0

So I had an idea for an LED lamp. I've got some old flashlights that have older model CREE LEDs.

My idea is that I'd put the LEDs in parallel, alternating polarity so that for each phase of AC, half of the LEDs will be conducting.

The array of LEDs would be in series with a resistor and a fuse to drop voltage from 110 VRMS from the outlet to 3VRMS over the LEDs and prevent the LEDs from breaking if something catastrophic happens (and to compensate for my own failings.)

The LEDs have a 5V reverse voltage and a 3.3V forward voltage at 3A at 80 deg C, so I figure if I limit to 3 VRMS I should't hit the reverse voltage and make badness happen.

With this in mind ... I am only passingly familiar with anything more advanced than kirchov's and ohm's laws. I can't apply ohm's law to the diodes, so what do I need to know to figure out the resistor value for n parallel diode pairs (Or maybe n sets of 5 in series and put those in parallel)? or is it just a bad idea?

Daniel B.
  • 444
  • 1
  • 6
  • 18
  • If you are only passingly familiar, throw a transformer in the mix. Transformerless designs are inherently dangerous. – jippie Oct 18 '13 at 19:16
  • I assume you mean to use a transformer to step the voltage down to the 3.3VRMS, but I'd still need a resistor to limit current wouldn't I? Why is it dangerous? – Daniel B. Oct 18 '13 at 19:18
  • Not necessarily step the voltage down to 3.3V, yes you'd still need a resistor. But you want the transformer for galvanic separation from the mains. Mains power can potentially kill people. An added safety is that you have to deal with a lower voltage, which is safer too. You want a voltage below 40V, that is commonly regarded as a safe voltage. – jippie Oct 18 '13 at 19:22
  • Your LED's are really rated for 3A? – jippie Oct 18 '13 at 19:25
  • 1
    @jippie: The XM-L's, yes. – Ignacio Vazquez-Abrams Oct 18 '13 at 19:28
  • http://www.cree.com/led-components-and-modules/products/xlamp/discrete-directional/xlamp-xml2 Yes :) There are graphs around for their efficiency somewhere, but since this is just a proof of concept sort of thing, I don't really care much so I'd probably want 2A. – Daniel B. Oct 18 '13 at 19:28
  • The old model from these flashlights might be less, I need to dig up a datasheet for them. – Daniel B. Oct 18 '13 at 19:28
  • Just so I understand galvanic separation, the current being passed through the secondary coil of the transformer will be only what's required by the circtuit, rather than the full voltage going through the mains line, right? Hence the safety? – Daniel B. Oct 18 '13 at 19:29
  • No, the point is that there is no direct connection between the two sides of the transformer, only the magnetic flux transmitting the power. So the voltage of either output wire is not relative to earth, but only relative to the other output wire. Even if it was a 1:1 transformer, the voltage at the output is no longer with respect to earth, so you won't get a shock by conducting it to earth - you get such 1:1 isolation transformers for working on/with mains equipment where you might be exposed to a shock risk. – Pete Kirkham Oct 18 '13 at 19:42
  • Thanks I will definitely implement an xformer :) Any idea on what principles re: interaction between diodes/resistors/voltage/ac/etc I need in order to make the circuit work? i.e. if the forward voltage of a diode is 3V, is the voltage dropped across three diodes 9v? If that's the case if I'm putting in 110VAC, do I try to make the resistor drop 101V? – Daniel B. Oct 18 '13 at 19:53
  • I recommend a current-mode controller to get the most out of your LED lamps. – HL-SDK Oct 18 '13 at 20:07
  • Someone else has done the "vast stack of LEDs in series with the mains": http://electronics.stackexchange.com/questions/47212/is-this-led-fixture-safe-for-use-with-its-current-fabrication – pjc50 Oct 19 '13 at 09:13
  • My question was about leds in parallel and that question had almost no information pertinent to my question. – Daniel B. Oct 19 '13 at 20:59

1 Answers1

2

First, using a resistor to drop from 120V to 3V is very, VERY inefficient. You will be burning off about 90% of your energy in heat. You need a transformer here, at the very least. Just doing the math: If you need 3A/3.3V for light, then 3A/120V is 360 Watts of power, of which you're extracting 9.9W for the actual diodes, and thus burn about 350W in heat in the resistors.

Second, LEDs in parallel is unreliable, because the actual voltage drop of the LEDs will vary a little bit because of process variance, and thus some of the LEDs will draw a lot more current than others. This will result in the lower-voltage LEDs burning out, and the higher-voltage LEDs being dim. To fix this, you need one current limiting resistor per LED, rather than a single one for the gang.

Finally, what's the point of wiring half of them each way? Why not use a bridge rectifier?

Jon Watte
  • 5,650
  • 26
  • 36
  • "Why not use a bridge rectifier?" - Indeed! And a switching mode current regulator and all diodes in series? Because the asker never heard of such things... – johnfound Oct 18 '13 at 21:03
  • Id' actually looked at current drivers beforehand (I'm assuming that's a switching mode current regulator?). My original plan was indeed to have the LEDs in series but then I decided I wanted to try running it from AC and that's where I went to this conclusion. I was thinking I could do this for less monies, but it's looking like I was wrong, once I buy transformers and rectifiers and things, I might as well just buy a manufactured current supply. – Daniel B. Oct 19 '13 at 01:29
  • As I said in the question, I wasn't sure it was a good idea. My fears were founded, so here's your accept and thank you for your help :) – Daniel B. Oct 19 '13 at 01:30