DIY Chatroom Home Improvement Forum

DIY Chatroom Home Improvement Forum (http://www.diychatroom.com/)
-   Electrical (http://www.diychatroom.com/f18/)
-   -   80 leds for led light. can i run to 230v mains (http://www.diychatroom.com/f18/80-leds-led-light-can-i-run-230v-mains-14743/)

donnol1983 12-25-2007 08:14 PM

80 leds for led light. can i run to 230v mains
 
Hi, Im trying to make my own lighting for a small room in my house..
I have 100 led's 18,000mcd, 3.o-3.8 vf,10mm.

Im not sure on how I should calcualte what resistor i need.
I was thinking as they are 3v forwards current, If I use 80 in sequence that would allow me to run it strain off mains current as 3*80 = 240.

I have the leds all mounted and wired on a board with no plug, in seequence but just wanted some help before I plug it in as i dont want to blow them all out..

Can anyone help. thanks

robertpri 12-25-2007 09:40 PM

I have good luck with this site calculating LED resistors.
Scroll down to the calculator.

robertpri 12-25-2007 09:41 PM

oops--forgot the link

http://www.bcae1.com/led.htm

davefoc 12-25-2007 10:12 PM

Hi donnol1983,
The situation is not quite as straightforward as you seem to suspect.

A few things that may not be aware of:
1. The voltage from a standard 220volts nominal outlet is about 230 volts RMS (root mean square). The RMS voltage is meant to approximate the amount of power that would be supplied to a resistor if the continuous DC voltage out of the circuit was 230 volts. The peak positive voltage from a nominal 220 volt RMS AC supply is about 1.4 time 115 volts (half of 230) or 161 volts measured to ground. The peak voltage difference between the two opposite hot legs of a typical US home power circuit is about 1.4 times 230 volts or about 320 volts.

2. The circuit that you propose to use to light your LED's may be dangerous in that dangerous voltages will exist throughout the chain of LED's.

3. Assuming you decide to go forward at this time the voltage drop across your hundred LED's will be something in the range of 300 to 380 volts depending on temperature, the nature of the LED's and the current through them. Unfortunately this voltage range doesn't work well with the voltage from the wall supply and it would be impossible to pick a resistor to provide the appropriate current in all situations. The simple solution would be to reduce the number of LED's which will make it so that a resistor can control the current better. As an example use 70 LED's. These will have a forward voltage drop of something in the range of 250 volts. So something in the range of seventy volts will need to be dropped across the resistor. Assuming that you are trying to get a current of 10 ma's the value can be calculated with ohm's law. R=100/.010 or 10K. (ETA: correction 70/.010 or 7K resistor should have been the equation I listed here, still a 10K resistor isn't a bad value to start with)

4. Current in your circuit will flow much less than half the time. Half the time the LED's will be reverse biased, so the current flow then will be almost zero. When the diodes are forward biased significant current will only flow when the voltage has risen above about 250 volts.

5. A conservative calculation for the power dissipated in the current limit resistor is .5 (period when diode is forward biased) times 10,000 ohms times (.010)^2 or about .5 watts. So a one watt resistor seems pretty safe.

6. The LED chain might not be as bright as you intended because of the limited time when significant forward current is flowing through them. You could make things better by adding a capacitor to extend the period when current flows or allowing more peak current to flow (reducing the size of the current limit resistor). There are issues with the capacitor so be careful if you go down that path. Increasing the peak current needs to be done carefully so as not to exceed the peak current rating of the LED and not to exceed the power rating of the LED.

7. I am an ex electrical engineer so I know something about what I described above but I did make mistakes in the past and I may have made some above so be careful.

8. You might start with a higher value resistor and work your way down to the value you decide to use permanently. You might also start with a much lower voltage from a wall transformer type supply and do some experiments like that or even use that kind of supply and drive your LED's in parallel. You might also just use a standard 115 volt outlet and have two parallel chains of LED's. Your circuit could have half the LED's facing one way and the other half facing the other way so that you were drawing current on the positive and negative phases. Nothing in this circuit should be all that hot so if you smell stuff or see smoke something is not right. Be careful and good luck.

- Dave

scorrpio 12-26-2007 01:24 PM

Don't you need DC voltage for the LEDs? Last I checked, AC didn't work.

davefoc 12-26-2007 01:56 PM

Quote:

Originally Posted by scorrpio (Post 83057)
Don't you need DC voltage for the LEDs? Last I checked, AC didn't work.

LED = light emitting Diode

The diode will glow when the AC voltage causes the appropriate forward voltage drop across the diode.

A problem with using AC is that the diode will glow brighter as the AC voltage rises and essentially won't glow at all until the voltage rises above the point that significant current flows through the diode. Of course the LED also won't glow when it is reverse biased.

AdamVocks 12-27-2007 05:28 PM

Wouldnt the OP be better off rectifing the AC for more efficient and higher light output? A simple full wave bridge and properly sized capacitor and resistor?

http://hyperphysics.phy-astr.gsu.edu...ic/rectbr.html

davefoc 12-27-2007 06:05 PM

The diodes themselves can be used as rectifiers so it might be reasonable to run this off of AC.

A capacitor could be added that would extend the time when current was flowing thereby increasing brightness without increasing peak current.

Whatever scheme is used that involves a current limiting resistor will have similar losses that will limit efficiency.

In looking around on the web a bit I saw a scheme that involved using a series capacitor to limit current. That would probably be more efficient but there are issues that need to be considered before a scheme like that is used.

donnol1983 12-27-2007 06:23 PM

Quote:

Originally Posted by donnol1983 (Post 82953)
Hi, Im trying to make my own lighting for a small room in my house..
I have 100 led's 18,000mcd, 3.o-3.8 vf,10mm.

Im not sure on how I should calcualte what resistor i need.
I was thinking as they are 3v forwards current, If I use 80 in sequence that would allow me to run it strain off mains current as 3*80 = 240.

I have the leds all mounted and wired on a board with no plug, in seequence but just wanted some help before I plug it in as i dont want to blow them all out..

Can anyone help. thanks


Okay, thanks for all the help.
I think Im going to try to use DC current if the same rules apply then I'll be using around 70led's I'll have to buy a resistor and power convertor but at least I didnt wire the Plug up yet so should have no problem fitting all the things I need in the circuit..
I'll post some pics when I have the light up and running, ma take a few weeks. thanks again guys.


All times are GMT -5. The time now is 12:01 PM.


Copyright 2003-2014 Escalate Media LP. All Rights Reserved