily disconnected one sensor and waited a few days, and so far it works fine.
Definitely the MCB that trips as there is no RCD connected to that circuit.
I agree with your current assessment. However, I'm still thinking something must go into the second sensor that it doesn't like.
What if I used the active out to switch another relay so that any switched lamp power is definitely not connected to any sensor once both are active?

That would work, or change the pirs for ones that use a relay but haven't got that type of override function.

 

[Avo Mk8]

Given post #6 saying tripping seems to stop with only one sensor, this may be irrelevant, but I'll post it anyway!
It's likely your floodlights are responsible for quite a significant inrush current, depending on where the mains cycle happens to be when a relay clicks in.
Some interesting figures in the article below, and comment on MCB types:
"For a typical 150W, 0.7A LED driver, Philips Model 9137012116, used in various CREE luminaires, the inrush current is stated as 130Amp for 165 Micro Seconds. It is suggested that up to 7 -10 of these drivers be used / per 20A C curve MCB"
What is your MCB rating and characteristic?

Advanced Lighting Technologies Asia

PS just a thought - if the sensors had solid state zero volt switching instead of a relay, or you could implement that as an addition to what you have, that might solve the problem!

 

[SquirrelSparks]

Hi @Avo Mk8,
interesting find! Although most likely not applicable in my case (as two sensors in parallel tripped the MCB with no lights attached), I opened a faulty LED that I still had and found its LED driver to be PT4515HA, but the linked data sheet does not reveal any Imax (as expected, given how hard it is to measure).

For my problem I'll try to add the additional relay which might just fit into available space below the sensor's circuit board. If that doesn't help, I'll invest in new sensors, as we really need one on each end of the dark path.

Thank you @Avo Mk8 and @mainline for your contributions!

 

[mainline]

A few years ago I used 6 pirs in an indoor beach, driving via a contactor 6 × 150w bay lights never been a problem.

 

[Avo Mk8]

found its LED driver to be PT4515HA, but the linked data sheet does not reveal any Imax (as expected, given how hard it is to measure).

Interesting - must brush up on my Chinese!