[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/diy/ - Do It Yourself

View post   
File: 129 KB, 1000x660, psu.jpg [View same] [iqdb] [saucenao] [google]
1267565 No.1267565 [Reply] [Original]

Hey /diy/,

I'm planning on building a 8^3 RGB LED cube build using discrete WS2811 LEDs. Using 60mA/LED draw at max brightness, I've calculated that the whole cube will require 30.72A in the worst case.

Looking at 40A supplies (pic related), the reviews often note that the supplies output a fraction of the rated current on each output rail. Would it be safe to join all the positive and negative connections in order to create a single rail capable of outputting the full rated current? Or is this even necessary if I'm only using a single output rail on the supply?

>> No.1267569
File: 1.61 MB, 2322x4128, 20170523_201559.jpg [View same] [iqdb] [saucenao] [google]
1267569

Seeing that you can splice in power anywhere with WS2812s I imagine you can do the same with those, and yes I have basically joined the positive and negative connections on my 20A power supply for my sign.

>> No.1267572

>>1267569
Okay, thanks. Since I'm using discrete LEDs I'm planning on just bridging all of the power and ground pins together anyways (both for power distribution and structural stability), so if it's safe to bridge the PSU outputs like you've suggested I think I'll be good just supplying power at one point.

>> No.1267692

>>1267569
YAAAAAAS.

Love this shit

>> No.1267706

>>1267565
>8^3 * 3 LED
>60 mA / RGB LED
>30.72 A
what universe are you trying to illuminate?

>> No.1267713

>>1267706
30.72A*5V=153.6W
Not to mention that thile WSxxxx are bright, they are still nowhere near as bright as those 100W cobs, because thermal limits.

>>1267565
>connecting 2 rails together
If it truly is a dual-rail power supply, this is a bad idea. You'd have 2 individual voltage feedback circuits, meaning that because of the manufacturing tolerances the one with higher voltage will cause the one with the lower voltage to "crowbar", depending on the built-in protections, it can cause anything from pulsing the output to a spectacular self-destruct.

But if it's just a single point feedback (being of chink origin, it probably is), you can join the rails no problem.

>> No.1267761

>>1267565
I'm pretty sure on the inside of those PSUs all the ground and Vcc terminals are already connected directly together. They give you more than one pair of terminals for convenience not because there's multiple independent power supplies in the box.

>> No.1267766

>>1267761

with a multimeter and 10 seconds of time, this can be confirmed or dis-proven.

>> No.1267769

>>1267766
I don't have one to test at the moment.

>> No.1267860

>>1267713
>You'd have 2 individual voltage feedback circuits,
I absolutely have no evidence or experience in smps design but I believe this is not correct, rails might have separate transformer windings, fuses, current limit etc but I don't believe the closed loop switching feedback operates individually on all rails, one particular voltage rail would be used to set the switching parameters, the other voltages and rails are set ratiometrically by windings.
To have individual closed loop control of multiple voltages but also each rail? Is that really done?

>> No.1267986
File: 1.00 MB, 2322x4128, 20170919_200349.jpg [View same] [iqdb] [saucenao] [google]
1267986

>>1267692

Unfortunately my phone doesn't show the purples and blues right.