The LED lighting project - Page 5
 

The LED lighting project

Started by bobofthenorth, February 22, 2009, 05:21:18 PM

Previous topic - Next topic
Print
Go Down Pages 1 2 3 4 5
User actions

JohnEd

#60
April 10, 2009, 10:21:00 PM
PVCCES aka Tom,

The voltage across the led would rise with the charge on the cap and at the same rate.  The circuit values would determine the time needed.(RC time constant..I think)   Same result of putting a coil in series with the led.  All of this is becoming to complicated as I expect to make a few different size "lights" composed of many leds.  They just seem to be the only way to go.  My problem in the past was the temp of the light and having to put 50 of them together to get the lumen's I wanted.  It seems everything has changed and I am liking it.

Thanks for you thoughts,

John
"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
"We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light."
—Pla

boogiethecat

#61
April 11, 2009, 08:37:23 AM
Capacitors and inductors absolutely aren't necessary with LEDs, because LEDs don't mind being turned on and off rapidly. 
Case in point, a lot of data is transmitted optically thru fibers using simple red or infrared LEd's modulated at high megahertz rates in music synthesizer systems (adat optical interface, etc) and in millions of other applications (your TV's remote being another example).  Nothing special about the LEds used in these apps- you can modulate any LED from full on to full off in nanoseconds and they're happy.
(white LED's don't care either, but due to their glowing phosphors they don't "optically" go on and off instantly, but that's nothing that a busnut would ever need to care about either).

You're probably thinking of incandescent lightbulbs, whose filament resistance is something like 10 times less when they're off than when they're on. They have a heck of an inrush current while they're coming up to operating brightness.  LEd's don't do that... ALL they care about is that they don't have too much current pushed through them, and that they aren't reverse biased beyond their specification (a static electrical spark hitting them before they are soldered in to a circuit for example)...beyond that they don't care much about anything else.
Cheers
G
1962 Crown
San Diego, Ca

JohnEd

#62
April 11, 2009, 12:07:00 PM
Boogie,

That isn't my lie, I'm just repeating it. :P

Lots of talk in this thread about Soft Start features in current regulating devices.  What am I missing?  I know the resistor was a current limiting device albeit a poor one in some of the aspects of performance.  I'm so confused ??? :'( :(  They never said it would be this hard :-[

Thanks,

John ;D
"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
"We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light."
—Pla

boogiethecat

#63
April 11, 2009, 02:05:30 PM
John, it really isn't this hard.  This thread has, in places, gotten blown way out of proportion.  The current limiting resistor is the industry standard, it's simple, it performs well, and it's only the surface tension of a few people here who aren't engineers that is continually making this to appear difficult.
  There are definitely places in life for active current regulators (switcher ic's etc) but in probably 95% of all LED applications on the planet, you'll find the lowly resistor doing the job effectively, simply, inexpensively and reliably. Electronic regulators are absolutely great when you're driving a luxeon 1 watt LED on 12 volts, because they require a few amps rather than a few milliamps.  Resistors in that case would be a bad choice- they'd work but you'd be dumping 10 watts thru them and the heat would be hasslesome. But for strings of small 20Ma led's or individual ones, a simple resistor is the way to go.

If people want to use switching electronics to drive their reading light LEDs  tail light LEDs, turn clicker LEDs etc, let em have at it. There's nothing more amusing than watching silly creatures bang their heads on a concrete wall when you can giggle knowing that they don't have to...  :)

Ok off to melting some glass for the afternoon...

G
1962 Crown
San Diego, Ca

Tim Strommen

#64
April 11, 2009, 05:48:45 PM
Quote from: bobofthenorth on April 10, 2009, 08:36:51 AM...I will however be trying some of the LEDs that Tim referred to.  I have difficulty believing that they are really 80x as bright as the ones I am using but even if they are 10x then they would be worth the extra bux...

Here is the conversion from candelas to lumens ("q" is the beam angle):

Lumens = (Candelas x 2pi (1 – Cos (q/2) )

The Lumileds Rebel 100lm Cool-White LED P/N: LXML-PWC1-0100, has a beam angle of about 140 degrees.  This is about 24200 millicandela (about 13x the brightness of the Hiteq LED for the same segment of beam angle - which is to say that if you poked a hole through some thick cardboard to block all of the light except for a 75 degrees beam {like the hiteq part}, it would be 13x brighter for that small cone, and in this example we'd be wasting the blocked light {more than half of the emitted light} that doesn't make it past the cardboard).

The Lumileds K2 TFFC 220lm Cool-White LED P/N: LXK2-PWC4-0220, has a beam angle of about 140 degrees also.  This works out to about 53200 millicandela (about 28x the brightness of the Hiteq LED for the same segement of beam angle).

The Lumileds "lambertian pattern" LEDs put more light over a wider area than the Hiteq parts.

The beam angle as well as the light output has the most to do with the usefulness of an LED - for a given work-surface you need so much light per square foot, and thus so many LEDs at a given height and direction to fully cover the surface with the appropriate amount of light for the task.  The light can either be directly used from the LED device (with a narrow beam) which gives a light fixture the "thousand points of light" look, or you can use a wider-angle LED and some form of optics to direct the most light where it is needed.  To place the LEDs closer to the work-surface, you need a wider beam angle.  I don't personally prefer the "thousand points of light" look, I like my light fixtures to be smooth and even (less eye strain, sexier appearance).  I use LCD backlight light-pipe material from 3M to even out the light in a very thin piece of plastic.  You can also use lenses and reflectors that are made specifically for LEDs (Future Electronics has several different types made specifically for Lumileds LEDs).

-T
Fremont, CA
1984 Gillig Phantom 40/102
DD 6V92TA (MUI, 275HP) - Allison HT740
Conversion Progress: 10% (9-years invested, 30 to go :))

JohnEd

#65
April 11, 2009, 08:00:00 PM
El BooGay,

Munch o grassy @$#. ??? ;D

You are gifted at turning a phrase. :)

John
"An uneducated vote is a treasonous act more damaging than any treachery of the battlefield.
The price of apathy towards public affairs is to be ruled by evil men." Plato
"We can easily forgive a child who is afraid of the dark; the real tragedy of life is when men are afraid of the light."
—Pla

bobofthenorth

#66
July 19, 2009, 08:43:28 AM
The naysayers will be relieved to hear that 2 of my resistorless LED modules finally failed.  Five months is not an acceptable lifespan but it is worth noting that nothing caught on fire and the cat was not injured.  I've still got modules running with 10, 22 & 50 ohm resistors inline with them but I won't put up any more with no resistors.   >:(  ::)

Compared to the commercial bulbs I'm still well ahead of the game.  I've replaced 5 of them in the same time span.
R.J.(Bob) Evans
Used to be 1981 Prevost 8-92, 10 spd
Currently busless (and not looking)

The last thing I would ever want to do is hurt you.
Its the last thing but its still on the list.

Tim Strommen

#67
July 19, 2009, 08:18:48 PM
Hey, we're not happy/relieved that your resistor-less fixtures failed - just not suprised.
Fremont, CA
1984 Gillig Phantom 40/102
DD 6V92TA (MUI, 275HP) - Allison HT740
Conversion Progress: 10% (9-years invested, 30 to go :))

bobofthenorth

#68
July 19, 2009, 08:53:13 PM
Quote from: Tim Strommen on July 19, 2009, 08:18:48 PM
Hey, we're not happy/relieved that your resistor-less fixtures failed - just not suprised.

You were always more of an another-way-sayer than a naysayer.   ;D
R.J.(Bob) Evans
Used to be 1981 Prevost 8-92, 10 spd
Currently busless (and not looking)

The last thing I would ever want to do is hurt you.
Its the last thing but its still on the list.

Melbo

#69
July 19, 2009, 08:59:40 PM
we are glad the cat is ok

sometime the innocent get injured

Melbo
If it won't go FORCE it ---- if it breaks it needed to be replaced anyway
Albuquerque, NM   MC8 L10 Cummins ZF
Print
Go Up Pages 1 2 3 4 5
User actions