led lamp assembly help

[12vdc]

hey all
I acquired a lamp assembly, 42" long, extruded aluminum case approx 2"x2"


it contains 6 boards in series, 12 vdc, each board has 2 rows of 18 LEDs (white)

each group consists of 6 leds in series with a 75 ohm resistor between led 3 (-) and led 4 (+)

the leds do not glow, yet the resistors get warm (120 F)

power is dell brick rated at 8 amp

I tried 5v on a single led and...... it died glow then yellow fade to burnt.



it is a simple set up yet nothing works, dead / smoked resistors (smt) ?

 

[Harald Kapp]

I tried 5v on a single led and...... it died glow then yellow fade to burnt.

This one is defect and the whole string will light up no longer. Never use an LED on a power source without current limiting (resistor)!

it contains 6 boards in series, 12 vdc, each board has 2 rows of 18 LEDs (white)
each group consists of 6 leds in series with a 75 ohm resistor between led 3 (-) and led 4 (+)

We'd need to know how the LEDs are the groups are interconnected.

Do you have part numbers, a schematic diagram etc?

 

[Colin Mitchell]

There are 3 LEDs in series NOT 6.

 

[Harald Kapp]

There are 3 LEDs in series NOT 6.

How do you know?

 

[Martaine2005]

6 leds in series with a 75 ohm resistor between led 3 (-) and led 4 (+)

@Harold Kapp, Perhaps Colin was refering to this..3 leds..resistor..3 leds
Martin.

 

[12vdc]

here is my cheesey attempt at a schematic.
6 diodes in series + 75 ohm resistor.



(+) ------o---o---o---^v^v---o---o---o----- (-)

o = diode
^v^v = 75 ohm resistor (SM package marked "750")

 

[12vdc]

2 pics, if they show up, top rail is 12V + the bottom is 12v - whick connects to each adjacent board

I made an error in my initial post, " it contains 6 boards in series, 12 vdc, each board has 2 rows of 18 LEDs (white)"
should have read :
6 boards are in parallel, (something like the LED ribbon strips that can be cut into small modules)

 

[12vdc]

Sorry to post like this in pieces, my connection "times out" randomly and I lose all I have written

I will try to explain better.
There are 5 lamps, daisy chain. I tried connecting power to each individual lamp assembly and chained. Each lamp has 216 LEDs. the chances that all 1080 LEDs are dead is slim (?).

I tried one LED with "AAA" battery 1.5v, with 2 (3.0v) and 3 (4.5v) and the LED would not "emmit". I then tried a 5V ATX PSU and the LED slowly glowed orange and then died.

I "googled" (another somewhat useless piece of garbage) LEDs and could not find any images of my device.(my fault maybe incorrect terminology?)

It is a clear lens approx 3/8" diameter base, 3/8" high, a small bright (chrome?) 1/8" diameter cup sits in the center with the chip/die inside.
Is this Cree LED?

This was a functioning "string" of the 5 lamp assemblies(powered by the dell (D220P brick), obviously tossed because they no longer function.

The PSU reads 11.87V when connected to one or all units. (its it possible the PSU could show ~12v yet supply inadequate current?) I also tried another "good" 12v supply to a single board & a complete lamp strip of the 6 boards with nno positive results.

 

[Martaine2005]

Has the lamp got a seperate pcb? Apart from the led strips.
Is there a model number for the lamp?
Sure you are using 12v from the atx including polarity?

 

[Martaine2005]

Think we just posted at the same time....

If they worked and now they don't, something has blown.
As they are in series, if one led blows the others wont work.
Try your 3v on individul leds on the boards.

 

[12vdc]

I messed this up.
I'll try again.

There are no decals, part numbers or any writing on the lamp housings or boards.
To simplify this, I'll tray rephrasing.
I have 6 white LEDs and a 75 ohm resistor in series that is fed by 12Vdc. I have 180 of these units, none work. Will post better pics when I get hispeed interent

 

[KJ6EAD]

Your pictures aren't clear enough to be sure but with one 75Ω resistor per six LEDs, the supply would need to be ~24V to provide a useful current level.

The situation you described where an LED glowed and then died is typical of a reverse polarity overcurrent failure event.

Do you have or can you get a multimeter?

 

[12vdc]

Yes many a meter.
I am usnsure of what voltage these LEDs operate at, knowing that would be the best start. . I know there are so many new developments daily so thses could be anywhere from 3.5v to 5 volts? Hmm....
(VE3 part of the world)

 

[KJ6EAD]

I'd suggest doing two things: first, draw out the circuit traces for a segment of the PCB to determine the actual topology and second, use the diode test feature on a meter to verify the correct polarity of one of the LEDs within that topology to extrapolate the correct polarity for the circuit supply nodes.

The individual voltage drop for any one LED of that color and type is going to be ~3.6V.

 

[12vdc]

I took some webcam pics, I did find an obscured (+) on the front of the board and the flat side of the LED case. As i have traced back from the power supply connections the polarity was correct.
The layout is as advertised in the cheesey text drawing above.
Now 3.5 x 6 = 21 v plus the 75 ohms resistor makes sense this could require 24V?

 

[KJ6EAD]

Even if the strips are designed for 24V operation, they should light up enough on 12V to determine condition.

 

[12vdc]

I bought a voltage/current regulating module off ebay. going to try 3.0v @ 150 ma for a single LED.. Sound safe?

 

[Colin Mitchell]

Include the 75 ohm resistor and gradually increase the voltage

 

[12vdc]

Is the purpose of the resistor not to regulate the current?
If i can do so with the regulating module can i not leave out the resistor?

 

[Colin Mitchell]

How can you regulate the current when you have absolutely no idea of the characteristic voltage of the LED at the current-level?
The resistor KNOWS. It has built-in intelligence and knows exactly what current to deliver at a range of supply voltages. Don't get the resistor too hot or it will lose this intelligence and you will finish up with a burnt out LED.