I wonder if it's a good idea to put LEDs in parallel, as below:
I've heard that it might not be because the voltage threshold won't be exactly the same for each LED, so they'll shine all with very different brightness and you don't have a way to balance that. Is that true? So does it mean it's a bad idea?
It's not a good idea. Look how a (generic red) LED conducts current when you apply a voltage to it: -
At 2 volts, the LED is taking 20 mA. If the LED was manufactured slightly differently it might require 2.1 volts or maybe 1.9 volts to push 20 mA thru it. Imagine what happens when two LEDs are in parallel - if they "suffer" from normal manufacturing variations, an LED that only needs 1.9 volts across it would hog all the current.
The device that needs 2.1 volts might only receive 5 mA whilst the 1.9 volt device would take maybe 35mA. This assumes a "common" current limiting resistor is used to provide about 2 x 20 mA to the pair.
Now multiply this problem out to 8 LEDs and the one that naturally has the lowest terminal voltage will turn into smoke taking the best part of over 150mA. Then the next one dies then the next etc...
Not a good idea.
Not only will you have inconsistent brightness (which you may not care about) you will have a cascade failure mode.
You've got 8 LEDs in your circuit with one resistor. 8 x 20mA is 160mA. As long as each LED takes roughly the same current we are fine.
Now lets say one of them gets a bit warm, and it now pulls more current. positive feedback increases the current until it burns out. Now we have 7 LEDs receiving current meant for 8. Rinse, lather and repeat until the final LED gets all 160mA and comes to a fiery end.
Yes you can do that, and to balance the brightnesses you can put a resistor in series with each individual LED (varying the value of the resistor to compensate for the difference in brightness). Also, if you put a single current-limiting resistor in series with the complete LED group, you will need to take into account the fact that you are using multiple LEDs and calculate the total required current (allowing for their differences in forward voltage drop and the values of their individual resistors if you have put those in), and you cannot use the same value of resistor that you would use if you had just one LED.
EDIT: As "Andy aka" has mentioned above, you need to watch out for such situations. But if you use the correct current-limiting resistors for each individual LED then there shouldn't be a problem (although you will need to determine the correct value for the resistor by measuring the forward voltage drop across each LED, which is not a very elegant approach to circuit design).
EDIT 2: In short, no. This is probably going to cause more problems than it is worth.
Yet another possible solution is this. More expensive than resistors, but it pretty much guarantees equal brightness for each LED with much more relaxed voltage requirements. Just be careful of power dissipation and temperature de-rating.
The best way to do this, if you have to, is to group the LEDS in series in groups of N = 1, 2, 3, 4, etc, depending on the maximum supply voltage and the LED voltage drop, with N giving the maximum voltage drop that will work with your supply voltage. For example, let's say you have 2-volt red LEDs, and you are running them in a car, with a supply voltage between 12V and 15V more or less. You can make strings of 5 LEDs in series, giving a voltage drop of 10V, and to anticipate the maximum voltage at 20ma current put the string in series with a (15 - 10)/.020 = 250 or 220 ohm resistor.
You can them connect as many of these modules in parallel as you want. Of course as your supply voltage falls, the LEDs will dim. But they wont all blow up if one fails.
For decent results you need to "bin" the LEDs. Nowadays, the voltage drop from LEDs bought from the same source under the same part number will be within a tenth of a volt. But it's always good to test.
