[ 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

Search:


View post   

>> No.1129297 [View]
File: 71 KB, 1492x620, Screen Shot 2017-02-11 at 10.47.54 PM.png [View same] [iqdb] [saucenao] [google]
1129297

>>1129220
If you want constant current at a constant voltage, your circuit should only draw that constant current when provided with the correct voltage(s). Except in the case of non-Ohmic resistors, which LEDs happen to be. They'll still increase their current when you increase the voltage across them, but the current increase will be much greater than an Ohmic resistor would have, as they have a hyperbolic curve rather than a liner one. Long story short, you need something to limit current, or else have a very precise voltage source that can account for temperature changes in the LEDs. It's not that big a deal if you don't do this, I've got a 3V coin cell powering a 3V LED and it works fine, but in extreme conditions your LEDs might not last as long. If your LED chip (assuming you don't own a massive 32V LED) already has resistors in there, you're in luck. If not, have fun searching for an online tool to let you calculate the correct resistor to put in series with your LED bank. The correct voltage drop and current drain across an LED is pretty much determined solely by its colour, so it shouldn't be too hard to find the resistor value, multiply it by the amount of LEDs in series (probably), see what power value it needs to be through P = I^2*R = 0.995^2*R, and buy it.

In contrast, the fan is fuck easy to run, just put the right voltage across it and it will be fine. You can experiment with putting potentiometers (variable resistors) going to both the fan and the LED bank to dim them assuming they can handle the power (max power through a pot is when its resistance is equal to that of what it's in series with). Chances are, it won't work with the LED bank, but they're pretty iffy with pots anyway, what with the nonlinear effective resistance curve and all that.

Cont.

Navigation
View posts[+24][+48][+96]