I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

On 13/02/2012 13:52, David Robinson wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

The present generation of high brightness white LEDs are just visibly
lit when there is a current of around 2uA flowing through them. As long
as they are not glowing brightly I wouldn't worry too much.

As they improve the lower current limit keeps on going down. Some are
more sensitive than others in any given batch. You can make an LED torch
that is always slightly on by bridging the switch with a suitable high
value resistor. This is useful in the dark during unexpected powercuts.

--
Regards,
Martin Brown

On Feb 13, 1:52*pm, David Robinson
wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

Probably stored energy in capacitors. How much depends on where the
cycle was when you switched off.

On Feb 13, 1:52*pm, David Robinson
wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

Cables have capacitance. Its not a problem in any way,


NT

On Mon, 13 Feb 2012 14:36:46 -0800 (PST), NT
wrote:

On Feb 13, 1:52*pm, David Robinson
wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

Cables have capacitance. Its not a problem in any way,

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. They cost 20x as much and only last about twice as long as the
halogens.
--
(\__/) M.
(='.'=) Due to the amount of spam posted via googlegroups and
(")_(") their inaction to the problem. I am blocking some articles
posted from there. If you wish your postings to be seen by
everyone you will need use a different method of posting.

In article , Mark i@dontget
lotsofspamanymore.invalid writes

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. They cost 20x as much and only last about twice as long as the
halogens.

What's the failure mode?

--
(\__/)
(='.'=)
(")_(")

On Tue, 14 Feb 2012 12:22:14 +0000, Mike Tomlinson
wrote:

In article , Mark i@dontget
lotsofspamanymore.invalid writes

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. They cost 20x as much and only last about twice as long as the
halogens.

What's the failure mode?

I had 3-4 and all but one just stopped lighting up. The other one
would flash on and off.
--
(\__/) M.
(='.'=) Due to the amount of spam posted via googlegroups and
(")_(") their inaction to the problem. I am blocking some articles
posted from there. If you wish your postings to be seen by
everyone you will need use a different method of posting.

In article ,
Mark writes:
On Tue, 14 Feb 2012 12:22:14 +0000, Mike Tomlinson
wrote:

In article , Mark i@dontget
lotsofspamanymore.invalid writes

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. They cost 20x as much and only last about twice as long as the
halogens.

What's the failure mode?

I had 3-4 and all but one just stopped lighting up. The other one
would flash on and off.

Unless you spend a lot of money, the LED products you find
in the shops are pretty universally crap.

Flashing on and off sounds like PSU failure.
The others could be PSU failure, or a single LED or single
series chain of multiple LEDs which have gone open circuit.

Stayed in a hotel a few months back which had fitted GU10
leds everywhere. They each had perhaps 15-20 LEDs. The two
in the bathroom had about 4 LEDs still working. I swapped
them with a couple in the corridor which had about 10 still
working, which was the most I could see still working in
any of the lamps.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

In article , Mark i@dontget
lotsofspamanymore.invalid writes

I had 3-4 and all but one just stopped lighting up. The other one
would flash on and off.

Not good. Can you remember the make?

Tried a LED GU10 in my kitchen the other day - it has 7 x 50W lamps.
Bluish light and nowhere near the 50W equivalent light output it
claimed. Looked absolutely pathetic compared to the other 6 halogens.

I might try 20W halogens - two for a quid in the pound shops - that's
140W total vs. 350W for an outlay of 4 quid.


--
(\__/)
(='.'=)
(")_(")

On 14/02/2012 10:05, Mark wrote:
On Mon, 13 Feb 2012 14:36:46 -0800 (PST),
wrote:

On Feb 13, 1:52 pm, David
wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

Cables have capacitance. Its not a problem in any way,

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. They cost 20x as much and only last about twice as long as the
halogens.

How do they fail?

The MTBF of an LED with *proper* heatsinking should be enormously long.

I presume the packaging constraints leave the things vulnerable to
overheating and thermal diffusion effects. They do tend to cook their
plastic encapsulation and darken the phosphor after a while.


--
Regards,
Martin Brown

On Tue, 14 Feb 2012 13:04:27 +0000, Martin Brown
wrote:

On 14/02/2012 10:05, Mark wrote:
On Mon, 13 Feb 2012 14:36:46 -0800 (PST),
wrote:

On Feb 13, 1:52 pm, David
wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

Cables have capacitance. Its not a problem in any way,

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. They cost 20x as much and only last about twice as long as the
halogens.

How do they fail?

See my other post.

The MTBF of an LED with *proper* heatsinking should be enormously long.

I presume the packaging constraints leave the things vulnerable to
overheating and thermal diffusion effects. They do tend to cook their
plastic encapsulation and darken the phosphor after a while.

FWIW visually the failed bulbs looked brand new.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

On 15/02/2012 10:18, Mark wrote:
On Tue, 14 Feb 2012 13:04:27 +0000, Martin Brown
wrote:

On 14/02/2012 10:05, Mark wrote:
On Mon, 13 Feb 2012 14:36:46 -0800 (PST),

How do they fail?

See my other post.

From your description it sounds like the control electronics died. I
suspect the nicely insulated fittings for halogen lamps allow LED units
to run rather hotter than they enjoy. I am surprised they expired so
quickly. I have seen plenty (not GU10) with one or two dead LEDs.

One of the great advantages of LED arrays in some applications is that
in the right packaging they are self collimating and a single point
failure just leaves a small gap in the array of working LEDs whereas a
conventional filament failure leaves you with ambiguous traffic signals.

The MTBF of an LED with *proper* heatsinking should be enormously long.

I presume the packaging constraints leave the things vulnerable to
overheating and thermal diffusion effects. They do tend to cook their
plastic encapsulation and darken the phosphor after a while.

FWIW visually the failed bulbs looked brand new.

OK. So probably not overheating failure of the LEDs themselves then.
They tend to develop black shadows over the emitting chip when seriously
abused. What you describe sounds like a bad design or a bad batch.

Follow the first link in the following document to see details of how
they could be designed for 50-100k hours MTBF according to Philips.
(assuming near infinite local heatsinking)

--
Regards,
Martin Brown

On Feb 14, 10:05*am, Mark
wrote:
On Mon, 13 Feb 2012 14:36:46 -0800 (PST), NT
wrote:









On Feb 13, 1:52*pm, David Robinson
wrote:
I've swapped the GU10 halogens in the centre of our kitchen for LEDs.

Strange thing: the LEDs sometimes glow slightly, even when switched
off.

Four different sets of lights are all switched from three locations
(2*4-gang 2-way switches, and 1*4-gang intermediate switches), and the
brightness of the glow varies depending on what other lights are
switched on in the kitchen.

I'm assuming that, because of the length over which all cables are run
in parallel, that somehow sufficient current is being induced to
illuminate the LEDs, dimly. It probably helps that the cables run
along a structural girder for 3 metres.

Do you think this is anything to worry about? I wonder if it's
possible to get a dangerous "nip" when changing a bulb? (Not that I'm
likely to put my fingers into a bayonet fitting, never mind a GU10).

I know you shouldn't bunch cables together unless suitably de-rated
(which these are), but never thought about one cable inducing a
current in another cable. Anyone ever take steps to avoid this?

Cheers,
David.

Cables have capacitance. Its not a problem in any way,

I was extremely dissapointed with the lifetime of LED replacement GU10
bulbs. *They cost 20x as much and only last about twice as long as the
halogens.

The ones in the kitchen (?8? SMD LEDs in each) have a slightly
greenish light, and a usually imperceptible flicker which gives rise
to strobing when anything moves quickly (e.g. chopping veg).

The one in the hallway (different sort - ALPHA brand - 3 LEDs in it)
died after 6 months. Just sat there flickering/flashing wildly. Guess
the 25 years predicted electricity savings won't be coming to
fruition. Still looking for the receipt to take the thing back. Shame.
As a spot light (pointing at a picture), it had a pleasant clean white
light and I really liked it. That one was useless for general lighting
though.

Cheers,
David.