Bipolar Bicolor LED

Bipolar LEDs are really two LED in one housing, mounted in opposite direction: depending on the polarity you apply to its two legs, either one of the internal LEDs lightens up.

The image below shows the bi-polar LED on the left side: it has two legs and is indistinguishable from a regular LED.

Alternatively, bi-color LED with three legs exist (on the right side in above image). These are not bi-polar. Instead, such LED have a common anode or cathode (middle pin), and each of the two internal LED exposes a separate pin.

• Bi-Polar: color is controlled by polarity. Only one of the two internal LED can be on at any given time.
• Three-Leg: each internal LED can be powered individually, and both LED can be on at the same time, producing a mixed third color if desired.

In this article, I am looking at bi-polar LED.

Challenges

Since bi-polar LED require reversed polarities, they are typically used in circuitry with bipolar power supply (dual-supply), or with AC (such as audio appliances).

Most DIY circuits are single supply (with a power supply ranging from a positive voltage to GND, and no negative voltages, i.e. battery or USB operated). This can make using bi-polar LEDs more complex.

Dual-Supply Circuits

In dual-supply scenarios (i.e. audio devices with positive and negative voltages), bipolar LEDs just require a single OpAmp to reverse the voltage as needed, and probably the only even simpler approach would be using two separate LEDs.

Here is the basic schematic:

• High: the OpAmp outputs a positive voltage

• Low: the OpAmp outputs close to 0V (GND)

  

However, this approach will not work with a single power supply (one that goes from GND to a positive voltage). It would require a positive and a negative voltage, enabling the OpAmp to produce the negative voltage difference required to drive the other LED inside the bipolar LED.

Testing Bi-Polar LED With Single-Supply

To better understand how bi-polar LED really work, let’s test some bi-polar LEDs.

Manual Powering

Here is the fundamental circuitry: the bi-polar LED is really a package of two LED in opposite direction. Like any other LED, they need a current-limiting resistor, i.e. 330R.

The actual resistor value depends on the type of LED and the voltage you want to supply, yet a 330R resistor should at least produce a visible light emission both for 3.3V and 5V power supply. You can then measure the currents, and adjust the resistor value so that the LED has the brightness you want.

Either use a breadboard, or simply connect the current limiting resistor to either leg of the bi-polar LED.

When you supply power to the circuit, one of the two LED lights up. Once you reverse the polarity of supplied power, the other LED lights up in a different color.

Fine Tuning

When I tested the bi-polar LED with 3.3V and the conservative 330R resistor, the emitted light was a bit dim. The measured current was 4.8mA for red and 3.8mA for green.

The vendor claimed a forward voltage of 1.8V for red and 2.2V for green, and a maximum current of 20mA each. Calculating the current limiting resistor with these values yields this result:

PS> Get-LedResistor -ForwardVoltage 1.8 -Current 20 -OperatingVoltage 3.3


Required Resistor (Ohm) : 75
Operating Voltage (V)   : 3.3
Led Current (mA)        : 20
Led Voltage (V)         : 1.8
Led Color               : yellow

I thus exchanged the 330R resistor with a 75R resistor, which resulted in 13.2mA for green, and 18.7mA for red. The LED now was (marginally) brighter, however the added current wan’t worth it, so I changed the resistor back to 330R.

Always start with a conservative resistor like 330R, then measure the current and note the LED brightness. Next, lower the resistor value gradiually without ever exceeding 20mA. You may notice that the visible brightness won’t change that much because the human vision is not a linear system. You would require a multitude of current to effectively double the visual brightness. Signal LEDs do not need to illuminate your lab. They just need to be clearly noticeable.

Using Two GPIOs

Manually re-wiring cables on a breadboard to reverse polarity is simple, yet how can this be done programmatically and without changing the circuitry?

The objective in this article is to attach the signal LED to an arbitrary GPIO and have it signal its state, without requiring special programming or dedicated additional GPIOs.

Before we use a single GPIO to drive the bi-polar LED, let’s first invest two GPIOs. Using two GPIOs is by far the easiest way to operate bi-polar LEDs.

Here is how it works:

Proof Of Concept

And here is a test setup using a ESP32-C3 (any microcontroller will do):

Note how the bi-polar LED is connected directly to one GPIO, and uses the current limiting resistor to connect it to the other GPIO. The polarity of the LED does not matter (it just reverses the sequence of emitted colors).

Here is the source code:

#include <Arduino.h>

// the output pins for the LED and resistor:
#define GPIO_A 20
#define GPIO_B 21

void Red()
{
  // polarity from GPIO_B to GPIO_A
  digitalWrite(GPIO_A, LOW); 
  digitalWrite(GPIO_B, HIGH); 
}

void Green()
{
  // polarity from GPIO_A to GPIO_B
  digitalWrite(GPIO_A, HIGH); 
  digitalWrite(GPIO_B, LOW); 
}


void Off_A()
{
  // one way to turn off the LED: no voltage difference
  digitalWrite(GPIO_A, LOW); 
  digitalWrite(GPIO_B, LOW); 
}

void Off_B()
{
  // another way to turn off the LED: no voltage difference
  digitalWrite(GPIO_A, HIGH); 
  digitalWrite(GPIO_B, HIGH); 
}


void setup() {
  pinMode(GPIO_A, OUTPUT);
  pinMode(GPIO_B, OUTPUT);
  Off_A();
}

void loop() {
  // let both colors light up for one second, then pause one second:
  Red();
  delay(1000);
  
  Green();
  delay(1000);
  
  Off_A();
  delay(1000);
  
  // let both colors light up in sequence ten times
  for (int i = 0; i <= 10; i++) {
    Red();
    delay(200);
    Green();
    delay(200);
  }
  
  // turn off the LED using the alternate method
  Off_B();
  delay(1000);  
}

Using Just One GPIO

The previous example illustrated that a bi-polar LED can be driven by two GPIOs. The second GPIO must always be inverted to the first. This effectively produces the reversed voltage.

Auto-Generating Second GPIO

In order to require just one GPIO (and ultimately simply use any GPIO that you may use for any kind of purpose), the second GPIO needs to be provided by someone else.

For this, you can use an operational amplifier such as the MCP6002 and use it as an inverter.

The MCP6002 supports sourcing and sinking currents of up to 20mA so you can drive the signal LED directly off the OpAmp. Just make sure you purchase a genuine MCP6002. Market places such as AliExpress routinely sell fake MCP6002 that do not work.

Here are the schematics:

The MCP6002 is wired as an inverter, so when the non-inverting input (pin 3) receives a low signal, the output (pin 1) is high and can source up to 20mA, lighting up the bi-color LED in one color.

When the GPIO turns high, the output switches to low and can now sink up to 20mA, effectively reversing polarity, and lighting up the bi-color LED in the other color.

I was unable to test this circuit yet because I, too, naively ordered MCP6002 at some AliExpress sellers. These OpAmps obviously turned out to be fake. I am now awaiting delivery of genuine components from Mouser. If you have the option to test this circuit before I can, please leave a comment below.

Comments

Please do leave comments below. I am using utteran.ce, an open-source and ad-free light-weight commenting system.

  Show on Github    Submit Issue

^{_{(content created Aug 25, 2024)}}