DIYbanter - Dimming an LED

DIYbanter (https://www.diybanter.com/)

- UK diy (https://www.diybanter.com/uk-diy/)

- - Dimming an LED (https://www.diybanter.com/uk-diy/686315-dimming-led.html)

Lonely Obnoxious Cantankerous Auto-contradicting Senile Ozzie Troll Alert!

On Thu, 29 Apr 2021 05:11:40 +1000, cantankerous trolling geezer Rodent
Speed, the auto-contradicting senile sociopath, blabbered, again:

FLUSH the trolling senile asshole's latest troll**** unread

--
Cursitor Doom about Rot Speed:
"I'm not the least surprised. The man is a conspicuous and unashamed
ignoramus."
MID:

"NY" wrote in message
...
"Tim+" wrote in message
...
Dave Plowman (News) wrote:
Think it's generally known that dimming a tungsten bulb reduced its
efficiency dramatically. In other words, a 100w dimmed to half
brightness
used a lot more electricity than a 50w on full. Although dimming did
extend the life of the bulb.

So how about LEDs? Since they generate far less heat are they still as
efficient when dimmed? And does dimming make a difference to their life?

Until they can make the colour temperature of LEDs drop as they dim, Ive
decided Im going to just avoid using dimmers with LEDs as the just look
ghastly as the dim IMO.

Why would you want the colour temperature of an LED to drop as it is
dimmed?

He prefers that effect.

I know it is an unavoidable side-effect of dimming a tungsten bulb, but I
don't see why you would want it to happen unless you couldn't avoid it. I
like the fact that LEDs keep the same colour as they dim.

I do too, I hate the very yellow effect.

I just wish LED bulbs could be dimmed further: most bulbs (eg Philips Hue)
have a minimum brightness below which they go out altogether. It makes
them less useful as a very dim night light.

I dont need a very dim night light.

Lonely Obnoxious Cantankerous Auto-contradicting Senile Ozzie Troll Alert!

On Thu, 29 Apr 2021 05:40:24 +1000, cantankerous trolling geezer Rodent
Speed, the auto-contradicting senile sociopath, blabbered, again:

FLUSH the trolling senile asshole's latest troll**** unread

Max Demian wrote:
On 28/04/2021 13:13, Paul wrote:
Max Demian wrote:

(I've seen idiot YouTube videos describing circuits that control LEDs
with potentiometers, either directly or via a power transistor; it's
inefficient and liable to overheat the pot and/or the transistor as
some of the comments say; also I think that controlling white LEDs in
this way is liable to change the colour of the light.)

Idiot YouTubes:
https://www.youtube.com/watch?v=wns8xrxTYhU
https://www.youtube.com/watch?v=OnjNXhxxc9w

This depends on the situation.

The beauty of design, is keeping an open mind.

It's a matter of whether you like economy and dislike smoke arising from
the circuit.

If you're referring to SMPS problems at low voltage,
an SMPS is supposed to have UVLO. It's not our fault
if a necessary safety feature is missing.

SMPS run at constant power. As the voltage is dropped on
input, the current on the primary side rises. If allowed
to go unchecked, the current flow could be too much
for components on the primary side. "Under voltage lockout"
prevents that. Even mains distribution can deliver the
wrong voltage (it happened here at the house one day), and
devices in your house should be protected. Unfortunately,
there's nothing for motors, and motors can be damaged
with too low or too high of a voltage. But an SMPS
(switch mode power supply, like in the base of high
power LEDs), there is room in the silicon for all
sorts of features, and no excuses.

The end result of "resistor" experiments carried out on a proper
SMPS bulb, the bulb should work (at constant light output),
over a range, and promptly switch off (at around half-voltage
or so, as an arbitrary design choice). Experiments carried
out by hobbyists, should leave them scratching their heads,
but no smoke in the room.

Just design your own lights, if you don't like the commercial
ones, then see how good you are. You can get nice array
LEDs, an inch in diameter, that run off 40VDC or so, and you
can use that as a starting material if you want to light
a room with just one bulb. But don't expect your project
to be small enough, to fit existing fixturing.

Maybe a dropper bulb, instead of an SMPS bulb,
would be a starting material. And preferably not
the Dubai bulb, as it has a bit of regulation
built in. Just a plain cheap Chinese dropper and
do your resistor-style experiments. With enough
dropper bulbs (each off a separate resistor), you
can light a room. Even if it costs a fortune for
a given amount of light. Dropper bulbs are preferred
by ham radio operators, as the dropper doesn't generate
RF hash like the SMPS does. With enough dropper
bulbs, you can just switch some off if you want
"less light".

https://www.powerelectronictips.com/...supplies-leds/

Paul

On 29/04/2021 02:07, Paul wrote:
Max Demian wrote:
On 28/04/2021 13:13, Paul wrote:
Max Demian wrote:

(I've seen idiot YouTube videos describing circuits that control
LEDs with potentiometers, either directly or via a power transistor;
it's inefficient and liable to overheat the pot and/or the
transistor as some of the comments say; also I think that
controlling white LEDs in this way is liable to change the colour of
the light.)

Idiot YouTubes:
https://www.youtube.com/watch?v=wns8xrxTYhU
https://www.youtube.com/watch?v=OnjNXhxxc9w

This depends on the situation.

The beauty of design, is keeping an open mind.

It's a matter of whether you like economy and dislike smoke arising
from the circuit.

If you're referring to SMPS problems at low voltage,
an SMPS is supposed to have UVLO. It's not our fault
if a necessary safety feature is missing.

The YouTube videos I linked were ways of controlling the brightness of
LED strips which typically worked on 12VDC. They may have been fed from
a SMPS, but that was irrelevant.

--
Max Demian