Advice regarding inrush current please?

[HappyHippyDad]

I have just fitted 5 x 200w LED floodlights on the same switch.
I fitted 3 yesterday all worked fine. However the owner said they tripped once after I had left, no further info as to if they tripped immediately or after a while.

I have just fitted the last 2, the MCB is tripping immediately when the lights are switched on.

It also trips immediately when only 4 lights are connected.

At present they are on a type B 6A MCB.

The RCD does not trip and there is no fault to earth.

I had zero time to investigate and am returning in the morning.

My thoughts are.

1. Short circuit L-N
2. In rush current.

I know LED's can have a high in rush current, but could 3x200W LED floods (power factor 0.9) cause the MCB to trip? Could 4 or 5 cause it to trip with regards in rush current?

 

[timbobelfast]

Might be worth getting the trusty clamp meter out and see what current is being pulled when switched on as an indication.

 

[Marvo]

Could have sworn I replied to this thread the other day....gawd knows where my post went.

This is a data sheet for a Meanwell 200w LED driver which is a popular unit used by quite a number of manufacturers.
I would imagine characteristics are similar for other brands.
They suggest 4 units on a 6A type B or 6 units on a 6A type C

ELG-200-MEAN WELL Switching Power Supply Manufacturer - https://www.meanwell.com/webapp/product/search.aspx?prod=ELG-200

Meanwell are one of the best brands of LED drivers available. I've used a wide range of their product line over the years and never been disappointed. It would be dangerous to assume that you could get insight into the performance or specs of another brand by referring to the datasheet of the Meanwell equivalent, I can almost guarantee the other brand will fall short and often by a large distance.

 

[1Justin]

I've fitted eleven x 28 Watt floods once (~300W total) and had to upgrade to a 16C breaker. (Phew..that was too close for comfort on R1+R2 for my already installed cables..). I'd began with 6B and taken every step up from there till I arrived at 16C. - These were Meanwell drivers on Ledzip floods.

 

[Wilko]

Hi - can you spell out the site Ze, the circuit design, cable sizes lengths and loads? I'll keep me devious thoughts to meself just now .

 

[Wilko]

Hi - and what lights are they please ?

 

[marconi]

You could use thermistor in-rush current limiters as I have done but you have to work out its required cold resistance and power rating and then faff about mounting it inside an enclosure which takes into account it gets hot.

You'd be better off using one of these ready made devices which can solve LED problems like yours:

ESB 101-23 - Peak current limiter, 23 A, 2000 µF - https://www.reichelt.com/gb/en/peak-current-limiter-23-a-2000-f-esb-101-23-p167632.html?CTYPE=0&MWSTFREE=0&PROVID=2788&wt_guka=54950995355_263550067601&PROVID=2788&gclid=EAIaIQobChMI9-a2i8za3QIVr7ftCh24QQA7EAQYBSABEgKcYvD_BwE&&r=1

 

[Andy-1960]

Might be worth getting the trusty clamp meter out and see what current is being pulled when switched on as an indication.

Your trusty clamp meter probably won’t react in time to give you an accurate reading especially if it’s digital, the inrush current drawn only lasts for microseconds

 

[HappyHippyDad]

Update!!

It was definitely in rush current!

I have not experienced it being a problem before as I only do domestic and Type C breakers are rarely needed.

I had taken in to account VD, worked out what the Zs was likely to be and used 6mm SWA for the approx 100m run.

My Zs is 1.45ohms.

I am picking up a Type C 6A and 10A tomorrow and will try those, however I did have a 16A Type B and this is working fine. The max Zs for 16A Type B is 2.2 ohms so I have been saved.

I can run another circuit for the next 5 x 200W lights or I might find that a 20A type B works for all 10 lights as long as the Zs < 1.75ohms.

I'm embarrassed that I didn't factor inrush current in to my design, it's not acceptable to be acting reactively like this to a problem that should never have arisen!

One last question..

What is the relationship between Type B and Type C MCB's with regards inrush current? What I mean is if for example a 6A Type B could take perhaps 100A for 0.2s without tripping what would the equivalent be for a 6A Type C be? Is there some data sheets on this or some interesting reading somewhere?

ps.. these are the lights http://www.belllighting.co.uk/fittings/SkylineProLEDFloodlightsWallpacks

Apparently the inrush current is approx 75A.

 

[Pete999]

Update!!

It was definitely in rush current!

I have not experienced it being a problem before as I only do domestic and Type C breakers are rarely needed.

I had taken in to account VD, worked out what the Zs was likely to be and used 6mm SWA for the approx 100m run.

My Zs is 1.45ohms.

I am picking up a Type C 6A and 10A tomorrow and will try those, however I did have a 16A Type B and this is working fine. The max Zs for 16A Type B is 2.2 ohms so I have been saved.

I can run another circuit for the next 5 x 200W lights or I might find that a 20A type B works for all 10 lights as long as the Zs < 1.75ohms.

I'm embarrassed that I didn't factor inrush current in to my design, it's not acceptable to be acting reactively like this to a problem that should never have arisen!

One last question..

What is the relationship between Type B and Type C MCB's with regards inrush current? What I mean is if for example a 6A Type B could take perhaps 100A for 0.2s without tripping what would the equivalent be for a 6A Type C be? Is there some data sheets on this or some interesting reading somewhere?

ps.. these are the lights http://www.belllighting.co.uk/fittings/SkylineProLEDFloodlightsWallpacks

Apparently the inrush current is approx 75A.

I had that problem once, inrush, you mucky people, some will get the joke some wont, not going to explain.

 

[Richard Burns]

Not a cheap solution but if you had sequenced switching so that only three lights came on at any one time, with a 0.5s delay or so between each bank this would alleviate the inrush current problem. You could fit delay on relays set to different times for each light or I believe you can buy specific staged switching systems.
The thermistors are a problem because of the heat and they ought to be switched out after start up which is more expense and messing about.

The current difference between B and C (and D) is not something I have seen data on but logic indicates that if on heavy overload an MCB trips in 10 ms, then any current over the trip rating but faster than 10 ms will not give it time to trip. Therefore you need to slightly average the current seen on start up so if an LED has 150A for 0.3ms if may seem to the MCB to be only 50A (or less) over 10ms, in this case a B type would trip, a C type may well trip and a D type would not trip.

 

[Marvo]

Glad you got it sorted out. Inrush current for switch mode power supplies / LED drivers is something that's not easy to calculate accurately. If you have a borderline case, which it sounds like yours was, whether it trips can actually boil down to how near the zero point the supply waveform happens to be when the switch closes and whether the lights have been recently run or not.

 

[Marvo]

Just as a footnote, if you're planning to add more lights onto the same supply I'd suggest supplying them via a separate switch or timer or PIR or whatever switching system they're on, even if it works out more costly. That way they won't all switch simultaneously which would add to the inrush current issue.

 

[HappyHippyDad]

Just as a footnote, if you're planning to add more lights onto the same supply I'd suggest supplying them via a separate switch or timer or PIR or whatever switching system they're on, even if it works out more costly. That way they won't all switch simultaneously which would add to the inrush current issue.

Thanks Marvo.
The plan is to have the next 5 on a separate switch.
I appreciate the detailed and informative responses, thanks again.

 

[camerabloke]

I had that problem once, inrush, you mucky people, some will get the joke some wont, not going to explain.

Just saw some tumbleweed roll past

 

[HappyHippyDad]

Well the 5 lights work fine on either a 10A Type B or a 6A Type C, so I should be ok with all 10 on a 20A Type B or possible a 10A Type C. Also the next 5 will be switched separately so that will avoid the inrush current of all 10 at the same time. Happy