Low profile downlights

[ConcreteCastle]

Hello,
I'm currently putting up a new ceiling and would like LED downlights, however under the current plan I would have a very shallow void of about 40mm plus 12.5 plasterboard. I've looked online but it's hard to find anything that will definitely work in this space. There are a couple that might just, but it would be very tight. Am I flogging a dead horse with this ?

If so my options appear to be lowering the ceiling further or scrapping the idea and having normal lights.

Just thought I'd ask in here if anyone either knows of a super shallow LED downright, or alternative set up to give the same effect? Or has any other ideas for making this work?

If it's relevant the room is 9'x9' with 8' ceiling and reasonable natural light so my plan was four 9w (c 400lm) LED downlights with a dimmer.

Thanks in advance

 

[ban-all-sheds]

scrapping the idea and having normal lights.

I like that one.

If it's relevant the room is 9'x9' with 8' ceiling and reasonable natural light so my plan was four 9w (c 400lm) LED downlights with a dimmer.

My advice would be to consider whether needing 4 lights in a room that small might indicate that the lights aren't very good at doing the job of lighting up a room...

Why is the void so shallow?

 

[ConcreteCastle]

I guess you're not a fan?

Interested to hear why as I don't have any particular axe to grind for them, I just like the look, the energy use snd the relative lack of maintenance.

I'm not sure it would actually need that much but having 4 spreads the light nicely and with a dimmer it can be adjusted to the right level.

The void is shallow because it's a concrete ceiling above and I'm planning on mounting plasterboard on 40x60 wooden joists. I could lose a bit more ceiling height, but I'd rather not a) have lower ceilings and b) have all that timber up there.

Thanks

 

[DetlefSchmitz]

Energy Use: 4 times as much as you need.
Lack of Maintenance: Until one goes wrong.
LEDs are quite difficult to dim reliably, you would need to check them out before fitting to see if the particular combination worked. When a lamp fails you may find that it is no longer available and you have to replace all 4 plus the dimmer.

A single thin surface mounted fitting would be much more appropriate.

 

[SFK]

ConcreteCastle
Would something like this work for you:
https://www.amazon.co.uk/TryLight-R...t+dimmable&dpPl=1&dpID=31U49wYwjRL&ref=plSrch

The search I used was
Round dimmable led panel light

Sfk

 

[ConcreteCastle]

Detlef
Fair point about maintenance. 25,000 hours sounds a lot but if I have to pull the ceiling down every couple of years it's a bit of a pain. I will look into surface mounted ones too.

SFK
That does look good. Thanks. I'm not wild about the style of that particular one but good to know they do exist. And anyway I don't need to be "wild" about the look of some lights!

Thanks

 

[JohnW2]

Fair point about maintenance. 25,000 hours sounds a lot but if I have to pull the ceiling down every couple of years it's a bit of a pain.

... and that's assuming that they do last for 25,000 hours. From what I understand, most don't and many fail far earlier than that.

Don't forget that even if they fail during warranty, the most you can hope for is a 'similar' (but probably not identical) replacement (the one you bought would almost certainly be 'out of production') or your money (for the failed one) back. If you have four, and want them all to look the same, you'd probably have to fork out for three more that matched the replacement - and, of course, the warranty would not pay for pulling down your ceiling and reinstating it

For what it's worth, if/when I buy multiple matching things (like light fittings), I try to buy a spare or three at the time, before they 'become obsolete'!

Kind Regards, John

 

[ConcreteCastle]

My maths was at 5 hours a day that gives you 5000 days use which is 13.5 years. If they last half this long it's not terrible. Though maybe that's still optimistic? A couple of spares is definitely worth while though.

 

[JohnW2]

My maths was at 5 hours a day that gives you 5000 days use which is 13.5 years. If they last half this long it's not terrible. Though maybe that's still optimistic? A couple of spares is definitely worth while though.

Don't forget that the 'life expectancy' quoted for this sort of product is usually the 'half-life' of a large number/batch of them. In other words, if the life is stated as, say, 25,000 hours, that will usually mean that half of them are expected to fail by (i.e. 'at or before') 25,000 hours - but that statistic gives no information about 'how early' will be the failure of those which fail before 25,000 and, in some cases, that will be 'very early'.

Kind Regards, John

 

[EFLImpudence]

I don't know how they get away with such a statistic.

If the OP has two fail in a relatively short time, does that mean the other two will definitely last that long?

No? so an entirely meaningless lie.

 

[JohnW2]

I don't know how they get away with such a statistic. If the OP has two fail in a relatively short time, does that mean the other two will definitely last that long? No? so an entirely meaningless lie.

The problem with survival/failure statistics of any sort is that they are of very limited value in relation to single items or small number of items - and that's as true of light bulbs as it is of human survival. It is all inevitably 'probabilistic', which means that nothing is ever 'definite' - but there is really nothing better which can be done.

The fact that a person of a certain young age has a certain (high) probability of still being alive at, say, 60 does not mean that they definitely won't die tomorrow. If one of a randomly selected group of such people does die tomorrow then that doesn't mean that another of that group will definitely not die in the near future.

If 'survival curves' are available, they will enable one to ascertain the probability (sorry!) of failure during any time period of interest. However, the nature of probabilistic processes is such that, even if the probability of failure between, say, 1 and 6 months were "1 in a million" doesn't mean that it definitely won't fail on the first day of that period.

These sort of survival/failure statistics are primarily of value to purchasers of large numbers of items, in relation to budget and maintenance planning - i.e. the figures tell them that they have to budget and plan to have to replace ('on average') a certain number of them over a certain period of time.

That doesn't mean that these statistics are useless for ordinary purchases of single items, or small numbers of items, but it still remains probabilistic. Since the survival curves for a particular type of item are generally all of a similar shape, a longer median survival time (e.g. 25,000 hours rather than 10,000 hours) will usually mean something like a pro-rata difference in the probability of a single item failing in any given period of time - but probabilities ('chances', if you prefer) obviously are not the same as certainties.

Kind Regards, John

 

[EFLImpudence]

It still seems wrong to just put 50,000 hours on the packet when people don't know that.

 

[JohnW2]

It still seems wrong to just put 50,000 hours on the packet when people don't know that.

It's certainly desirable that people should understand the information they are being given.

However, I think that, in general, people do understand the concept of 'on average' - and, although there are different ways of expressing that, the method used for the life expectancy of lamps/bulbs etc is probably as reasonable as any. For example, if I told a 20 year-old that their ('on average') life expectancy was a further 62 years (i.e. live to age 82) I think it would be generally understood and accepted that that 'on average' life expectancy didn't preclude the possibility of their dying of some natural cause 'tomorrow' (or, at least, in the relatively near future).

As I wrote before, I don't see that there are really any alternatives - unless you have one to suggest? The probabilistic nature of the issue is such that there is absolutely nothing 'definite/guaranteed' that can be said (e.g. "this lamp is guaranteed to last for at least X hours of use"). To say that, 'on average' a lamp will last for 25,000 hours (which will probably be taken to imply that about half will last <25,000h, and about half >25,000h, with an 'average' of 25,000h) seems as good a way to convey the information as any - if you disagree, what would you prefer?

Kind Regards, John

 

[EFLImpudence]

They could then state that the figure is an average.


I have some which state simply "30,000 hours Lamp life".

 

[JohnW2]

They could then state that the figure is an average. I have some which state simply "30,000 hours Lamp life".

Agreed. If it doesn't say 'average', it probably would be clearer if it did say so explicitly.

However, one has to wonder what people could think it could be other than an average of some sort. The great majority of people surely have more sense than to think that "30,000 hours Lamp life" means that the product is 'guaranteed' to last at least 30,000 hours?

People have got used to figures based on averages (which don't explicitly state that they are based on averages) in relation to the labelling of the weights/volumes/whatever of consumer (and other) products. You and I are old enough to remember the days when, say, a box of Corn Flakes would be labelled something like "contents not less than 8oz". However, that was pretty unrealistic and was, decades ago, replaced (with legislated approval) by a probabilistic figure (related to the average), designated only by the 'e' symbol (the meaning of which is probably unknown to nearly all consumers). However, if the Cornflakes are now labelled "250ge", I think most people probably realise that that does not mean "not less than 250g", even though there is no explicit statement as regards what the figure means.

Kind Regards, John