Contactors for controlling outdoor lighting

[jamesBen80]

Good evening all I've got an interesting problem. I've been asked by a customer to wire up a series of outdoor flood lights and PIRs around a large property (over an acre in all).
There are currently 4 large floodlights 2 with PIR's. They have asked for 3 additional ones and they have asked for each floodlight to have its own PIR which will, in normal operation trigger just that light, but when they are out/away can they turn a switch (possibly to be controlled by a spare output from the alarm panel in the future) that will link all the PIRs so any 1 PIR will trigger all the lights.

I think I've worked out how to do this but i would need a 6 pole single throw switch, which I've never heard of. Can this be done with contactors?

I've never done anything with lighting contacotrs and wondered if anyone could offer some suggestions? I've attached a diagram of my thoughts so far. It only shows 3 lights and PIR's for simplicity. All the lights and PIR's will be wired back to the switch panel in 4 core, mostly SWA but a couple will be 1.5mm T&E ran internally in existing containment. I might also have an issue with voltage drop as the furthest light will be over 400m from the DB but this is doable in 16mm SWA. I just have to find a way to break the news that the cable alone for that light will be £2500 !!

 

[jamesBen80]

i forgot to say the flood lights are going to be 50/100/150W LED's https://www.wholesaleledlights.co.uk/150w-led-floodlight-15000-lumens.html or something similar

 

[JohnW2]

I think I've worked out how to do this but i would need a 6 pole single throw switch, which I've never heard of. Can this be done with contactors? ... I've never done anything with lighting contacotrs and wondered if anyone could offer some suggestions? I've attached a diagram of my thoughts so far ....

I can't think of any other conceptual approach which would achieve the functionality you want.

As you imply, I imagine that it could be difficult (or impossible) to find a 6-pole mains voltage switch (but someone may come up with a source!). If I were doing it for myself, I might 'link' (with a bit of plastic and some superglue!) three DP grid switches - but one can hardly do that for a customer!

However, as you say, you could simply replace that switch with one or two relays or contactors (you may, again, have difficult in finding a single one which had 6 sets of contacts) - and then have a single SP switch which switched L to the coil(s) of the relay(s)/contactor(s). I see no problem with that, so wonder if you have thought of some potential issues?

Kind Regards, John

 

[SimonH2]

At that power rating, you can probably just link the PIRs together like that. For the multi-pole switch, just use a relay* (or more than one relay) operated by the switch/alarm panel output/whatever. As it is, use a 240V AC coil, if they use an alarm output, you could replace the relay with a 12V DC coil model from the same range.
* A contactor is really just a specific type of relay.

 

[Robin0577]

If you don't mind a clunky industrial rotary switch, six pole is no problem... https://uk.rs-online.com/web/p/rotary-switches/8662205/
You can get up to 16 pole in that style.

 

[bernardgreen]

Or use Change over switches on the Swithched Live lines



Centre Off switches are available if that function ( PIR only controls the lamp ) is required

the furthest light will be over 400m from the DB

That may require a heavy duty switch due to the in rush current into the capacitance of the cable if ( when) the switch closes at the peak of the AC cycle ( 250 volts )

 

[jamesBen80]

Thanks for the reply's, i think the light that is 400m away will be dropped as the cable costs are far too high. The 6 pole switch from RS is perfect and i might even go for more poles if they do it so there is spare capacity for the future. I was going to mount everything into an enclosure and put an MK grid switch along side that controls the individual lights so the rotary switch can just me mounted in the enclosure . i will post some pictures when its all done if you interested.

 

[JohnW2]

I might also have an issue with voltage drop as the furthest light will be over 400m from the DB but this is doable in 16mm SWA. I just have to find a way to break the news that the cable alone for that light will be £2500 !!

I think you could probably save a lot of money!

The floodlights you are considering require an input voltage of 85 - 240V, so voltage drop is not really an issue.

If you had 6 x 150W floodlights, even if the totall current for all of them flowed along the entire 400m of 16mm² SWA, that would only be a VD of about 4.5V (1.9%). With a similar arrangement using 2.5mm² SWA, the VD would be a bit under 30V, which is hardly an issue given the acceptable input voltage range of the lights.

BS7671 merely requires that the VD should not be so high as to "impair the safe functioning" of the load, which would clearly not be an issue.

Kind Regards, John

 

[SimonH2]

Hmm, another thought that should save a bob or three on cable. Not only that, but it also reduces the overall core lengths the power flows through - the initial proposal could have power running out to the furthest light, back along it's SL core, and then out to the other lights ...

How about running two cables daisy chained along the lights. One provides power to all of them, the other provides control.
Ignoring earths, the power cable has 3 cores - L, N, and SL. Each light has it's own PIR, and a relay that connects it's PIR switched line to the common SL. The control cable (which can be small cores as it only powers a few relay coils) provides power to the relay coils.
So the very simple control is you power two cores in the control cable and the relays at each light pull in and any PIR will power all the lights. If you take the common SL core back to the control panel then you can have a switch to power the common SL and turn on all the lights.
After that, you can add as many relays as you like - eg to turn on (or off) any light individually and so on.

Alternatively, there are many multiplexed/remote control options such as DMX and DALI which would reduce the setup to one power cable, one data cable, a controller and user inputs, and a module at each lamp. You then have the option to create any control scheme you want - limited only by your imagination and the capabilities of the controller. Not only that, but there's no risk of wanting to add some control scheme and realising that you don't have any spares cores left.
Unfortunately there are a lot of "non-standard" (read: closed and proprietary) control systems around - standards are wonderful, each manufacturer has some of their own So once you depart from one of the open standards, you are into deciding which standard might stick around and what might disappear overnight - especially things that depend on some external server to work, ask former Revolve users what they think about that You might think the vendor can be trusted - but in the Revolv case, Google bought the company and shut it down to kill one of it's competitors.

 

[JohnW2]

Hmm, another thought that should save a bob or three on cable. Not only that, but it also reduces the overall core lengths the power flows through - the initial proposal could have power running out to the furthest light, back along it's SL core, and then out to the other lights ... How about running two cables daisy chained along the lights. One provides power to all of them, the other provides control. ...

That's certainly an approach, albeit significantly more complicated. However, given that I'm not sure that one could find SWA smaller than 1.5mm², and given that 1.5mm² would be adequate both in terms of CCC and VD (given the voltage requirement of the lights), the simple ('brute force') approach would seemingly be OK with 1.5mm², even if some of the current paths were long. In practice, despite the sums, I would personally probably be hesitant to go below 2.5mm² (even though I would probably struggle to find a rational reason for that!).

Kind Regards, John

 

[ban-all-sheds]

but when they are out/away can they turn a switch (possibly to be controlled by a spare output from the alarm panel in the future) that will link all the PIRs so any 1 PIR will trigger all the lights.

Why, in this modern era, don't they want to be able to monitor and control their lights using their mobile phone from anywhere in the world?

What's wrong with them? Bl***y luddites....

 

[sparkymarka]

Just fit a gjd lighting system ....google gjd digizone , 4 zone switching per controller manual override separate timers per zone , radio linked pirs if required ect....ect ....

 

[jamesBen80]

I've originally suggested some kind of lighting control system like DALI to the customer but they quickly highlighted they have very poor WiFi and mobile data at the premises possibly due to thick stone/masonry walls blocking signals, and they had had given up on mobiles and WiFi long ago and would prefer a robust physical cable installation. I didn't believe them at first but I've been working doing some other jobs for them on site for a few days and its terrible no mobile signal, and the DAB radio won't even pick up a station outside most of the time! Thanks for the reply's above I've costing a few up and am just waiting for them to give me the OK to start and i'll post some pics of the final job for your perusal.

 

[SimonH2]

That does not sound like a valid reason - the state of mobile and DAB reception is irrelevant, and poor wifi could simply be that they haven't put the right equipment in the right places (which is more or less the norm )
Firstly, if you did use a wireless system, then you'd just need to make sure that you sited components in such a manner that the signal wasn't blocked by walls etc. But then, I'd recommend against using radio if you can avoid it - in the grand scheme of things, a wired network connection (if it were a network based system) would be relatively inexpensive and is almost certain to be more reliable.
If knocking up a system with relays etc, then remember that you can buy signalling and control cable that is MUCH smaller than 1.5mm². The only "thin" stuff I've worked with is network cable - 8 cores (2 twisted pairs) of about 0.5mm dia (0.2mm² CSA) - for network applications. But it could also be used for low power control applications (it's specced as ≤ 0.188 Ω/m).

 

[ban-all-sheds]

1) You don't need to use wireless, but you may need a fair bit of control cabling.

2)

I can't think of any other conceptual approach which would achieve the functionality you want.

Oh, I can, and it is an example of where "tech" could actually be of genuine use - some sort of microcontroller (Arduino, MSP430, Teensy...) - this is basically a simple IFTT application.

3) And true, there could be scope built in for future Ludd-corpse-spinning features such as fiddling with the settings by using a phone whilst p*ssed in a Palolem beach bar....