Multiple LED Driving

[bigkim100]

Why?
It absolutely doesn't matter whether the resistor is connected to anode or cathode. At least functionally.
Things become a bit more interesting when you look at EMC, specially at ESD-proofing your circuit. Then the resistor is best placed at the side where the driving transistor sits to reduce ESD influence on the transistor. But in that case there are again two possibilities:

• common anode
• common cathode

Like so:
View attachment 55205
When you use multiplexed LEDs where a driving transistor sits on both sides of the LED, it becomes even more complex. In that case it is often helpful (although from a purely technical point of view not necessary) to split the resistor like so:
View attachment 55206
Again note: This is sometimes used to improve resilience of a circuit against ESD. This is not the standard application a hobbyist will use.

THANK YOU HARALD!!!!...the second circuit is something that I have been looking for!!!

 

[bertus]

Hello,

You can find a lot of information on leds on AllAboutCircuits in the following thread:
https://forum.allaboutcircuits.com/threads/leds-555s-flashers-and-light-chasers.19075/

Bertus

 

[hevans1944]

For a Non-Professional guy, just screwing around, The LEDS there are fine for my purpose, and the prices are in Canadian with free shipping

Yep. LEDs ain't very demanding in how you use them, or even difficult to manufacturer, so they are dirt cheap. I do strongly agree with @Audioguru that if you want components that are built to some specification, then it is quite difficult to obtain them from Asian sources.

We have an excellent resource here on EP that instructs how to wire up and use LEDs. You need to know just two things about your LEDs

Bottom line is this: never try to wire two or more LEDs in parallel. Always use a single current-limiting resistor in series with one or a few LEDs. Everything, resistor and LEDs, is connected in series. If you connect two or more LEDs in series, the voltage required to light 'em up will be the sum of the voltage across each LED, which is of course larger than the voltage drop across a single LED.

I use a "rule of thumb" of allowing about 1.5 volts for each LED. This gets me in the ballpark. If I want a home-run, I have to measure the voltage drop across each of my LEDs. Then I subtract the voltage across the entire series-connected string (1.5 volts multiplied by the number of LEDs) from whatever power supply voltage I happen to have handy. If the result is a positive value, then I probably have enough voltage from the power supply to light up that string. If I want to light up more than one string, each string of LEDs, plus their series current-limiting resistor, is connected in parallel across the same power supply. I just make sure the PS can handle the current.

You need only know the voltage drop across all the LEDs in a string, plus the current required by any one of those LEDs, to calculate the value of the current-limiting resistor. For example, if you were to want four LEDs connected to a six-volt battery, you can either wire all four LEDs in series with one resistor, or wire four resistors, one in series with each LED, to make four strings. Connect all four strings in parallel across the same battery.

So which way should you wire the four LEDs? Applying the "rule of thumb" to one LED dropping 1.5 volts, means you need at least 6 - 1.5 = 4.5 volts from your 6 volt battery to light up one LED with its single current-limiting resistor. Gud enuf! You just need to make four strings.

OTOH, if you were to wire all four LEDs in series with a single current-limiting resistor, then you need roughly (1.5 x 4) = 6 volts to light up the string from a 6 volt battery. NOT gud enuf! There is zero voltage left over to drop across a current-limiting resistor! Depending on the LEDs (you don't know their characteristics since you bought a mixed bag really cheap from Asia), four of them connected in series might not drop the "rule of thumb" voltage. It could be less than six volts or more than six volts. So, to be on the safe side, you decided to invest in three more small current-limiting resistors, one for each LED, and wire each resistor-LED series-connected string in parallel across the battery.

That just leaves the problem of what size current-limiting resistor you need. Its purpose is to limit the current in the LEDs to less than or equal to the rated current of the LED. But wait! You don't know that either since the LEDs are without a datasheet specification! Most small LEDs require only about 0.02 amperes to light up visibly. A single LED with a single current-limiting resistor will need to drop approximately 4.5 volts across the resistor when used with a 6 volt supply. With 0.02 amperes flowing through the resistor, it's value would be R = V / I = 4.5/0.02 = 225 ohms. Again, these are all "ballpark, back of the envelope" values. You should measure the actual voltage drop across each of the LEDs you plan to use. Base your current-limiting resistor calculation on those measured values. Ahem. I almost NEVER follow my own advice. If I let the "magic smoke" out of a few LEDs, so what? Plug and pray again. LEDs are dirt cheap and so are low-wattage film resistors.

The amount of power dissipated in the resistor is P = V x I = 4.5 x 0.02 = 0.09 watts or about a tenth of a watt. This would be a very tiny resistor in carbon composition, carbon film, or metal film varieties, but larger 1/8 watt, 1/4 watt, and even higher-wattage resistors are perfectly acceptable. Just use whatever you have on hand that is "close" to the computed value.

I keep a goodly supply of 100 ohm, 1/4 watt resistors available to "play" with, connecting them in various series and series-parallel arrangements until I find a combination that works. Then I measure the resistance of the combination and search through my junque box for a single resistor of similar value. You can purchase bags of assorted resistors for just a few cents per hundred. Radio Shack used to sell specific values in their "mixed bag" of resistors, and you can probably find Asian vendors who will do the same. I have not found any problems buying small, fixed value, resistors from Asian sources, usually via Amazon.

If you have a bag of assorted LEDs, and a rainy Saturday morning in which to sort through them, it might be a good idea to measure the voltage drop across each LED, write that number down, then "bin" the lot according to color and voltage. Small paper envelopes work well for this. The voltage drop across an illuminated LED depends on the color, starting low with red and infrared LEDs and increasing in voltage as the color progresses towards blue and ultraviolet. So-called "white" LEDs are actually UV LEDs illuminating a phosphor that then emits "white" light... sort of like a solid-state "curly bulb" or vacuum fluorescent lamp only smaller and more efficient.

 

[bertus]

Hello,

Have a look at the attached OSRAM application note on parallelling leds.

Bertus

 

[hevans1944]

@bertus: That is a very interesting article for anyone needing or wanting to wire up a large number of LEDs, especially if the LEDs are unmounted chips, ready to be soldered to a massive heat sink for use as an automobile headlight. For that application, a hard failure of any one LED should not cause the entire headlight to become inoperative. OTOH, failure of any single LED should also be readily detectable by the electronics driving the headlights, or as a visible reduction in the light intensity of the headlight.

Failure of an LED usually occurs because the LED becomes overheated. Detachment from the heat sink or a huge spike in supply voltage causing excessive current are two typical reasons for failure. Clearly, the driver of the automobile having a headlight with one or more dark LEDs needs to replace that headlight as soon as possible. Back in the day, cars had a pair of headlights on each side. When one headlight with incandescent filaments burned out, you could still drive safely down the road at night using the other three headlights for illumination. When LEDs started replacing headlights with incandescent tungsten filaments, the automakers dropped the redundancy of a pair of headlights on each side. Well, that's my observation, folks. Your visions or hallucinations may differ.