Fluorescent Lighting Surge

[MrDuck]

I have installed eight 6ft 75W fluorescent lights in my garage, fed via a 5A fused spur off the garage ring main, which itself is protected by a 32A RCBO in the main fuse board. I have installed a single on-off switch for all the lights.

Surprisingly (for me at any rate) the RCBO trips in about one on five instances when I switch the lighting on. Please can someone explain what is happening and suggest what I should do about it.

Many thanks

 

[ricicle]

If the non-rcd part of the RCBO is type B then the inrush current of the lights could be enough to cause instantaneous operation of the breaker.
This might not happen every time due to the largest inrush only happening at instances when the supply is at a certain place on the AC sine wave (as it is moving away from zero)

 

[MrDuck]

Is there any information on how big the inrush current can be on fluorescents? It must be a whole lot bigger than I had expected.

I could get a type C RCBO but the current type B RCBO is almost new and it's required to protect all loads in the garage so going for type C operation doesn't feel ideal. I guess the alternative is to fit a two way lighting switch but that's a PITA for other reasons!

Thanks for your comments.

 

[JohnW2]

If the non-rcd part of the RCBO is type B then the inrush current of the lights could be enough to cause instantaneous operation of the breaker.

With a B32 and 8 lights, that would imply an 'inrush current' (I presume the current through filaments/starters) of about 20A per tube. If that is the case, then one would expect to commonly see this problem at switch-on of two (maybe only one) such tubes if protected by a B6 - is that something you commonly encounter?

I have certainly not come across it such a phenomenon with one or two tubes on a B6. Indeed, in my cellar, I have seven simultaneously-switched 5ft tubes running off a B6 and, in 20+ years, cannot remember switching them on (which happens at least once per day, more often several times) ever having caused the MCB to operate.

Kind Regards, John

 

[ricicle]

It might be the case that the RCBO (overload part) is a bit on the sensitive side or that the switching arc is causing the earth leakage part to operate.
I suppose I went for a more 'theory' type explanation first although it may be a rare occurence albeit possible if the sine wave is in the right (wrong) place.
Maybe an IR test of the fittings and/or wiring along with a good brand switch may alleviate it.

 

[JohnW2]

It might be the case that the RCBO (overload part) is a bit on the sensitive side or that the switching arc is causing the earth leakage part to operate. I suppose I went for a more 'theory' type explanation first although it may be a rare occurence albeit possible if the sine wave is in the right (wrong) place.

Indeed - your 'theory-type explanation' sounded very credible until I realised that, by extrapolation, if it were the sole explanation we ought to be seeing it happening all the time, particularly in my cellar!

Maybe an IR test of the fittings and/or wiring along with a good brand switch may alleviate it.

I suppose it could just be a fault in one of the fittings or even tubes - e.g. that allows arcing to occur across one of the filaments during start-up. I'm not too sure how one could 'IR test the fittings' without removing the starters - in which case a fault such as I've just mentioned presumably wouldn't show up. It might be necessary to disable the fittings (starter removal should be enough) one at a time to see if the problem can be attributed to just one of them.

Kind Regards, John

 

[ricicle]

Like I said though John, it won't happen all the time - it is sine wave dependant !

 

[JohnW2]

Like I said though John, it won't happen all the time - it is sine wave dependant !

As I think you probably well realise, I didn't mean that it would happen 100% of the times that fluorescent tubes were switched on!!

What I meant was that. with probably hundreds of thousands, if not millions, of pairs of fluorescent tubes on B6s out there, each drawing 40A between them on switch-on, even if it only happened on a small percentage of switch-ons (due to the chance point in the sine wave), that could still probably be many thousands of incidents of MCBs tripping every day - so that we would all be 'hearing about them all the time'!! The OP says it's happening to him on about 20% of switch-ons. The seven in my cellar have probably been switched on at least 15,000 times without a single B6 trip, let alone 3,000 trips (and taht's with a B6, not a B32)!

Kind Regards, John

 

[MrDuck]

Thanks everyone for your comments - they really do help. It's an odd fault isn't it! For now I intend to try and fault find it by removing one starter at a time to see if it's fitting related.

It would be good if I could establish whether the tripping is earth leakage or overcurrent driven, but I'm not sure whether that's possible. Any ideas?

Thanks again everyone.

 

[JohnW2]

Thanks everyone for your comments - they really do help. It's an odd fault isn't it! For now I intend to try and fault find it by removing one starter at a time to see if it's fitting related.

As I said, I think that would probably be a good start (if you'll excuse the pun!).

It would be good if I could establish whether the tripping is earth leakage or overcurrent driven, but I'm not sure whether that's possible. Any ideas?

Only a sophisticated, expensive and rare RCBO woul tell you which 'part' had tripped - 'standard' ones certainly wouldn't. Temporarily changing the B32 RCBO for a B32 MCB would obviously be one approach, but even that would not necessarily exclude 'over-sensitivity' of the over-current part of the present RCBO as being the problem. I suppose a more complete approach would be first to change the present B32 RCBO for an 'identical' one - and if that doesn't alter anything, then temporarily change to a B32 MCB.

There is at least one other possible approach which some might be tempted to suggest, but it's potentially dangerous.

Kind Regards, John

 

[JohnW2]

duplicate deleted

 

[333rocky333]

How old are these 75 watt fittings, old ones tended to have very large metal capacitors across the supply and i seem to recall this increased the start load.
could you alter the wiring to a 2 gang switch

 

[MrDuck]

Several of the fittings do date back 20 - 25 years in case that's a possible issue. I could rewire to a two gang switch but it's quite a fiddle as my garage, like most peoples, is full of 'stuff', and I would prefer to avoid it.

I'm now thinking that an earth leakage fault is most likely and will investigate this further, initially by selective starter removal.

 

[JohnW2]

Several of the fittings do date back 20 - 25 years in case that's a possible issue. I could rewire to a two gang switch but it's quite a fiddle as my garage, like most peoples, is full of 'stuff', and I would prefer to avoid it. I'm now thinking that an earth leakage fault is most likely and will investigate this further, initially by selective starter removal.

Selective starter removal would not necessarily be a very effective way of eliminating fittings if the problem were 'an earth leakage fault'. Similarly, if the issue were arising because of 'large capacitors across the supply' in old fittings (as suggested by 333rocky333) then, again, starter removal might well not eliminate the fault. In either case, you would probably have to selectively disconnect the fittings one at a time.

Kind Regards, John

 

[MrDuck]

A new theory I have is that the metal cased capacitors may have their cases connected to earth, in which case they could contribute to an earth leakage spike at switch on. Fixing that will require insulating the cases of the capacitors. Not impossible, I hope, and worth a try.

Thanks everyone for the continued flow of advice.