On 15/11/2017 23:36, Johnny B Good wrote:
On Wed, 15 Nov 2017 13:57:50 +0000, Martin Brown wrote:

On 15/11/2017 13:48, Brian Reay wrote:
On 15/11/2017 12:18, Clive Page wrote:

assume that the claimed lifetime of 50,000 hours (a decade or two at
our typical usage) is just a wild extrapolation from their testing
programme and wouldn't rely on that at all.

Their demise can also be hastened by putting them in unsuitable fixtures
where heat builds up sufficiently to dry out the PSU capacitors. The
sealed glass globe type are particularly bad for them.

That would certainly be true of the older, less efficient (81Lm/W) types
we've had to contend with these past 5 years or so - during the last 3
years of which we've been kept waiting for the more efficient LEDs to
land on the shop shelves any time during the 18 to 24 months the Cree
lighting spokesperson had promised that the 303Lm/W lab samples would
take to get into production. Only now are we seeing improved LED lamps
appearing with moderately improved efficiencies of 125Lm/W which falls
woefully short of Cree's laboratory achievements of... let me check... 26
March 2014, way over 3 1/2 years ago.

https://www.forbes.com/sites/peterde...will-get-even-
more-efficient-cree-passes-300-lumens-per-watt/#65748cea2611

Read it and weep (sorry for the line wrap)!

The 300Lm/W was achieved on a near infinite heatsink under perfect
laboratory conditions with a fabrication technique that doesn't easily
scale up - basically a one off research device made of unobtanium.

Around 130Lm/W is the current production device which is still
impressive and better than HPS (but still short of 220Lm/W LPS).

I'm unconvinced re LED lights in the domestic setting, not so much due
to the life (although it is good to see a seller honouring his
warranty) but more the light output.Â* We have them in our motorhome
and, in the confined space, they are OK. However, we tried them in the
kitchen (replacing some halogen bulbs - 2x40W halogen per fitting) and
with LEDs they were hopeless. I forget the rating of the LEDs but they
were, supposedly, equivalent to 40W bulbs. At a guess, they were more
like 25W.

Odd. I would agree where CFL are concerned that equivalent light output
was vastly overstated, but all the LED bulbs I have ever bought have
produced something approximating their claimed equivalent brightness.
The first one I bought a nominal 60W equivalent was far too bright in a
small bathroom (the "60W" CFL was about right but failing due to age). A
40W equivalent LED bulb did the trick.

You can now buy such "40W" (6W 510lm) LED GLS lamps in Poundland these
days (both LES and BC22 types)... for just a quid each (not *everything*
is priced at one pound in Poundland and Poundworld shops these days).

It is astonishing how the prices have fallen.

I suspect when Philips first marketed their SL 'Comfort' 13 and 18 watt
range of CFLs, the industry wasn't facing any regulatory need to specify
the details of the "Incandescent GLS" reference lamp they were pitting
their product's performance against.

Quite possibly they were using a long life/rough service 220v (or
perhaps, the even lower efficacy 240v UK standard) filament lamp as their
"Standard", relying upon the characteristic 120% of "Design Lumens"
output from new of the fluorescent tube light output life curve to help
mask this discrepancy between these CFLs and regular 1000 hour rated GLS
220/240 volt filament lamps.

I think the CFL claims were just bare faced lies on the part of the
marketing men and slimy salesmen.

By the time LED lamp technology started to become a viable alternative
to the CFL, Much tighter consumer protection regulations came into force
(presumably in the US of A judging by the use of the American 806lm 60W
120v 750 hour lamp standard chosen for the sale of LED lamps claiming to
have a 60W equivalency). It's for that reason alone that we in the UK
(and likewise for Europeans no doubt) have a much better experience when
upgrading old 60 and 100 watt incandescent GLS lamps (even more so when
replacing ageing CFL versions which claim the same incandescent lamp
wattage equivalence).

I think part of it is that there is a global standard now and US rules
based on their lower voltage incandescent filament bulbs have made LED
nominal equivalents brighter as a result. Thicker low voltage filament
can run hotter and be more efficient.

The real benefit of efficiency improvements in LED lamp performance lies
not so much with reduced electricity bills so much as being able to
upgrade from a 10W 810Lm LED to a 10W 1520Lm LED in a light fixture that
provides barely sufficient cooling for the 10W 810lm lamp with the
confidence that the newer brighter lamp will run just that little bit
cooler (and quite possibly realise the 25 to 30 thousand hour promises
typically made these days) since even less of that 10W input gets turned
into waste heat.

I think in an ideal world new LED luminaires should be designed to take
into account the strengths and weaknesses of the physical devices. This
is now happening but the change over will be slow. Some types of older
sealed glass bowl lamp shades spell a short hot life for LED bulbs.

--
Regards,
Martin Brown