Cheap multi-color LED Christmas light strings.

[Winfield]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

 

[Spehro Pefhany]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

42mA is a lot of average current for a small LED, especially the blue
and green ones.


Best regards,
Spehro Pefhany

 

[T]

42mA is a lot of average current for a small LED, especially the blue
and green ones.


Best regards,
Spehro Pefhany

I hooked up 10 high brightness LED's to my power supply. They're all run
through 1K resistors and it's drawing 30ma at 5VDC, so if if you divide
it out by the 10 LED's it would mean approximatley 3ma per LED.

 

[Jim Yanik]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

42 ma seems a bit high for the LEDs,doesn't it?

 

[[email protected]]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

Where did you get the lights? Price sounds good.

You might be buying a few more strings if the lights burn out from the
higher current. Do a burn-in test yet?

M

 

[D from BC]

42 ma seems a bit high for the LEDs,doesn't it?

Maybe for indoors, but chilled LED's (outside in the cold) can
probably last longer with higher current.


D from BC

 

[Spehro Pefhany]

I hooked up 10 high brightness LED's to my power supply. They're all run
through 1K resistors and it's drawing 30ma at 5VDC, so if if you divide
it out by the 10 LED's it would mean approximatley 3ma per LED.

Fine, now suppose you were to put them in series at 42mA as in the
above situation, what would you calculate the current per LED to be?


Best regards,
Spehro Pefhany

 

[Jim Yanik]

Maybe for indoors, but chilled LED's (outside in the cold) can
probably last longer with higher current.


D from BC

most common,small LEDs max at around 25ma,would cold allow almost double
that current? It doesn't seem likely.
But a pulsed LED would allow a greater max current,as the duty cycle would
be less than for 'pure' DC.
less than 50% for half-sine.

I guess it all depends on what LED chip they are using.

 

[Spehro Pefhany]

most common,small LEDs max at around 25ma,would cold allow almost double
that current? It doesn't seem likely.
But a pulsed LED would allow a greater max current,as the duty cycle would
be less than for 'pure' DC.
less than 50% for half-sine.

Given a fixed maximum power dissipation, the average current has to be
*less* for pulsed current because there is more voltage drop at higher
currents.

I guess it all depends on what LED chip they are using.

And on the packaging.

Best regards,
Spehro Pefhany

 

[Paul Hovnanian P.E.]

42mA is a lot of average current for a small LED, especially the blue
and green ones.

"The light that burns twice as bright burns half as long ..." -- Eldon
Tyrell

 

[Martin Griffith]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

could you use something like this?
http://www.reuters.com/news/video?videoId=72584&newsChannel=technologyNews


martin

 

[WJLServo]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

My wife sweet talked me into making her an LED XMass string, some
years back. She had tagged along with me on a visit to the old Eli
Heffron / Solid States Sales surplus shop in Cambridge, MA, US, and,
while I was looking at used test equipment, bought herself a couple
bags of red and green LEDs. She raided my stash of blues when we got
home (Eli's didn't stock them) and demanded that I make them into
XMass lights for her.

So, got a 5 volt wall-wart and a bag of 150 ohm resistors, one ballast
resistor per LED, and soldered it all together. She wrapped the string
around a small evergreen tree that she's had growing indoors,
"bonsai" style, for a good many years, and lit it up for the holidays.
Looked pretty, but, took up way too much time soldering....

W Letendre

 

[Don Lancaster]

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

This trick dates from at least the 1960's, when you could drmatically
improve an electric shaver's performance with a diode and a capacitor.


--
Many thanks,

Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: [email protected]

Please visit my GURU's LAIR web site at http://www.tinaja.com

 

[Spehro Pefhany]

My wife sweet talked me into making her an LED XMass string, some
years back. She had tagged along with me on a visit to the old Eli
Heffron / Solid States Sales surplus shop in Cambridge, MA, US, and,
while I was looking at used test equipment, bought herself a couple
bags of red and green LEDs. She raided my stash of blues when we got
home (Eli's didn't stock them) and demanded that I make them into
XMass lights for her.

So, got a 5 volt wall-wart and a bag of 150 ohm resistors, one ballast
resistor per LED, and soldered it all together. She wrapped the string
around a small evergreen tree that she's had growing indoors,
"bonsai" style, for a good many years, and lit it up for the holidays.
Looked pretty, but, took up way too much time soldering....

W Letendre

The little incandescents (~1mm diameter) are more attractive than most
current LEDs on a mini tree.


Best regards,
Spehro Pefhany

 

[Don Klipstein]

My wife bought some multi-color LED Christmas-light strings,
50 lights for $10 per string. We used them to decorate our
traditional wreath over the door, but I was disappointed with
their brightness. Checking them on the bench, I found that
having no bridge rectifier, they operated on only one half of
the AC line cycle, and with 50 LEDs in series, they needed
125 to 135 volts to draw a even modest 10mA current.**
Their conduction angle was also less than one might like.

BTW, half-cycle operation makes a 50 or 60Hz flicker, not
a 25 or 30Hz flicker as many web-sites state.

Clearly, adding a bridge rectifier would double both the LED's
flicker rate, and light output. But I elected to add a storage
capacitor after the bridge (200uF, for 2 to 3 strings), and feed
the LED string with DC. This would make the conduction
angle 100%, and further increase the light output.

My in-situ measurements: 162 volts and 42mA per string,
about 5x the original estimated current. Yep, they looked
about five times brighter. At least that's what I told my wife,
after she said they didn't look _that_ much brighter.

** The string has 20 yellow LEDs, and 10 each red, green
and blue. I guess the band-gap voltages add up to ~ 110V.

42 mA would usually appear less than 4.2 times as bright as 10 mA
because:

a) Nonlinearity in human vision. Keep in mind that a 60 watt
incandescent lightbulb looks 60% as bright as a 100 watt one, while being
only 50% as bright.

b) The LEDs would run hotter and their efficiency would suffer

c) Efficiency of the LEDs vary with instantaneous current. I expect the
reds and yellows to mostly operate fairly linearly, possibly be
extremely slightly improved except for their higher temperature because
their efficiency is less at less than a few mA, and without the smoothing
capacitor a small percentage of the energy going into the LEDs goes in
during low current operation that impairs the reds and greens.
However, I expect this to be outweighed slightly by higher temperature.

Same story for the green ones if they are the dimmer very yellowish
green. If they are brighter non-yellowish green, then they will behave
like most blue ones - see below:

Most blue LEDs, especially turquoise blue ones and very deep blue ones,
have a different nonlinearity where efficiency is maximized at a few mA
(for the lower power size ones, as opposed to ones rated 150 mA or more).
With a smoothing capacitor, they will be receiving current near the
maximum instantaneous current received without a smoothing capacitor, so
their efficiency will probably be reduced measurably (though maybe as
little as a few percent) by use of that capacitor.

Of these factors A, B and C, I would give A the most weight.

- Don Klipstein ([email protected])

 

[Don Klipstein]

most common,small LEDs max at around 25ma,would cold allow almost double
that current? It doesn't seem likely.
But a pulsed LED would allow a greater max current,as the duty cycle would
be less than for 'pure' DC.
less than 50% for half-sine.

In general, modern high brightness low power LEDs, when given 25 mA
average current, give maximum efficiency with steady DC as opposed to
pulsing. This is especially true with the InGaN ones, and really
especially with blue and green InGaN ones.

I guess it all depends on what LED chip they are using.

- Don Klipstein ([email protected])

 

[Winfield Hill]

42 mA would usually appear less than 4.2 times as bright as
10 mA because:

a) Nonlinearity in human vision. Keep in mind that a 60 watt
incandescent lightbulb looks 60% as bright as a 100 watt one,
while being only 50% as bright.

That's true, but these LEDs really do look much brighter
in-situ, I suppose because the LED light output is being
compared with the surrounding reflected light from street
lights, porch lights, the moon, city sky light, auto
lights, etc., where the absolute light level certainly
does count, as opposed to being compared with no light.

b) The LEDs would run hotter and their efficiency would suffer

Hasn't been to be too bad at 40mA. I'll take measurements
and report back.

BTW, many years ago I made a very intense light source from
hundreds of "ordinary" red LEDs (with the exact wavelength
we needed to process bags of blood), and ran them at 150mA.
The light output vs current began to drop above 80 to 100mA,
but it was still effective to run them at 150mA for maximum
light. I soldered the LEDs to a copper bar and used a huge
fan to remove heat as we could manage. It worked quite well,
lots of light, far more than you could stand to look at (and
we got very rapid blood processing, minutes instead of hours).
We called it the "LED Blaster" and my technician affixed a
sign, like the laser warning signs, saying DANGER LED LIGHT.

It worked well, except every now and then one of the LEDs
would explode. No warning or dropping light level, etc., it
would just explode into many small bits of plastic. (The
guys at the Red Cross, who were performing the blood tests,
were not amused.)