[Solved] Help powering WS2812b, have I fried it ?

--Edit : Since the first sketch attached to that first post wasn't that clear, I posted another one 3 posts below, here it is, or scroll down a bit https://hostpic.xyz/files/15614665461939651637.png--

Hi,
I'm fairly new to electronics, but I've worked with arduino before.
I'm starting an ambilight project with a WS2812b strip, and got stuck at the first step... powering it.
I'm facing one of the most obvious things for most people but I maybe too scared to act or do anything.

Any help would be really appreciated, thanks in advance !

wellshir.png

So you start by connecting a regulated 5 V power supply to the power connections, and you connect the 5 V, ground and data wires to your Arduino Nano (the 5 V will then power the Nano), and you run some test code.

What is the problem?

It goes without saying that if you connect it wrongly, you will destroy at least one part.

Hi !
Thanks for your reply.

I've tried to learn a bit more about all this, and this is the setup I've followed :

I'm using the Adalight basic code, not modified in any way except for the number of leds and pin number.
Uploading it to the nano, then running Prismatik, and going to go through the setup. The LEDs never turns on.

I will get my hands on a multimeter tomorrow, I should have done this earlier.
I will verify the output of my dc adapter with this.

But apart from this, I don't know what could have gone wrong in this scenario. Might just be the leds all along

Thanks in advance

I forgot to add that my first scheme was not clear at all.
There it is, a bit more explained

The 3 pin connector was not connected and I'm here wondering if maybe there were a problem in the connector itself, creating a short circuit ?

But apart from this, I don't know what could have gone wrong in this scenario

Well apart from your wiring not actually matching up to your diagram, and your power supply not being able to produce 7.2A, and your breadboard tracks trying to carrying over 1A, and you not having any capacitor across the strip and not having a resistor in series with the data signal, then I cant see anything that could have gone wrong either.

The 3 pin connector was not connected and I'm here wondering if maybe there were a problem in the connector itself, creating a short circuit ?

How can wires not connected to anything cause a short circuit?

Hey,

So my first diagram is the first thing I've tried to check if the LED strip was working. The second one is a second setup I'm using right now after failing miserably on the first attempt... I now realise that I wanted to power a bit less than 2 meters of LED and that's why I got a 6A 5V adapter. I guess that amperage difference would bright up leds yellow when too far away (as I've read across forums ?)

But apart from this, I don't know what could have gone wrong in this scenario

Yeah honestly that was not very clever on my part :expressionless:

your breadboard tracks trying to carrying over 1A

Does it means that breadboard tracks has an Amperage limit ? I don't know anything about this

Also, a capacitor would help having a clean signal from a source not powerful enough ? I omitted it thinking my amperage was ok, but it wasn't.
Same thing for the resistor, I did not consider putting one on it because I've followed something that someone did without too much thinking. That's my biggest mistake here.

Finally, about the connector and short circuit, I imagined that maybe wires were touching inside but that's like, nearly impossible I guess.

So now I'm lost again. Can I ommit the capacitor if I want to power 100 LEDs (needing 6 amp exactly) ?

I don't understand how a resistor could help, but I've seen sketches where they were used (470 ohms). I don't want to blindly follow other sketches, but I feel like it's not the forum's job to just give all the answers, etc.

So I should either learn everything, and maybe try later. But I look forward for advices, if there's any

Can I ommit the capacitor if I want to power 100 LEDs (needing 6 amp exactly) ?

No.
You need to understand what the capacitor does. It smooths the voltage in the face of rapidly changing current demand. With the sort of current you have you need one. The power supply will not cope with that much ripple.

The resistor is there for two reasons, to try and match the impedance of the line so there are no reflections on it, and also to limit the current into the first LED in the event that the supply voltage drops below the drive signal, which can happen especially when you have no capacitor.

Does it means that breadboard tracks has an Amperage limit ? I don't know anything about this

Yes you should not use it for heavy current, it is not designed for it.

I guess that amperage difference would bright up leds yellow when too far away (as I've read across forums ?)

No you are miss understanding that as well. You need to be able to supply sufficient current for the LEDs, even if your power supply is capable of this then you need to take precautions so that you don’t get a voltage drop due to the thin PCB tracks having a highish resistance which causes a voltage drop that becomes significant at high currents. Precautions included at least connecting the power and ground at both ends of the strip and perhaps taking the power and ground to some point mid strip.

Hi,

I've took notes of everything, thanks again.
I've decided to drown myself in books and start to learn. I'll be attempting my project in the next weeks / months when I have the knowledge required

Thanks again for the help and have a good day !

Grumpy_Mike:
You need to understand what the capacitor does. It smooths the voltage in the face of rapidly changing current demand. With the sort of current you have you need one. The power supply will not cope with that much ripple.

Actually, the power supply will cope with that much ripple. The problem will be that the power supply is on one end of a length of cable, and the LEDs are on the other end! The cable in the middle introduces series resistance and inductance which allows all sorts of funny things (transients) to happen. So the capacitor is needed at the very start of the strip where you are feeding the data.

Mike has explained the resistor - it also needs to be at the very start of the strip for much the same reason.

Now your diagram in #3 is a bit silly. :astonished: If the data connector is not connected, then nothing is going to happen, is it? Except that the first LED or two may interpret your switch-on transient as random data and light up. If this is a trouble when you actually have an Arduino connected, just put a 10k resistor between the Arduino pin driving the display, and ground.

The problem with your diagram in #2, is that it does not emphasise that the wires between units must always run as pairs, power together with ground, and data together with ground. As I explained originally, the wiring of the strip as you get it, is for a reason. You feed power to the two power wires that terminate at the beginning of the strip, and you feed data in at the 3-pin connector (you should have received the matching connector with the strip - if not, you have been ripped off!). But you use the power wire at the 3-pin connector to bring 5 V power back to the Arduino, you do not power it separately so you do not create "loops" in the wiring.

However there is the matter of getting adequate power to the full length of the strip. If you only ever activate a few LEDs at a time, you will not notice a problem, but if you propose to operate most or all of it at full intensity, then it will correspondingly draw the current calculated at 60 mA per LED. This would result in yellowing of the distant parts of the strip (and may even burn out the power foils) so you need to run a power cable (at least 1 mm2 conductors, both 5 V and ground) along with the strip and tap it in to the strip at least every 50 LEDs (3 Amps worth) including both ends.

Well that's a lot of info. As I said, I'll be on my way and learn everything I can, but thanks for your insight !

Actually, the power supply will cope with that much ripple.

That is a point I would disagree on.

I have made many system, both home built and commercial using commercial power supplies and all I have seen need extra supply decoupling when the load gets choppy, like it does here.

@Rarduinoo - I have developed a saying from my time in industry.

"Given n engineers, there will always be AT LEAST n+1 firmly held opinions."

This game might look like a science but there is a bit of art in it as well. :slight_smile: