Replacing 70W T8 Fluorescent tubes with LED equivqlent

[andytheflyer]

Knowing that there are some significant electrical skills available in this community, I thought I'd ask for advice.

My garage/workshop has a twin T8 70W luminaire with fluorescent tubes.  It gobbles about 120W (so 0.5A) according to my energy monitor.  This luminaire is on for a good few hours a day, and since we have Economy 7 (we have a couple of night storage heaters and also set the washing machine and dishwasher to run overnight), I pay around 60p a unit for that daytime electricity (IIRC).

I think I'd like to change the tubes for LED versions. I know that there's an issue with the ballasts, and some LED tubes need a different starter, or no ballast at all and that some re-wiring of the luminaire may be required  I can cope with that.

However, will the LEDs be significantly less power-hungry, and will I get the same light output?  What LEDs should I buy - I see that there are different outputs available.

Looking up the current tubes both are A rated, so are presumably quite efficient to start with.  But the LED tubes claim to use only a third of the power that the fluorescents do.

Almost all of the other light fittings in the house are now LED and our base power consumption in the daytime is around 200W, without the garage luminaire.  It goes up to around 320-350W with the garage lights on (and my wife is very efficient at leaving the bloody thing on).

Should I change the garage twin luminaire to LEDS?

[Brucey]

decent 70W T8 tubes each have a nominal light output of around 6300 lumens. So the light fitting you have at present ought to produce a total output of 12600 lumens.  IIRC the actual light output reduces over time, so new tubes are likely to be better than old ones.

In the laboratory LEDs can produce well over 200 lumens per watt.

However most commercial LED lamps for domestic lighting produce around 100 lumens per watt.

Read the specifications for any replacement lamp very carefully; on the face of it there isn't much electricity to be saved if you need the same amount of light.  However there is much more to it than lumens alone; different lamps direct the light better or worse for particular situations. In just replacing the old tubes with new 'tubes' you are keeping the compromises that are inherent in the original light fitting.

It is probably the case that a different approach to the lighting may work equally well for your purposes but use less electricity.

cheers

[grams]

AIUI the LED tubes are only interested in mains voltage so you have to rewire the fitting to bypass everything else and feed the mains directly to one end of the tube.

But if it were me I’d rip out the whole fitting - they sell striplight-shaped self-contained LED lights that will be less hassle  to install and look better.

Have you thought about installing a PIR sensor to turn the light off automatically?

[rafletcher]

I've just bough, but not yet fitted, retrofit T8 LED "tubes" for my shed. AFAIK there are no wiring mods to make (for these particular tubes), although I may need to remove the starter - I've asked the supplier and they should hopefully reply today.

ETA: Just spotted an illustration which is among one of the pictures of the tubes - and yes, the starter needs to be removed, but no wiring mods need to be made 

The tubes are 1200mm 18W, with 1900 lumen, 6500K. I'll make a subjective decision on light quality/quantity once fitted - which will probably be tomorrow afternoon.

[andytheflyer]

I've just bough, but not yet fitted, retrofit T8 LED "tubes" for my shed. AFAIK there are no wiring mods to make (for these particular tubes), although I may need to remove the starter - I've asked the supplier and they should hopefully reply today. The tubes are 1200mm 18W, with 1900 lumen, 6500K. I'll make a subjective decision on light quality/quantity once fitted - which will probably be tomorrow afternoon.

I'll be very interested in how the LED light compares.  I don't need a darker workshop, but ceiling space is limited to add more luminaires as my model aeroplanes are in racks hanging from the ceiling, and there's not much space left.

Brucey has given me some theory to think about, and grams has the bright (sorry..) idea of a PIR - now why didn't I think of that?

All thoughts and suggestions gratefully received.  New LED luminaires are a possibility, but a couple of quid on new wiring and an hour of my time looks like good value if I can retain the existing fitting.

[rafletcher]

I'll let you know how I get on - I got mine from LED ME.  Assuming yours are 1800mm tubes this would be the replacement:

https://led.me.uk/t8-led-replacement-tube-6-1800mm-in-cool-white-6000k.

The rubric does state "Magnetic ballast compatible (fluorescent fittings fitted with a starter)" so it would seem to be a simple swap, which is why I chose them.

Re lumens, remember that a standard tube emits light  equally through 360 degrees, whereas an LED "tube" is restricted, in this case to 270 degrees, so more "light per degree" IYSWIM.

[Gattopardo]

I've just bough, but not yet fitted, retrofit T8 LED "tubes" for my shed. AFAIK there are no wiring mods to make (for these particular tubes), although I may need to remove the starter - I've asked the supplier and they should hopefully reply today. The tubes are 1200mm 18W, with 1900 lumen, 6500K. I'll make a subjective decision on light quality/quantity once fitted - which will probably be tomorrow afternoon.

I did the same with some osram tubes from homebase.

Feels like more light but that could be because the lamp it replaced was old and quite yellowed.

[jsabine]

All thoughts and suggestions gratefully received.  New LED luminaires are a possibility, but a couple of quid on new wiring and an hour of my time looks like good value if I can retain the existing fitting.

I suspect that time, money and effort will be pretty much equal between the two options. I had a quick look at Screwfix and, assuming you've probably got 4 foot tubes, LED replacements looked to be £12 or £13 each, while a whole new fitting, including lamps, was £30. (I didn't look hard, so cheaper ones may well be a possibility.)

[andytheflyer]

decent 70W T8 tubes each have a nominal light output of around 6300 lumens. So the light fitting you have at present ought to produce a total output of 12600 lumens.  IIRC the actual light output reduces over time, so new tubes are likely to be better than old ones.

In the laboratory LEDs can produce well over 200 lumens per watt.

However most commercial LED lamps for domestic lighting produce around 100 lumens per watt.

I had a chat with the guys at my local electrical factors yesterday, and they said exactly the same.  They reckon my existing fitting is p0roducing about 13,000 lumen, so like for like I need about 130W of LED capacity, which they say will need almost the same power input - so not much to be saved by changing the tubes to LEDS.

Plus, to get the same light output, I'd need 2 twin 5ft luminaires - so a) not cheap, and b) not sure where I can put a second one without totally revising my racking for my model aeroplanes.

But, there could be a big benefit to fitting a PIR so that the light does not get left on, and they have them for around £20.

Goes off for a think...………..

[Brucey]

in fairness as others have pointed out the existing fitting may not use the lumens very efficiently; only a small fraction of the light goes downwards directly, the rest gets reflected (somewhat imperfectly) from the back of the thing, or is cast sideways (depending on how the fitting is designed).    On the plus side this casts less abrupt shadows but on the minus side for some uses a more direct downwards illumination may be better.  The other thing that makes a surprising difference is the colour of the walls; white walls make a huge difference so if they are a darker colour then painting them is a good idea.

Coating the shade inside (eg with aluminium tape) can improve the amount of light cast by this sort of fitting, and angling the shade differently (easier with some fittings that others) helps too. New tubes help.  Maybe you can try some tweaks and then use lower power/output tubes in the same fitting?

IIRC there is some very high inrush current when you switch a fluorescent light fitting on. Do check but I think this means that it isn't a great idea to switch them on and off too often (a few times a day is OK, a few times an hour is less good) and I think it also means that the switchgear needs a higher rating than you might expect.

cheers

[rafletcher]

IIRC there is some very high inrush current when you switch a fluorescent light fitting on. Do check but I think this means that it isn't a great idea to switch them on and off too often (a few times a day is OK, a few times an hour is less good) and I think it also means that the switchgear needs a higher rating than you might expect.

cheers

Actually switching them frequently doesn't seem too much of an issue, we had PIR sensors in my last office for occupancy switching of fluorescents. It may of course shorten their life. 

I'll be swapping my tubes over this afternoon, so I'll soon see if the reduction in Lumen is noticeable. As an aside I watched a programme about the human bodyclock recently, and part of that was the effect of sunlight on us. Remarkably the detectable light level (in what units I can't recall) was significantly reduced even by sitting inside a  car, but I don't practically notice it.  I hope the same applies in the shed in practice (the walls of which are white painted, though the floor is black)!!

[andytheflyer]

IIRC there is some very high inrush current when you switch a fluorescent light fitting on. Do check but I think this means that it isn't a great idea to switch them on and off too often (a few times a day is OK, a few times an hour is less good) and I think it also means that the switchgear needs a higher rating than you might expect.

cheers

Actually switching them frequently doesn't seem too much of an issue, we had PIR sensors in my last office for occupancy switching of fluorescents. It may of course shorten their life. 

I'll be swapping my tubes over this afternoon, so I'll soon see if the reduction in Lumen is noticeable. As an aside I watched a programme about the human bodyclock recently, and part of that was the effect of sunlight on us. Remarkably the detectable light level (in what units I can't recall) was significantly reduced even by sitting inside a  car, but I don't practically notice it.  I hope the same applies in the shed in practice (the walls of which are white painted, though the floor is black)!!

I'm awaiting your results with great interest and anticipation!  Even more so as we seem to have the same family name!

[rafletcher]

Well, as one fletcher to another (no relation I’m sure!), I can say that despite the replacements being 18w vs the original 40w, subjectively there is more than sufficient illumination (due in part I’m sur3 to the reduced spread as a result of the inbuilt reflector), there aren’t any sharp shadows, and the 6000k light temperature isn’t, in the shed situation, too “cold”. In all, a success at minimal cost. 

[PhilO]

I have the 5ft single ‘tube’ (39w) version of this fitted in my kitchen: http://www.sylvania-lighting.com/product/en-int/category/interior-lighting/industrial/families/sylbatten-led/sylbatten-led-pir

It’s far brighter than the twin-tube fluorescents in the garage, and I wouldn’t be without the automatic switching now.

[Quisling]

It sounds like you have the old magnetic ballast T8's in which case just upgrading to high frequency T5's would give you 25%+ energy savings and would be a much cheaper install than full LED upgrade.

LEDs will give you a very decent energy saving BUT you need to look beyond comparing lumen output (which others have noted is onmidirectional (and therefore includes a good deal of waste in a conventional T8) and look at comparing usable lux at floor or worktop height.  Improved colour rendition can also deliver improvement in perceived light for the same Wattage.

Ideal solution for me would be to replace the whole luminaire with a new LED unit designed to deliver appropriate lux rather than just a like for like replacement based on crude tube length.  And I'd add a PIR, but make sure it has an adjustable delay-off time function. 
However, given the low absolute burn hours each day it'd never pay back from the energy cost savings on a single light replacement.  But think about the environment darling 
I've been involved in re-lighting multistorey car parks and hospital corridors with LED tubes and we've seen anywhere between 50 and 85%+ kWh savings depending on what was being replaced and what additional controls are added.

[andytheflyer]

Having looked at the costs, going LED does not seem worth it in terms of my electricity costs - and, whilst, as a geologist, many of my projects were about environmental protection and I did my bit, I can't see any payback for me in reduced energy costs.  So the tubes will stay.

I can, however, see the benefit of making sure the tubes are not on when they are not needed, so I've bought a PIR (with a variable time delay adjuster) and will now fit that.  I would have fitted it on Sunday, but having taken the wiring to bits, I discovered I needed 3-core (not 2 and earth) for the PIR as it needs a permanent live as well as neutral and the load, so had to put it all back together again.

I now have the cable, the wife's out all day tomorrow, so that seems like the time to get the job done.

Thx for the advice given to my OP.  Much appreciated.  Changing normal house bulbs for LEDs is a given but changing the tubes seemed to be a different question.

[andytheflyer]

PIR fitted, adjusted and seems to work as wanted.  That will save SWMBO the bother of switching the garage light off when she comes back from walking the dog first thing, so that the light's left on all day if I happen not to go in there to fettle something that day.

Now, having worked out how to wire a PIR, what else would benefit from automatic swithcing off?  Hmm.....  (goes off to attempt deviousness......)

[Wowbagger]

I just replaced a fluorescent tube above my desk. I had a look at the possibility of an LED substitute and bought one from B & Q. But when I fitted it, it didn't work. The LED tube came with its own starter, but the starter clearly had a bit of plastic broken off its case. I took it back and the only other one they had accessible (apparently there was more stock on a high shelf, about 6 metres above ground level, but there was no means for anyone to access it) was already broken, so I just bought another fluorescent tube.

I could have driven out to Wicks to get one, but that's over 5 miles away and it would take a long time to save the electrons expended on that journey.

[Polar Bear]

Having looked at the costs, going LED does not seem worth it in terms of my electricity costs - and, whilst, as a geologist, many of my projects were about environmental protection and I did my bit, I can't see any payback for me in reduced energy costs.  So the tubes will stay.

...

How about the bigger picture of reducing energy use?  I'm sceptical as to personal cost savings given the much higher purchase price but I was happy to see that our electricity consumption fell when we replaced incandescent bulbs with led and how much less heating we do when we Improved the insulation of the house.  Repayment times for both stretch into decades.

Of course, I realise that we could damage the profits of large energy companies and the dividend payouts to their shareholders but that's another debate.  🙂

[Valiant]

I've slowly been replacing old 8ft (magnetic 120w) and 6ft (HF 78w triphosphor) tubes at the town hall with 6ft LED units using iirc 48w. The instant ramp alone is worth it, but also I seem to have way more useable light and it's a much higher quality too.