Lighting relays overheating. Solution?

[sweet212]

Evening all,

Been to a customers unit recently due to some high level lighting that had stopped working. The cause was a relay that had overheated causing the neutral pole to fail open (they are 2 pole relays). When opening the panel I noticed the relays had signs of thermal damage. Some relays have already been replaced previously.

There are 24 din rail mounted relays in one panel. Each relay switches groups of 3/4 lights with 3 x 6ft fluorescent tubes in each light. All relays are 20Amp rated.

Due to the lighting being on for 14/15 hours a day my idea is to install NC relays so to prevent the coils being energised during daytime hours so effectively using the coils to turn the lights off rather than on. Would this prolong the life of the relays?

Any others thoughts appreciated.

Best regards.

 

[James]

14 hours is so close to 12 hours it’s not worth worrying about.
Get some good quality lighting contactors. Only switch the live unless you really have to double pole switch them.
Run the live through both relay poles in parallel and you will find the contactors last a lot longer.

 

[Lucien Nunes]

Long periods of operation can take their toll on contactors, but more usually the coil and housing than individual contacts. If one pole has burnt out, unless the unit is mechanically jamming or stiff and not closing the contacts fully, it may be a symptom of overcurrent, at least for the particular switching duty and load type. Ultimately that pole has ended up with excessive contact resistance through wear, pitting, carbonisation and loss of spring tension, hence overheated. It's worth trying to identify the root cause of the failures before choosing replacement contactors.

As per James, you can significantly improve the survival rate of contacts by paralleling them, as an alternative to using a larger contactor, although each of the paralleled contacts must still be rated for the full load.

 

[WallaceP]

Pretty much on the theme as the other replies. I would avoid switching the neutral as there is no certainty that it will make first and break last.
I don’t think it is your problem but the rating of the contacts will likely be the resistive load and if you are running an fluorescent fittings it will need to be derated. But you should be fine if each relay is 20amp and running 3 fittings.
You do get contractors specifically for lighting loads and are designed for longer times being energised but that is normally months and years at a time.
I don’t know how frequent they are causing problems but depending on the nature of the site it may be worth offering a scheduled replacement service at a frequency to be agreed.

 

[sweet212]

Afternoon guys,

Thanks for the input.

I will have a look into better quality lighting contactors, any recommendations?

I will also remove the neutral and run the live through both relay poles. I have space in the panels to fit a din rail connector block for each neutral connection so that would work well.

Thanks, James.

 

[Marvo]

Is 20A the AC1 rating? For fluorescent lighting I'd suggest limit the load on each contactor according to the AC3 rating.

 

[marconi]

Are the fluorescent lights choke - glow starter types?

Your idea to use the NC contacts is not good practice because it employs ‘negative logic’. Better to use use simple positive logic ie something has to be on (control variable) in order that something else is on (dependent variable) rather than the converse.

 

[fairlight]

In my experience thermal damage to din-rail mounted contactors can often be attributed to the devices being butted together which results in coils (not the contacts) overheating. Spacing the contactors apart is sometimes all that's needed.

 

[jamie-spark]

Snap.
That’s what I usually do and use these

Merlin Gerin 27062 Contactor Spacer - John Cribb & Sons Ltd, UK Electrical Distributors, Dorset, Hampshire, Wiltshire, Somerset & the Isle of Wight.

Merlin Gerin 27062 Contactor Spacer - available online and in store from John Cribb & Sons Ltd

www.johncribb.co.uk

In fact if you read manufacturers instructions ,some often advise to install spacers or select low loaded mcbs next to fully loaded ones

 

[WallaceP]

Snap.
That’s what I usually do and use these

Merlin Gerin 27062 Contactor Spacer - John Cribb & Sons Ltd, UK Electrical Distributors, Dorset, Hampshire, Wiltshire, Somerset & the Isle of Wight.

Merlin Gerin 27062 Contactor Spacer - available online and in store from John Cribb & Sons Ltd

www.johncribb.co.uk

In fact if you read manufacturers instructions ,some often advise to install spacers or select low loaded mcbs next to fully loaded ones

£5each? ? I think I’ll stick with connector end stops between loaded contactors.

 

[marconi]

Could you give me details of the relays and I will look up their specification for you?

There are three events which cause damage to the contacts apart from continuous or short -circuit overcurrent through them. The first is the surge current when the contacts close. The second is contact bounce - a brief period when the contacts make/break/make. The third is when the open and inductance attempts to maintain the current and sparking/arcing occurs between the contacts. Marvo mentions using relays with AC3 rating which applies to inductive loads, specifically squirrel cage induction motors being turned on and off during steady state running. For a predominantly inductive load like a conventional choke ballast fluorescent light fitting they may be we satisfactory in many cases.

Utilization categories - Wikipedia - https://en.wikipedia.org/wiki/Utilization_categories

The utilization category for discharge lighting is AC5a - see table at the end of the link above. So a relay with a suitable AC5a rating may be better suited to your application.

Assuming you have inductive choke fluorescents and without knowing more it is hard to know which event is causing most damage to the contacts. The surge current - depending on when in the sine wave the contacts close - can be estimated and the contacts suitably rated to cope with it repeatedly (I'd use 10 x steady state current which in your case for 4 x 3 x 75W 6ft tubes /230 x 1/0.9 is 4.3A so 40 odd Amps. For the opening event, it is the voltage rating of the contacts that matters because they are not breaking 230Vac but 230V plus the back emf kick of the choke - the quicker the opening of the switch so current is made to change from 4.3A to zero over a shorter space of time, -dI/dt, the much higher the back emf kick = -dI/dt x L the inductance of the choke. A 'soft' or slower switch off would actually help reduce the back emf and stress on the contacts. So a relay with contacts having a higher voltage rating will be better suited. As a broad generalisation, if the contacts have been rated for the surge current but not the 230V plus the back emf kick as well then they will have a hard time repeatedly turning off banks of fluorescent light fittings.

If the fluorescent fittings are electronic types, I'd expect - need to confirm or be told - that they are predominantly capacitive in nature thus it is the starting event which causes most damage not the switch off - so surge current rating matters most.

There are snubbers and voltage limiters which could be connected across the load or contacts - but for now I think better to use more robust relay contacts (current and voltage ratings) and keep things simple..

Last for now, the relay contacts will need to have some tungsten in their make up to harden up the contact surfaces to better withstand the heat generated by arcing.

 

[fairlight]

£5each? ? I think I’ll stick with connector end stops between loaded contactors.

Personally I dont think it's anything to do with the loading, if the contactors are butted together it's the heat from the coils that does for them.