Best way to wire LED's

[collectors]

Hi, i want to wire up some scenery lights & i see that i can wire led's in different ways. IE:
Power to 1 led with 1 resistor “390 ohm”. all in parallel. OR
Power to 1 resistor “100 ohm” with let's say 3 led’s wired in series
Is there a preference, or is it just best to put a resistor on each one.

Also, how do i know the wattage of each 5mm led if i just have a box of white leds.

Thanks

[Steve M]

Best practice is to wire them with a resistor for each LED but this calculator will help should you choose to vary from that and will indicate the ‘correct’ resistor to use.

http://led.linear1.org/led.wiz

[collectors]

Thanks Steve, i thought as much. Out of curiosity, if i did any in a series with just one resistor & one led out of 3 failed, would this stop the other 2 led’s working in this group? As this used to happen with the old xmas tree lights, but not now days, or is there somthing that bridges this problem.

[Brian]

LED resistor Ohm rating will depend on several factors.
A) Supply volts
B) LED Forward rated volts known as VF
C) LED Forward rated max current (Normally in milliamp) known as IF
D) Number of LEDs to be in the single circuit.

So for example, using a typical 12 volt regulated DC power supply. LED VF of 2.2v. LED IF of 20ma. One LED in circuit (Note the reference to a REGULATED power supply) 12-2.2=9.8/0.02 = 490R Use nearest higher value of resistor. In this case a 510R resistor would be the lowest value to use. However by using a higher Ohm value resistor the LED current can be reduced and its light output still remain excellent. A 1K0 (1000 Ohm) resistor would reduce the current drawn to around 10ma while a 2K2 (2200 Ohm) would reduce it to around 5ma. After 1000 Ohm the illumination level of the LED may become dimmer. Note this is based on using a 12 volt DC regulated power supply. "Regulated" means the output volts are always constant regardless of load from minimum current loading to the rated PSU maximum

There is a very simple to use LED resistor calculator web site here. Well worth book marking for future reference... http://led.linear1.org/led.wiz

The problem with series connected LEDs is that should one LED in that chain ever fail all others in that same chain will go out too!

You haven't mentioned what the supply volts is to be? 390R (390 Ohm) is fairly low unless your using 5 ish volts
Wattage is extremely low and not normally factored in. Watts is Amps x Volts

Edit to add..., Beaten to the LED calculator link by Steve
There is no linking of blown LEDs available that I know of.. Its actually quite dangerous where the old type filament lamps were used, as when one blows and shorts all others have an increased voltage applied to them by the amount the blown one was rated at. Two or three blow and all the rest are likely to pop their clogs too, due to excessive voltage on their feed!

Your 'Box of White LEDs' probably have a Forward Voltage of somewhere between 3 and 4 volts. Normally this is obtained from the LED data sheet for the appropriate LED. White LEDs normally have a higher rated forward voltage (VF) than coloured LEDs. I would work on 3 volts and a current of 10 milliamp (0.01A) per LED. But we need to know the voltage of the supply and whether or not its regulated??

[Gordon H]

Given the extreme reliability of modern LEDs, properly treated, there is no real reason to expect a single LED to fail in a series connected set.
If conditions do arise which cause one to fail, chances are that the others in the same chain would have suffered to some extent too, so would need attention. Essentially, the only likely failure mode is extreme overcurrent for some reason.
However, if you are really that concerned about single failures, it is possible to allow for it to some extent by the addition of a reverse biased Zener diode across each LED.
In normal operation the LED forward voltage would be lower than the Zener voltage, so the Zeners would have no current flowing through them.
If an LED was to go open circuit, the Zener across it would start to conduct instead, giving a minimal effect on brightness but current would still flow.
2.7V Zeners could work reasonably well, producing something like this:

LEDs and Zeners.PNG

[collectors]

Thanks Steve, Brian & Gordon, it feels like a nice school lesson “if i can remember back that far” That link to the very impressive calculator makes life a lot easier & is a great help with the alternative answers it gives & will help a lot with the layout. I will keep it a simple as a little extra work now will make it better in the long run.

Chris