This should be able to drive 1200 WS2812B’s without any issue right?
I assume you’re looking at the 5V 200W model? If so, that’s 40A max (200W / 5V = 40A). 1,200 LEDs can pull 72A max (1,200 LEDs * 60mA each = 72,000mA = 72A) at 100% brightness solid white. You’ll need to limit the max brightness and/or power use in software. You’ll also likely need to inject power every few hundred LEDs.
That depends on how bright you run them and how well you handle power distribution to avoid voltage drop. 5V at 200W is 40A, which requires quite heavy wires. 1200 pixels at roughly 50mA each full brightness is 60A, so you couldn’t do that. But for most patterns it will be OK.
Very good feedback, i definitely dont plan to use all white i need some colorful patterns for christmas. So if i limit the power in FastLED that will limit the amount of current the LEDs will pull?
The pattern will only be set for about 300 LEDs thou the rest will branch out from it using the same data line.
@Jason_Coon out of interest have you ever measured WS2812’s drawing 60mA each? I’ve only ever seen about 50mA, although it has been a couple of years since I did any serious measurements. Of course it is good to have some headroom when doing these calculations!
Just did 2 quick tests at full 0xFFFFFF white on two different groups of WS2812 LEDs…
10 LEDs (of the type which require a shorter T0L pulse) draws 0.53 amps, or about 53 mA per LED.
64 LEDs (of the type which allows a longer T0L pulse) draws 2.15 amps, or about 33.6 mA per LED.
Also, with both types, if I leave them on at that brightess, the current seems to decrease somewhat as they heat up. The 10 LEDs went down to about 0.49A after several minutes.
Interesting findings. Thank you for sharing 
wrote a quick blog article… with photos of the tests
https://www.pjrc.com/how-much-current-do-ws2812-neopixel-leds-really-use/
Would 14 AWG wire be thick enough for this thing?
It depends on the length…
http://www.rapidtables.com/calc/wire/voltage-drop-calculator.htm
@PaulStoffregen interesting! Thanks for doing that.
WorldSemi’s die shrinks have been the cause of various incompatibilities in the past, with the well known timing “adjustments”, but also increased ESD susceptibility. Most manufacturers would renumber the part, add a suffix, or at least publish a PCN. Their handling of the first die shrink back in 2013 meant an urgent trip to China for me, and a total lack of faith in them as part of any supply chain.
So what happens if the LEDs pull more power than what this power supply can supply?
Can’t be sure with the cheap “no-name” brand, but the similar Meanwell NWS-150 or NES-350 go into hiccup mode or current limiting protection mode, with auto-recovery. So the voltage will drop until the current decreases… which will typically happen anyway when the voltage drops.
But be careful, unless you use fairly heavy gauge wiring you’re going to melt your wire insulation off if you get anywhere close to the maximum current rating.
I dont plan on it as not all the lights will be on at the same time and ill probably never use white (im still considering it) just dont want my tree suddenly bursting into flames in the middle of the night lol
A live tree in the yard by the way. And not a pine tree lol
You can run several “feeds” from the supply and put in-line fuses on each feed - e.g 10A each
I use the generic no-name ones from China, and they do the hiccup mode thing with auto-recovery.
I should mention that fuses will generate a voltage drop, as they are effectively a resistor that will burn out if stressed beyond the specification. And at 5V you don’t have a lot of tolerance for volatge drops. So you tend to need to overrate any fuses, just like you would go with heavier gauge wire to minimise voltage drop on longer runs. But the fuse should always be rated lower than the cabling can sustain.