Jaycar Electronics Reference Data Sheet: LEDUSEJ.

PDF (1)

Using LEDs – It’s Easy!

LIGHT EMITTING DIODES or ‘LEDs’ are now very
widely used in almost every area of electronics, mainly as ANODE CONNECTION EPOXY ENCAPSULATION WITH
indicator and display devices — in effect, ‘solid state lamps’. WIRE WEDGE BONDED MAGNIFYING DOME LENS
They’re very well suited for this kind of use, because they TO POST, BALL
are physically quite rugged and hence much more reliable BONDED TO CHIP
than filament-type incandescent lamps. They also run much LED CHIP ATTACHED VIA
cooler and are much more efficient, requiring far less CONDUCTIVE EPOXY
electrical power input for the same amount of light output.
Other common uses for LEDs are as a source of either REFLECTOR DISH COINED
visible or infra-red light, transmitted as a carrier for data ANODE POST INTO TOP OF CATHODE POST
and other information over short ‘line of sight’ distances.
A LED is basically just a specialised type of P-N junction FLAT ON CATHODE SIDE
diode, made from a thin chip of fairly heavily doped OF BASE FLANGE
semiconductor material. When it is forward biased to ANODE LEAD
reduce the potential barrier provided by the junction’s (LONGER) CATHODE LEAD
narrow depletion layer, electrons from the semiconductor’s (SHORTER)
conduction band can combine with holes from the valence
band, releasing sufficient energy to produce photons of
light. Because the chip is thin a reasonable number of these
photons can leave it and radiate away as its light output. Fig.1: The construction used for
most common ‘leaded’ LEDs.
Unlike diodes made for detection and rectification, which
are generally made from either germanium or silicon, LEDs
are made from compound semiconductor materials such as current level of 20mA. The construction of a typical LED is
gallium arsenide (GaAs), gallium phosphide (GaP), gallium shown in Fig.1 above. LEDs with leads like that shown are
arsenide-phosphide (GaAsP), silicon carbide (SiC) and made in a variety of package shapes and sizes, of which the
gallium indium nitride (GaInN). The exact choice of 3mm, 5mm and 10mm diameter ‘bullet’ type with a
semiconductor determines the wavelength of peak emission spherical front lens are the most common. Other much
of photons — and hence the colour of the light emitted, in smaller packages are used for surface-mount LEDs.
the case of visible light LEDs. It can also determine the
electro-optical conversion efficiency, and hence the light How they’re used
output for a given amount of forward conduction current. In most cases LEDs are operated from a low voltage DC
Another parameter determined by the compound supply, with a series resistor (Fig.2) to limit the forward
semiconductor used in a LED is the forward voltage drop current to a suitable value — from say 5-6mA for a simple
for a given forward conduction current. pilot lamp or status indicator application to 20mA or more
where more light output is needed. As you can see the
Table 1 shows the peak emission wavelength for the series resistor value is easily worked out knowing the
most common types of LED, with the nominal colour required operating current I F, the supply voltage and the
indicated and also the typical forward voltage drop for a LED’s forward voltage drop at this current level.
The LED’s voltage drop V LED can usually be estimated
Table 1: Typical LED characteristics from the figures given in Table 1 for a current of 20mA,
although the actual voltage drop will be a bit lower for
Nominal Semiconductor Peak Emission VFORWARD @ 20mA
Colour material wavelength
much lower current levels. Note that the resistor value is
Typical Range
Infra-red GaAs 870-940nm 1.2V 1.1 - 1.6V
Red 1.5 - 2.6V RESISTOR
GaAsP/GaP 650-700nm 2.0V VRESISTOR Fig.2: In most
1.7 - 2.8V + –
Orange GaAsP/GaP 620-650nm 2.0V + + cases LEDs are
Yellow InGaAlP 580-620nm 2.4V 1.7 - 3.0V operated from
A LED DC, and a simple
Green GaP 510-560nm 2.8V 1.7 - 4.0V series resistor is
VSUPPLY CURRENT l VLED used to set the
Blue GaN/SiC 420-470nm 3.6V 3.2 - 4.3V
IF forward current
White GaN/SiC 460 +570nm 3.6V 3.2 - 4.3V K
level. The value
– – of the required
resistor is easily
ORG

INFRA RED ULTRA VIOLET

GRN
RED

BLU
YEL

VLED + VRESISTOR = VSUPPLY calculated

IR

So VRESISTOR = VSUPPLY – VLED knowing the

And VRESISTOR = IF x RRESISTOR LED’s forward
So RRESISTOR = VSUPPLY – VLED voltage drop, the
supply voltage
IF and the current
Where VLED is found from LED data you want, as you
1300 600 400 300 200 80
1000 800 100 and IF is chosen to give desired can see.
WAVELENGTH (NANOMETRES)
300 FREQUENCY (TERAHERTZ) 3000
light output (brightness)
 Jaycar Electronics Reference Data Sheet: LEDUSEJ.PDF (2)
even though current is still drawn from the supply. So the
Rs light output and efficiency are both effectively halved.
The method shown in B is more efficient, and also
K A LED maintains the LED’s light output. Here a bridge of four
DIODE CURRENT diodes is used to ensure that the current always flows
VS (AC) DIODE l A through the LED in the forward direction, regardless of
supply polarity. But notice that the voltage drop of two
A K diodes — about 1.2V — needs to be taken into account
LED CURRENT
when the value of series resistor Rs is being calculated.
Need more light?
The maximum continuous light output from a LED (usually
~ measured and rated in millicandelas) is essentially limited
Rs by the maximum average forward current which it can
LED handle, which is determined mainly by the LED chip’s
l power dissipation rating — typically less than 100mW for
VS (AC) + A K
– B plastic encapsulated devices.
DIODE When higher light output is required, the usual approach
BRIDGE is to operate the LED not from a steady DC supply, but
from a pulsed current with a fairly short duty cycle (on-off
~ ratio). This allows the current and hence the light output
to be increased significantly during the actual pulses, while
Fig.3: If you do need to run a LED from an AC still keeping the LED’s average current level and power
supply, there are a couple of ways of doing so. In A dissipation within its ratings (Fig.4).
above a reverse-connected diode is used to limit Why does this pulsed output give an advantage? Partly
the LED’s reverse voltage to about 0.6V, while in B because the efficiency of LEDs actually tends to increase
a diode bridge rectifier is used to ensure that the with current level. So short pulses of significantly higher
LED receives only unidirectional rectified current. output separated by periods of no output actually result in
a higher average light output, for the same average current.
not especially critical, so once you calculate the theoretical Rs
resistor value you can pick the nearest preferred value. +
It’s important to realise that LEDs are designed to
operate ONLY in forward conduction mode, and should IPEAK A
not be subjected to reverse voltage. They have a reverse
l LED
voltage rating of 5V or less, so they can be damaged by IAVERAGE
accidental reverse connection. K
If a LED does need to be operated from an AC supply, 20%
or from a signal source which cannot be relied upon not to 100% –
reverse its polarity, it can be protected by one of the
methods shown in Fig.3. The simplest approach is shown in Fig.4: To get more light output from a LED, it can
A, where a reverse connected silicon diode is connected be run from a higher current but pulsed, so the
directly across the LED to limit any reverse voltage to average current is still within the LED’s ratings...
0.6V. This protects the LED, but of course no light is
emitted for the negative half-cycles of the AC waveform —
Also, the human eye’s persistence of vision tends to ‘fill
+9V
in the gaps’ between the light pulses, providing the pulse
repetition frequency is higher than the eye’s critical fusion
A frequency (CFF). So light pulsing at a frequency of 100Hz
R1 l LED or more actually appears steady, but brighter than the light
8 4
resulting from the same continuous average current.
Vcc RESET K
RLED
7
DIS OUT
3 A simple low cost LED pulser using a 555 timer device is
shown at left in Fig.5. The value of capacitor C1 can be
6 IC1
R2 THR
555
chosen to set the pulse rate, after which resistor R2 can be
2 VC
5 chosen to set the LED’s on time. Then the value of R1 is
TRG selected to give the desired duty cycle. For example R1 has
a value of nine times that of R2 for a duty cycle of 10%.
C1 1 10nF
If you use this circuit to obtain the highest average light
output from a LED, for example, you might use a value of
say 100nF for C1 and values of 120k and 15k for R1 and R2
respectively. R LED could then be chosen to give a peak LED
TLOW (LED on) = 0.693 x C1 x R2 current of 200mA (i.e, about 33Ω for a typical red LED).
THIGH (LED off) = 0.693 x C1 x (R1 + R2) The same circuit can also be used to make an ordinary
LED ‘flash’ quite slowly and visibly, so it’s easily seen even
Fig.5: How to use a 555 timer chip to operate a LED when running at a very low average current. For this kind
in pulsed current mode. Capacitor C1 can be chosen of use you’d make C1 470nF, R1 2.7MΩ and R2 330kΩ, and
to set the pulsing/flashing frequency, while resistors make RLED about 330Ω. This gives only 2.1mA of average
R1 and R2 are chosen to set the LED on and off LED current, but it will clearly ‘flash’ once per second.
times. For a 10% duty cycle for example, you’d
make R1 nine times the value of R2. (Copyright © 2005, Jaycar Electronics)