How to explain inrush and why a submain blows before a circuit MCB

[SUNRAY]

Todays little just job; customer has complained of frequent fuse blowing since changing to LED tubes in recent fittings from Flou.

The situation is far from domestic or DIY and in a bid to stop it happening customer recently instructed an electrician to change the submain fuse box to all AFDD Schneider, 6x 6A AFDD for lights with each running between 3.5 to 5A (total ~25A), plus 1x 16A AFDD feeding several DSSO, basically not used. It is fed with app 30m of 4mm² 3C SWA and 32A BS88.

The problem is the building is powered down daily and they expect to light it by pressing the green button (DOL starter). They don't understand: Why this wasn't a problem with their fluo light but is with LED, why the circuit fuses never blew or the new AFDDs have never tripped but the supply fuse blows sometimes and certainly don't understand inrush.

I've suggested delay step starting which they don't like the idea of as they have previously had problems with a similar system, getting the staff to switch the lights off before they close up which they describe as impossible.

I suggested upgrading the fuse and cable, However that made 2 people shudder at the thought as the building is in a row of industrial units and the submain passes through a unit to which they have very restricted access as it's supposedly sensitive/secure.

As a last resort I've suggested DOL starters on some of the circuits but their risk assessment says no (I assume due to a problem in the past).

 

[RandomGrinch]

Completely not my area, so forgive me, but would an inrush current limiter work?

 

[bernardgreen]

Delayed step switching is probably the best option and can be achieved quite easily by simple relays.
Pressing green button applies power to a few lamps and the coil of Relay 1
Contacts on Relay 1 close which applies power to a few more lamps and the coil of Relay 2
Contacts on Relay 2 close which applies power to a few more lamps and the coil of Relay 3
Contacts on Relay 3 close which applies power to the rest of the lamps,

Dividing the lamps into 4 groups should be enough.

Switching of the power to will drop out all the relays and so pressing the red button on the DOL starter will turn off all the lamps.

 

[SUNRAY]

Delayed step switching is probably the best option and can be achieved quite easily by simple relays.
Pressing green button applies power to a few lamps and the coil of Relay 1
Contacts on Relay 1 close which applies power to a few more lamps and the coil of Relay 2
Contacts on Relay 2 close which applies power to a few more lamps and the coil of Relay 3
Contacts on Relay 3 close which applies power to the rest of the lamps,

Dividing the lamps into 4 groups should be enough.

Switching of the power to will drop out all the relays and so pressing the red button on the DOL starter will turn off all the lamps.

Oh I know the how's, and the lamps are already divided onto 6 AFDDs so no problem on that score. I've never looked at the inrush decay on LED drivers but having tried simple relays with CRT monitors (including the degauss of course) and filament lamps and found they were not slow enough I assume the same could be true for LED too.

 

[jj4091]

Oh I know the how's, and the lamps are already divided onto 6 AFDDs so no problem on that score. I've never looked at the inrush decay on LED drivers but having tried simple relays with CRT monitors (including the degauss of course) and filament lamps and found they were not slow enough I assume the same could be true for LED too.

I don't know anything about your application but can't you slug the relay coils to delay their actuation.

 

[plugwash]

Ultimately there comes a point where you have to say "sorry, I can't help you if you won't let me help you".

 

[ericmark]

As said a starter normally has a timer, clearly needs reconfiguring, but multi relays with a delay on each one should work. Other option as with motors is a soft start.

I found the fluorescent did have an in rush, they charge the capacitors on the HF ballasts on start up, so can't see what has changed, the wire wound ballast went out years ago, but they also had an in rush, I know in large rooms we would have the lights on 2 or 4 switches, as there was always a delay between turning on each switch.

Maybe smart switches which can auto turn lights on when they arrive and off when they leave, then can't be left on, only over ridden.

 

[SUNRAY]

I don't know anything about your application but can't you slug the relay coils to delay their actuation.

Yes there are things to use for the purpose, with loads as small as these I'd simply use timer relays with settable delays and probably use about one second.

Ultimately there comes a point where you have to say "sorry, I can't help you if you won't let me help you".

Yes this is exactly the sort of point I'm trying to get my head round on this one.

As said a starter normally has a timer, clearly needs reconfiguring, but multi relays with a delay on each one should work. Other option as with motors is a soft start.

I found the fluorescent did have an in rush, they charge the capacitors on the HF ballasts on start up, so can't see what has changed, the wire wound ballast went out years ago, but they also had an in rush, I know in large rooms we would have the lights on 2 or 4 switches, as there was always a delay between turning on each switch.

Maybe smart switches which can auto turn lights on when they arrive and off when they leave, then can't be left on, only over ridden.

A basic starter has no delay, simply a push button to energise the contactor (As is the feed to the building) but of course there are all sorts of variations.

I get the impression the flou fittings were quite old so very well may have been chokes and yes they would have created a heavy inrush current, the 'maintenance guy' I was dealing with has categorically stated it wasn't a problem before they were replaced and seems to be looking for a 'repair solution' rather than an alteration.
There are already 6 circuits and a fairly big grid switch, I'm sure wider than 4, so possibly 18way. Turning that lot on by hand would cure the problem of blowing a fuse.


From the comments from him I'm quite sure that if a bigger fuse fitted in the carrier it would have been implemented a while back.

 

[SUNRAY]

Ultimately there comes a point where you have to say "sorry, I can't help you if you won't let me help you".

Yep, I believe that point has been reached.

 

[JohnD]

I used to work in glossy offices with fancy lighting.

It had occupancy sensors (PIRs) at entrances, lift lobbies and in floor sections. If anybody walked in, the lights came on for the room or entire floor. Timers held them on for quite long periods. The corridors and walkways held on for longest, and the WCs for an hour ir so, and seldom went out.

There might have been timers to turn on safety lighting in stairwells and walkways during normal hours of occupation.

Only if you were working late, alone, would you see sections of the floor go dark.

It would be impossible to have a switch-on surge for the whole building, or even a floor.