LED volt drop- maximum LEDS on 6amp??

[snipersmith87]

First post all so go easy lads....
What's people's views on maximum amount of LEDS on a 6amp mcb lighting circuits?? 6w LED GU10s throughout. Wired in 1mm twin and earth. Maximum cable runs??? Nearly every spark I speak to has a different verdict. Should we still be considering if 50w halogens are to replace the LEDS on circuit design?? Practically everyone uses LEDS now.
As for cable runs/volt drop it would take roughly 38 6w lamps to pull an amp.
1 x 44 x 140=6160
6160/1000= 6.1 volts

A cable run of 140m with 38 LEDS on??

Do people consider this good practice???

Am I missing something???

Obviously if your r1/r2 ZS readings satisfactory.

 

[kingeri]

I see what you're saying but using gu10s always gives folk the option to fit halogens. Using fittings with inbuilt led lamps might be a better option.

 

[Octopus]

Interesting subject - I wonder how long it will take the OSG to catch up with this, with respect to diversity...

More to a point, I wonder how many sparks consider start up in rush currents?

 

[snipersmith87]

Interesting subject - I wonder how long it will take the OSG to catch up with this, with respect to diversity...

More to a point, I wonder how many sparks consider start up in rush currents?

My thoughts exactly...the chances of having all lights on at once also pretty slim.

 

[The Ghost]

Yes indeed it is an interesting subject. I asked one of the reps from LED firm what do they think, not a lot really. One of the things about LED is the surprisingly high inrush current per LED. Fitting 8 LED panels in a conference room and thinking about JCC saying inrush per panel was 10 a. so eighty amps going through a 10a MCB. How is that going to work? Although operating A. =1.5 a. Apparently there is a nanosecond as each light is lit up and the inrush is staggered sufficiently to avoid an instantaneous inrush current. And again we are talking microseconds inrush. It did work and still does fine by the way. So if you investigate literature and manufacturers instructions it does say "you may need to fit a type C MCB" if there are problems with this.

 

[kingeri]

Had to change a b6 to a c6 a few weeks ago because some bright spark had wired 40 odd gu10s through a single light switch. Still got to go back at some point to rewire the lot and split them up a bit.

 

[Midwest]

I concur with Vorts (if I may be so familiar), think the limiting factor of LED's is the inrush current. Check out the manufactures web sites on led D/L to type B or C mcb's.

Edit and King'ies (too familiar)

 

[The Ghost]

You may be familiar @Midwest, there are of course inrush limiters that can be fitted between control gear (switch, which usually suffers the most) and the LEDs. Although it seems this is less of a problem for DALI systems.

 

[Midwest]

You may be familiar @Midwest, there are of course inrush limiters that can be fitted between control gear (switch, which usually suffers the most) and the LEDs. Although it seems this is less of a problem for DALI systems.

Not something I need to consider in Mrs Miggins kitchens I do, but worth knowing.

 

[darksideby182]

Maybe we will start to see a soft start on LED lamps would mean more cost per lamp as this would be manufactured into the lamps PCB

 

[The Ghost]

For a typical 150W, 0.7A LED driver, Philips Model 9137012116, used in various CREE luminaires, the inrush current is stated as 130Amp for 165 Micro Seconds. It is suggested that up to 7 -10 of these drivers be used / per 20A C curve MCB, however it will depend on the individual circuit impedance.

For a typical 220W, 1.05A LED driver, CREE Model LE098X01 used in CREE High Output Edge luminaires, the inrush current is stated as 80 Amp for 1000uS (1mS). It is suggested that up to 6 – 8 of these drivers be used /per 20A C curve MCB, however, it will depend on the individual circuit impedance.
Something to think about especially where relays may be involved.
https://www.osram.com/osram_com/new...and-emergency-lighting-certificates/index.jsp nice bit of bed time reading for you.

 

[snipersmith87]

I'm not talking them all coming on at the same time. So you wouldn't have the inrush current. I'm talking seperate rooms. Seperate switches. 36 x 6w LED d/l in various rooms all on 1mm cable for a 140m run on 1 circuit. Can anybody technically say it's wrong???

 

[1Justin]

As others have said, I think the issue is potentially inrush current. I have had problems with commercial-type LED's, and was forced to fit a 16C breaker on an install needing just 0.75 A, with eleven 28 watt lamps in circuit. Inrush current for twelve Meanwell drivers turned out to be over 600A for 500uS. I was very lucky to just squeak into Zs limits since I'd designed the circuit for 6A B type.

Volt drop, at least as far as lamp performance goes, is a silly reg these days given the fact that LED's all contain SMPS power supplies which have a wide input voltage range and performance and efficiency will be quite decoupled from supply voltage until it gets stupidly out of hand. I regard the 3% figure as quite "bendable" personally, though that rarely comes up given practical current consumption. The guideline to assume 100W bulb in every fitting is equally silly, - especially since 100W domestic bulbs are illegal in EU!

New 18th edition will I think contain stuff about installation efficiency which will be impacted by volt drop since that's all just resistive heating we could do without.

 

[The Ghost]

Can anybody technically say it's wrong???

According to BS7671, yeh. If you calculate the way they say you are supposed to then yes it will not meet regs, as above the 100 watts assumption per fitting (althought I agree with @1Justin). I think as per the second post stated @kingeri, use integrated fittings, as someone could come along (new owner/tenant)and think "I hate LED I love tungsten I am going to change them all back to tungsten", how would your calcs. then stand? One has to provide for future possibilities no?

 

[1Justin]

Your main and biggest concern is safety. If it's safe for both possibilities, then should some dork from the stone age come along and fit halogens and incandescents everywhere, provided the circuit handles the load (which might be worth five minutes of your time), then volt drop in a fictitious scenario isn't major concern as far as I say. If the horrid old bulbs end up running dim, well tough..

 

[snipersmith87]

I may fire this one to the niec. Be interesting to see their opinion. I think the 18th edition will definitely have something regarding this. I very rarely design a lighting circuit to halogen fittings now. I know you should go worse case scenario but I can't help but think it's such overkill. I don't know anyone that pops to the shop to buy a halogen bulb.

 

[Nightmare1308]

Had this 18months ago with robus LED strip, Could only have 3 60w transformers fitted per B type 6A on one switch pattern and 4-5 on a type C.
When consulting Robus who told me there was no issue with this they said it has 70A inrush for nano secounds.
Also speaking of your issure speaking to the NICEIC the told me a sufficient risk assessment and covering letter stating only designed for LED and only LED to be fitted was acceptable in there eyes.
Depending on the installation method 140m of 1mm may not be suitable for anything more than 1A?

 

[dmxtothemax]

A cable run of 140m with 38 LEDS on??

Do people consider this good practice???

Am I missing something???


Do-able, and will work only due to the smps in the lamps making allowances for the voltage drop.

Probably not. But sometimes you don,t have the luxery of perfect instalations.

No, you got it !

your concerns show that.

Tell them your concerns.