Nicols Par led 363 RGB - Green leds illuminating very low

[matb]

Hello everyone, I'm trying to repair a PAR LED 363 RGB from Nicols. https://drive.google.com/file/d/1o19Ej_4FivIdIYNoKH4Wo5Y-wVEvFpW5/view?usp=sharing is the reference of the product.

The folder of images is https://drive.google.com/drive/folders/1nltcPlk61XfHIye755DryKLG1mFIceP5?usp=sharing so you can go around the images.

The Red and Blue light work fine, the Green however doesn't light up. Seems like it's what gets killed on this product. The previous owner changed 2 Green leds and apparently it fixed the problem.

Now I'm trying to find a Red, Green and Blue leds to replace the faulty leds. Just to be clear only the Green leds fail it never occured on the Red and Blue.

A brief overview : the main (+230V) goes in to the "primary PCB" (Brown and Blue cable) then it is converted to DC and goes to the "secondary PCB" which then drives the leds. They are 2 circuits for the leds one called "LED1" and the other "LED2".
https://drive.google.com/file/d/1nxdpjEHbPyT_10bu5ObjyQmM9oklXE60/view?usp=sharing is the general image.

https://drive.google.com/file/d/1oR_6Ki-4itasBrUN_tmfrvRsDDdqnUL6/view?usp=sharing and https://drive.google.com/file/d/1oPqz-YnKvUVz-12K_h9qOYUcOEiho0V3/view?usp=sharing are the "primary PCB" images.

https://drive.google.com/file/d/1ocT9RStZhf-UjyR-lvNjQBZJSI65Lat2/view?usp=sharing are the "Secondary / LED driver PCB" images. It doesn't seem damaged that why I'm guessing it is a dead Green led.

https://drive.google.com/file/d/1oF0ZU8h5Ik7TlkGAPQdXNEF8mm225uPV/view?usp=sharing shows that some of the Green Leds light up. G3 and G8 have already been changed so there is that ...

So 2 questions :

• Do you think the G7 led is bad and should be changed ?
• Can you link a replacement for each color ? (in the case another color finally fails)

[mariush]

In case it helps, the LED drivers seem to be PAM2681 , a 1A max step-down led driver: https://www.diodes.com/assets/Datasheets/PAM2861.pdf

It looks like the max current resistor is 0.14 ohm which would mean the led driver at 100% brightness would allow a current of I = 0.1 / Rs = 0.1 / 0.14=  ~0.71A  (datasheet says 0.13 ohm = 0.76A on page

Looks like there's 6 led drivers, so I can only assume it's 36 / 6 drivers = 6 leds per driver ....

I don't see the output voltage of the power supply, but I can only assume it's 20-24v or something like that, because the led drivers are 40v max and there's only 6 leds ... 6 x 3v..3.2v is around 18v, making 24v more realistic than 6 x 6v forward voltage leds, because that would mean at least 38v or so input voltage.

You can also see figures 2 and 5 on page 5 in the datasheet, where they use 0.15 ohm resistor and 47uH inductor, and  which shows 24v as good voltage for 6 leds.


With around 0.7A current, you'd be looking at 3.2v x 0.7A = 2.25w per led, not 3A ...but who knows.

As for leds themselves ... I'd look at digikey, mouser, newark, other big sites ... you have to figure out that package shape, you have the beam angle 25 so you can filter by that

On Digikey I see 2-SMD with gull wing as a potential package type, but there's only some 700mA max leds with wide lens (140 degree), for example here's a blue one : https://www.digikey.com/en/products/detail/broadcom-limited/ASMT-AB31-NMP00/2347268

As you can see, 3.2v forward voltage, max 700mA

Here's some red leds, but not in stock and also wide angle : https://www.digikey.com/en/products/detail/seoul-semiconductor-inc/R42180-01/2504442

[matb]

Thanks, would it be of any use if I share the voltages of Green, Blue and Red ?
Red leds light up if I use my multimeter in continuity mode but not the other colors.

Are the led driven from the low side ? I tried to connect my oscilloscope on the ground of one led but that light up the array so I'm guessing it is low side driven.

[mariush]

The forward voltage of the green and blue leds is probably too high, and your meter can't measure that high (or push constant current at high enough voltage to open the led).
Red is usually much lower forward voltage, in the 1.7v..2v , the others are around 2.8v...3.2v

[matb]

Ok thank you, can I apply maybe 3V on one LED and see if it illuminate or is that a no-no. Would in circuit / out of circuit make a difference ?
The resistors of the LEDs are on the LED board or on the power supply board ?

[mariush]

You can damage the led if you apply 3v directly to the led, because your power supply would have a very low resistance so the individual led would consume as much current as the power supply is willing to give, and then blow itself up.

You can always use a CR2032 battery because the battery has enough internal resistance which will act as a current limiting, so at most the led would probably consume 30-50mA of current, and not hurt itself. 

A simple solution would be to simply add a 10 ohm or higher resistor in series with the supply you have (ex 5v from USB charger or 2 AA / AAA batteries in series). That resistor will limit the current and not hurt the LED.

Each led driver has the sense resistor of 0.14 ohm (R0140 on the resistor on the circuit board). The LED driver chip monitors the voltage drop across this resistor and constantly adjusts output voltage so that the voltage is a specific value, limiting the current to a desired amount as I explained above (formula is 0.1 / resistor value  so 0.1 / 0.14 works out to around  0.714 A or 714mA

[matb]

So a power supply of 5V and a series resistor would do it ? For every color of LEDs ?
I will try that I have 1k laying around I don't need full brightness so I guess I will adapt depending on the luminosity and see if one led doesn't light up.
Thank you.

[james_s]

Yes 1k should be very safe and it ought to be enough to make just about any modern LED light up.

[matb]

Ok so I tested with +5V 1k ohm resistor and all the green LEDs light up.
I'm thinking of a bad solder, is that frequent on aluminium based PCBs ? What do you guys think ?

[themadhippy]

I'm thinking of a bad solder, is that frequent on aluminium based PCBs ? What do you guys think

cant comment on these particular pars,or aluminium pcbs,but dodgee solder was certainly a feature in the early days of led cans,especially if they lived there life on the road.

[m3vuv]

You mention bad solder,tbh ive yet to see a good joint made with lead free crap!.

[matb]

So should I try to melt again the solder joints, I'm thinking with a hot air gun and a soldering iron. (pre-heating the board then going with a soldering iron and pushing on them)