Blasted Cyan Leds

[**DONOTDELETE**]

Well after working on my video light for EONS now, it is on its 3rd revision, and is just as stupid an idea as it was when i started it

of course somone has to do it

if links work here here is a link to the project. http://www.candlepowerforums.com/cgi-bin/ultimatebb.cgi?ubb=get_topic&f=3&t=000661

anyways ONE of the continuing problems is the blue/green cyan leds, two different batches of them now, and they break down, all the whites on the same bank, same voltage do not.
At First i thought it was because i reversed them, as i toasted a LS by reversing it one time for miliseconds, now i am positive that it isnt from reversing them, they just dont cope with thier own voltage specs.

What happens to them is they start to Flicker after time being on, blinkey blinkey blinkey, makes a cute xmas light but it keeps being very frustrating for this project

Do others have this problem, Like do any of the turquoise/cyan blue/green type of flashlights or minilights ever do this ?

right now thier configuration is like this electrically
whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| cyan
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|-O-|-O-|-O-| whites
|___11.50v__|

there are 2 banks of these
the leds are the ones they sell at TheLEDLIGHT

specs cyan http://www.theledlight.com/led-specs6.html

specs white http://www.theledlight.com/led-specs5.html

same voltage ? same current?
any suggestions ?
this even happens when they are driven even lower voltages but peaked at the 11.50v at one time or another

this is begginin to

me off

even though the whites were bought from WHITELEDS.net for way less money

 

[**DONOTDELETE**]

I should add that the cyans were put in in small quantity to get the hole in the whites spectrum filled, i really dont even think the camera notices, it was a matter of Specs of the specturm, very small quantities of the cyan and red were added back to the spectrum to achieve actual white light because the blue leds with yellow phosphors are missing that in thier spectrum.

i use to have a LIFE before i started doing this stuff

 

[**DONOTDELETE**]

AND
the exact same thing happened to me when i put 4 cyans with 4 whites for making one of them radioshack 4AA battery holder flashlights, i used 3 nimhy batteries and blinkey blinkey blinkey, that one is in the trash now.

 

[jeff1500]

Hmmmmmmm,

1. Blue-greens blink when mixed with white.
2. Do blue-greens blink when run by themselves?
3. What is measured voltage drop for blues when running by themselves?
4. What is measured voltage drop for whites when running by themselves?
5. Are there any resistors in your circuit?

I wonder if the current is oscillating because the voltage drops in each parallel path are similar but not equal? On a high peak the blue-greens light. When it's low they go off.

I wonder if an led without a resistor in series is unstable? Maybe put a resistor or two in there someplace. dV = I * R

I don't know, maybe something that only drops the voltage a little tiny bit in series with the led strings.

Leds don't have any resistance factor. Resistance helps stabilize parallel paths.

 

[**DONOTDELETE**]

1. Blue-greens blink when mixed with white.

not nessisarily BECAUSE of being mixed with whites, they just are in this case

2. Do blue-greens blink when run by themselves?

after they "break" they blink again when applying thir normal voltages

3. What is measured voltage drop for blues when running by themselves?

When i had made a flashlight using the cyans with 3mimhy batteries the voltages would have been up to 3.8v they fried in 20 seconds

4. What is measured voltage drop for whites when running by themselves?

each bank gets as high as 3.85 most banks sit at below 3.7, but i have driven the whites as high as 3.9 with no problem

5. Are there any resistors in your circuit?

the wires offer SOME resistance,then it routes through a wire wound potentiometer and the power originally comes from a regulated dc dc 12v power source that provides max 11.80v

I wonder if the current is oscillating because the voltage drops in each parallel path are similar but not equal? On a high peak the blue-greens light. When it's low they go off.

they blink after they are toast so to speak, after replacing it it works for a while

I wonder if an led without a resistor in series is unstable? Maybe put a resistor or two in there someplace. dV = I * R

the resistance NOW is set for the whites, and works for the whites, to add resistance to just the cyan would be to remove them from the banks, which is probably what i have to do.

I don't know, maybe something that only drops the voltage a little tiny bit in series with the led strings.

Leds don't have any resistance factor. Resistance helps stabilize parallel paths

ahhhhh, so it would be normal/better to have each led to have its own resistance, what a royal pain in the . . .
that would indeed make more correct the current/voltage going to each item in the banks
but then reduces the capacity to keep minimum resistance, minimum parts, and to layer in series and parellel sets for overly bright lights, and it keeps me from being able to dim it as a set.

 

[jeff1500]

2. Do blue-greens blink when run by themselves?

after they "break" they blink again when applying their normal voltages

------
Can you measure current? Sounds like something is getting fried.

I have a red led that wants just about 2.4 volts according to the package it came in, so I just hooked it up directly to two nimh batteries. I figured it would draw about the 20 mA it was rated at. Eventually I put a meter on it and found that it was drawing something like 65 mA. Then I put in a 10 ohm resistor and the current dropped down close to 20 mA.

 

[**DONOTDELETE**]

yup i think that is the issue. the current hitting the cyan vrses what goes to the whites in the exact same condition
ALMOST as if the voltage for this cyan falls between the voltages for the red and the voltages for the blues and whites.

is there a different metals/Material used in CYAN/turquoise than there is in white?
and has anyone else experienced a current diferential at the same voltage for them vrses the whites, in general, say from any company? or say even in the luxeous ?

from the talk here the curent is more important than the voltage, but when thinking simplistically what do you all think the voltage for the cyan/turquoise really should be for a nice burn without a burnout?

 

[jeff1500]

I made a variable voltage supply with a max757 chip and a potentiometer like this:
http://edusite10.tripod.com/led3/dimmer.html

I'm using it to replace R1 and R2 in this circuit:
http://edusite10.tripod.com/led3/max757.html

I'm seeing each led wanting to conduct 20 mA at a slightly different voltage. So I just adjust the voltage until I get the current I want.

If you want to use batteries and resistors, you can adjust the voltage by trying different resistors until you get the current you want.

Voltage pushes the current through the led. More voltage means more current flow. I think that when an led has a voltage rating of 3.6 to 4.0 volts, that means that at some voltage within that range, the current will be 20 mA.

Power is volts x amps. Too much of that is what blows them up.

 

[MrAl]

Hi there video guy,

You need to understand something very
basic to LED's. LED's should be driven
by a current source, not a voltage source.
Driving LED's with voltage is bound to
cause problems sooner or later.
Since voltage sources are much more common
to get a hold of, the simplest way to
drive the led's is with series resistors.
This is even more important for different colored led's. Voltage sources typically
have very very low series impedance so
when the temperature of the led changes,
it's current changes drastically when
driven from the voltage source alone.
Stick a few resistors in series with each
series string of led's and you'll see
the difference right away. You should also
measure the current through each string just
to make sure you are operating within spec.
Of course if an led is already blown out
it wont help :-(

Good luck with it,
Al