Jump to content
  • Sign in to follow this  

    LEDs and LED mountings


    If you think you know everything about LEDs then I invite you to read the following article where we have gathered the most important useful information about LEDs. Also in this article we have gathered some electronic schemes that can be used to power LEDs.

    1. What are LEDs?
    An LED ( English : light-emitting diode, means a light-emitting diode) is a semiconductor diode that emits light at the direct polarization of the pn junction. The effect is a form of electroluminescence.
    An LED is a small light source, most often accompanied by an electrical circuit that allows modulation of the shape of the light radiation. Most of the time they are used as indicators in electronic devices, but more and more have started to be used in power applications as light sources. The color of the light emitted depends on the composition and state of the semiconductor material used, and may be in the infrared, visible or ultraviolet spectrum. LEDs are used to provide white and color light in compact lanterns, bulbs and luminaires as well as in a wide range of electronic devices.

    2. Brief history.
    Electroluminescence was discovered in 1907 by HJ Round, using a silicon carbide crystal and a primitive semiconductor metal detector. The Russian Oleg Vladimirovich Losev was the first to create the first LED in the 1920s. His research has been around the world, but it has not been used for decades.
    In 1961, Bob Biar and Gary Pittman, discovered that by applying current to an alloy of gallium and arsenic, it emits infrared radiation. The first LED in visible spectrum (red) was made in 1962 by Nick Holonyak, when he was working at General Electric Company. A former student of his, M. George Craford, invented the first yellow LED and improved the illumination factor of the red and red LEDs - orange about ten times in 1972. By 1968 the visible LEDs and the LEDs infrared cost a lot, almost $ 200 and could not be used only for minor applications.

    The first large-scale corporation to manufacture LEDs was Monsato Corporation, producing LEDs for display in 1968. These were taken over by Hewlett Packard and integrated into the first alphanumeric computers. The first widely marketed LEDs were used to replace incandescent indicators, first on expensive equipment such as labs and tests, then later on TVs, radios, telephones, computers, even watches. These red LEDs could only be used for indication because the light emission was not sufficient to illuminate a surface. Over the years other LED colors have been discovered, with higher lighting capabilities.
    The first LED with high lighting capacity was made by researcher Shuji Nakamura in 1993 from an InGaN alloy. It was awarded in 2006 with the Milennium Technology Prize for its invention.
    In 2008, the most powerful LED commercialized belonged to the South Korean company Seoul Semiconductor. A single LED in the Z-Power P7 series achieves 900 Lumen performance at 10 watts, so an efficiency of 90 lm / W, equivalent to a regular 75W bulb.

    On May 12, 2010, Nexxus Lighting presented the most powerful LED lamp available on the market, with an efficiency of 50 Lumen / Watt. The brightness of the PAR38 Array LED lamp is comparable to that of an ordinary / standard 75 Watt bulb reaching 985 Lumen at a consumption of only 18-20 Watt, while being variable.
    On April 12, 2010, Toshiba presented the prototype of the most powerful LED lamp for domestic and industrial use, with an efficiency of 120 Lumen / Watt. [4] The brightness of the led lamp is comparable to that of a standard / standard 100 Watt bulb, reaching 1690 Lumen.
    On December 18, 2012, Cree presented the XLamp MK-R LED Lamp with an efficiency of 200 Lumen / Watt and a dimension of 7 x 7 mm. [5] The brightness of the led lamp is comparable to that of a 120 Watt incandescent bulb, reaching 1769 Lumen at 15 W and 85 ° C.

    3.Simbol. Classification. Construction.

    3.1 Symbol


    3.2 Classification. Construction.
    The LEDs fall into two broad categories:
    a) Low power LEDs (<1W);
    b) High power LEDs.
    Depending on the type of construction the LEDs are divided into:
    a) THT LEDs (case a below);
    b) SMD LEDs (case b below);
    c) Power LEDs (one or more high power SMD LEDs).


    Depending on the number of colors rendered we distinguish: single, two-color and three-color LEDs (the latter is also called RGB LEDs).

    4. Efficiency and operating parameters of LEDs Typical indicator LEDs are designed to operate at most with 30 ... 60 mW of electricity. Around 1999, Philips Lumileds introduced power LEDs capable of running continuously with one watt. These LEDs are based on much larger semiconductor to be able to absorb high power. Also, the semiconductor molds were mounted on metal supports to allow heat transfer from the LED mold. 

    One of the key benefits of LED lighting sources is the high light efficiency. The white LEDs match quickly and have successfully replaced standard incandescent lighting systems. In 2002, Lumileds made five-watt LEDs with a luminous efficiency of 18-22 lumens per watt (lm / W). In comparison, a conventional incandescent bulb of 60 or 100 W has an efficiency of about 15 lm / W, and standard fluorescent lighting up to 100 lm / W.

    Since 2012, the Lumiled catalog presented a table with the most efficient light sources according to color.


    In September 2003 the company Cree introduced a new type of blue LED that consumes 24 mW with a consumption of only 20 milliamperes. The commercial variant that produced white light with an efficiency of 65 lm / W at 20 mA, became at that time the brightest white LED on the market being four times more efficient than the standard incandescent bulbs. In 2006, they presented a white LED prototype with a light efficiency of 131 lm / W at 20 mA.

    Nichia Corporation has developed an LED with a luminous efficiency of 150 lm / W at a current of 20 mA. By comparison, XLAMP Cree LEDs, which have been commercially available since 2011, have an efficiency of 100 lm / W consuming 10W and can go up to 160 lm / W with only 2W consumed. In 2012, Cree launched a white LED capable of 254 lm / W.
    General lighting requires high power LEDs, one watt or more. Typical operating currents for such devices start at 350 mA.
    Note that these levels of efficiency are achieved only by the monobloc LEDs and could be obtained at a low temperature in a laboratory. The lighting operates at higher temperatures and with losses in the supply circuit, so the resulting efficiency is much lower. The US Department of Energy, after testing the commercial LED lamps intended to replace incandescent or CFL lamps, showed that the average efficiency in 2009 was approximately 46 lm / W (the performance of the tested LEDs varying between 17 lm / W and 79 lm / W).

    On February 3, 2010 the company Cree issued a press release regarding a prototype LED laboratory that has an efficiency of 208 lm / W, at room temperature. The color temperature was reported at 4579 K. In December 2012, Cree issued another press release announcing the commercial availability of LEDs with an efficiency of 200 lm / W at room temperature.

    5. Lifespan and failure rate
    Solid-state devices, such as LEDs, are subject to quite low wear if they operate at low currents and low temperatures. Many LEDs designed in the 1970s and 1980s are still in service and in the early 21st century. The typical lifetime of an LED is between 25000 and 100,000 hours, but the heat transfer with the environment and the way we choose the operating current (see section 8 the notion of ILED_OPTIM) can significantly prolong or shorten this time.

    6. Colors and materials
    Conventional LEDs are made from a variety of inorganic semiconductor materials. The table below shows the colors available with a range of wavelengths, voltage drops and materials:


    Blue and ultraviolet

    LEDs Blue LEDs are based on a semiconductor band gap made of gallium nitride (GaN) and indium gallium nitride (InGaN). They can be combined with red and green LEDs to produce the impression of white light. The modules that combine the three colors are used in large video screens and in adjustable color programs (using RGB LEDs).

    The first blue LEDs were produced in 1971 using gallium nitride by Jacques Pankove of RCA Laboratories. These devices had too little light to be useful in practice and research into gallium nitride (GaN) devices slowed down. In August 1989, the Cree Company introduced an indirect semiconductor tape gap made of silicon carbide, resulting in the first commercially available blue LED. The silicon carbide (SiC) LEDs had very low efficiency, no more than approx. 0.03%, but they emitted in the blue portion of the visible light spectrum.

    At the end of the 1980s, significant progress was made in epitaxial growth and doping with gallium nitride-type carriers, which led to the launch in the modern era of gallium nitride (GaN) optoelectronic devices. Starting from this base, high-brightness blue LEDs were demonstrated in 1993. High brightness blue LEDs were invented by Shuji Nakamura of Nichia Corporation, using gallium nitride, which revolutionized LED lighting, making high power LED light sources feasible.

    At the end of the 1990s blue LEDs became widely available. They have an active region consisting of one or more quantum gaps in indium gallium nitride (InGaN) which is between several thick layers superimposed on gallium nitride (GaN), called plating layers. By the relative variation of the In / Ga ratio in quantum gaps of InGaN, the light emission can theoretically be varied from purple to amber. Various Al / Ga ratios of aluminum gallium nitride (AlGaN) can be used for the manufacture of tiles and quantum layers for ultraviolet LEDs, but these devices have not yet reached the technological efficiency and maturity level of InGaN blue / green devices / GaN. If pure GaN is used in this case to form the active quantum layers, the device will emit near-ultraviolet light with a maximum wavelength centered around 365 nm. Green LEDs manufactured using the InGaN / GaN system are much more efficient and brighter than green LEDs made from materials that do not contain nitride, but practical devices still have too low yields for high brightness applications.

    White light

    There are two main ways to produce white light emitting diodes (WLEDs) or LEDs that generate high intensity white light. One is to use individual LEDs that emit the three primary colors: red, green and blue, and then blend all the colors to form white light. The other way is to use a phosphor material to convert monochromatic light from blue or UV to wide-spectrum white LEDs in the same way that fluorescent tubes work.

    There are three main methods of blending colors to produce white light using LEDs:
    - Blue LED + Green LED + Red LED (mix color, can be used as backlight for screens);
    - in the vicinity of UV or UV LED + RGB phosphor (LEDs that produce white light with a shorter wavelength than when using blue to excite an RGB phosphor);
    - Blue LEDs + yellow phosphorus (two complementary colors combine to form white light - more efficient than the first and most frequently used methods).
    Due to metamerism (or changing the color of an object viewed in different lights or with different power spectral distributions), it is possible to produce quite different spectra that appear white.


    7. LED performance The

    table below highlights the performance parameters for three types of lamps analyzed and a forecast of the performance of LED lamps in 2017. Also in the table was calculated the "total lamp life". This parameter represents the measured luminous flux accumulated over the entire life of the lamp and is measured in megalumen-hours. The light efficiency for an LED lamp in 2012 is 65 lm / W. The last row in the table is calculated by making the ratio between the luminous efficiency of the other lamps and that of the LED lamps from 2012.
    The scalar impact of future LEDs is expected to be smaller as future LED performance will improve and designers will continue to seek and increase the quality of materials and components used in their construction.


    8. How do we control the LEDs?

    Light emitting diodes (LEDs) are constructed from PN semiconductor junctions. When the LED is directly polarized the electrons are able to recombine with the holes inside the device and release energy in the form of photons. The current passing through the PN junction of the LED will have to be limited and influence the brightness of the LED.

    The setting and limitation of the current can be done in several ways:

    • using an external resistor to limit the direct current through the LED;
    • using a constant current source to establish a definite and stable current through LED (Constant Current - CV);
    • a DC-DC converter in constant voltage mode (Constant Voltage - CV).

    The above methods will be detailed in point 8 of this article.

    8.1 Powering the LEDs using a resistor
    One of the simplest ways we can control the lighting of an LED is to attach a resistor to it to limit the LED current to an appropriate value (a value that indirectly respects and results from the catalog data. LED, which ensures the proper functioning of the LED). If we attach to this circuit a simple switch to stop and start the LED operation, it is the simplest circuit to power an LED (fig.3a).


    a) without a resistor or a similar ballast circuit that limits the LED current to a certain value, the LED will burn;
    b) the battery plus will have to coincide with the electrical path that makes contact with the LED anode, respectively minus the battery with the LED cathode;
    c) the identification of the anode and cathode terminals for an LED is done as in figure 4.


    Figure 3b shows the case where we have several identical LEDs that we want to power from a battery using a single resistor.

    The mathematical formulas for calculating the value of the resistor are the following:
    - for figure 3a: R 1 = (U battery - U LED1 ) / I LED1 or taking into account that we have a safe LED the simplified formula can be applied: R 1 = U battery / I LED1 ; - for figure 3b: R 2 = (U battery - U LED2 - U LED3 - U LED4 ) / I LED      
       , where I LED  will be selected so that it is supported by all LEDs and will have to ensure the proper functioning of all LEDs. This scheme applies when all LEDs are of the same type, so they have the same characteristics. For example, in the case of LEDs with a diameter of 3mm but which have different colors - as in the figure above - a current of 5mA can ensure the correct functioning of all LEDs.
    Attention, it is not recommended to power the LEDs by applying the solution in fig.3b when we have LEDs that do not have the same LED_OPTIM .
    In the case of ordinary low power LEDs, such as those in Figure 2, the optimal LED current recommended for use in calculations is shown in Table 5.


    Note:  ILED_MAXIM may differ from one LED model to another, respectively from one color to another. TheLED_MAXIMvalues higher than those presentedcan be found in the catalog. Note, in the calculations, the values in column ILED_OPTIMwill always be used, which will ensure an appropriate voltage drop on the LED for its or their proper functioning.
    Figure 5 shows a method of identifying the operating voltages of the LEDs (noted in the technical documentation in English with VF).


    8.2 Powering the LEDs via a constant current source (DC - Constant Current) or a constant voltage source (CV - Constant Voltage)

    We have previously seen the simplest method by which we can power an LED using a common resistor. The method is simple but has a major disadvantage: due to temperature variations and taking into account any variations in supply voltages (that is, if we do not use a battery but a power supply with transformer and rectifier bridge with capacitive filter), the LED current can record considerable variations leading to the decrease of the LED life. To combat this phenomenon, several types of circuits have been designed to provide a constant current or a constant voltage at the LED terminals.

    Figure 6 shows graphically the functions corresponding to the three modes of LED power supply: using a constant voltage circuit (fig. 6a), using a constant current generator (fig. 6b) and the CC-CV version that combines the properties of the circuits of constant current with those of constant voltage.
    After analyzing the three graphs in fig.6 a question will appear quite common among the developers of LED lighting applications, namely: which is the best method of controlling the LEDs: CC or CV or CC -CV?

    Throughout this article I will try to bring enough arguments to complement your current experience to make the appropriate decision in various situations. I said earlier that LEDs are semiconductor devices that need a certain current in order to operate, then you will surely wonder why companies supply both LEDs with constant current (DC) power supply and power solutions. constant voltage supply (CV)?

    The main reason is that companies want to give designers enough options to enable them to optimize the lighting system. If several LEDs are connected in series, the most efficient way to power them is to connect them to a constant power supply. If the LEDs are connected in parallel, there may be a problem with the current distribution through each LED. A possible alternative to this power mode is the placement of an external component or an active electronic component that controls the LED current. Although this strategy provides the same current through each LED, the method results in a less efficient lighting solution,
    What is the difference between a constant current source (DC) and a constant voltage source (CV)?

    Figure 6 shows the three characteristics of the three distinct modes of operation of the LED power supply. The X axis shows the load increase, and the Y axis shows the output voltage of the LED power module. The blue line represents the voltage and the green line is the output current.

    To begin with, we will analyze the performance of the constant voltage power supply (fig. 6a). As the name suggests, the circuit returns to the output a constant voltage as the load current increases (symbolized in the figure by the English term "load"). The load current will be able to increase until a moment when the circuit enters the current limiting mode, in order to prevent circuit damage.
    Figure 6b shows the characteristic of a constant current source. In this case, if the load changes (increases or decreases), the current will remain constant.

    Figure 6c shows the characteristics of a circuit that combines two modes of operation. Initially, the circuit acts as a constant voltage source. Once the maximum permissible load current is reached, the circuit control loop will adjust the load current to a constant value while simultaneously reducing the output voltage. This type of approach has many benefits and allows the designer to achieve greater efficiency using modern solutions based on CV-CC sources. In the last period, a lot of types of integrated circuits have been developed that use the characteristic of fig. 6c, some of them being presented in this article.


    8.3 Series and / or parallel LEDs?

    --- Serial connection --- 

    Powering multiple LEDs in series avoids uneven brightness due to current variation. So, all LEDs will see the same current for the same brightness level. The output voltage of the driver will be equal to:

    V OUT  = V F  X n

    where  V F is the nominal operating voltage of the LED and n is the number of LEDs connected in series. For example, if V F   = 2V and we have 5 LEDs connected in series then the output voltage of the LED driver will be 10V. Most LED drivers are DC / DC type low voltage converters. Care must be taken to keep the input voltage within nominal limits so as not to exceed the output level above the appropriate limit.
    When LEDs are connected in series, the output current of the driver will be equal to:

    I OUT  = I F ;

    where I F is the nominal current of the LED, a very important catalog date. So all LEDs connected in series will see the same current, in our example I will consider: I F = 30mA. Advantages of LED series connection:   

    • low circuit complexity;
    • each LED sees the same current;
    • high efficiency (no ballast resistor required).

    Disadvantages of LED series connection:

    • the output voltage of the driver can become quite high for LEDs connected in series;
    • over the lifetime, LEDs can unevenly change their operating parameters, leading to overcharging of some and undercharging of others, which will cause much faster LED string failure or brightness. uneven;
    • if a LED fails, the brightness of the entire serial connection is interrupted. A short-circuited LED has a reduced effect on the overall brightness of the circuit but can cause overvoltage of the other LEDs in series, if the LED driver is not provided with a current reaction by which to automatically adjust the output voltage to the corresponding value - I mention this because most LED drivers made with linear voltage stabilizer integrated circuits, output a fixed voltage, which does not automatically adjust according to the nominal current consumed by LEDs. In this situation, it may happen that at one point the fixed output voltage is too high for the n-1 LEDs remaining in operation.

    --- Parallel connection --- Suppose we have three rows of LEDs connected in parallel. Through each row or string of LEDs a current circulates that I will note with: IF1, IF2and IF3. The LED driver will need to provide a constant output voltage equal to nsx VF, where nsis the number of LEDs in a string. Then the output current of the LED driver will have to be the sum of the currents that flow through the three LED strings connected in parallel, namely:I OUT = I F1 + I F2 + I F3 If: IF1= I 

    F2 = I F3 = I F , then: I OUT  = 3 * I F .

    The output voltage of the driver will remain the same as initially calculated, ie those 10V, if we have five LEDs connected in series on each row with VF = 2V. If we consider IF = 30mA then the output current of the driver will be 90mA. So, by connecting the LEDs in parallel, the output voltage may be the same as in the case of serial connections if we have the same types of LEDs, but the current required for their power supply will increase depending on how many LEDs we connect in parallel.

    The major advantage of using parallel LED connections is that we can use a larger number of LEDs which, if connected in series, would require a higher supply voltage than the nominal output of the driver. of LEDs. For example, if the upper voltage limit of an LED driver is 28V and V F  = 2V, we can connect at most 14 LEDs on a single string. If the driver has a current capacity greater than IF, suppose 0.3, and I F  = 0.03A, that means we can connect at most 0.3 / 0.03 = 10 LED strings in parallel. Thus the total number of LEDs connected to the output of a driver, both serial and parallel, will reach 14 x 10 = 140 LEDs.

    The biggest problem with parallel connections is that the small differences in tolerances or manufacturing dispersion of the components of a circuit can lead to significant differences in the current absorbed by each string of LEDs. This will have repercussions on the perception of the intensity of the color or the brightness of an LED, reaching even in extreme cases when the failure of one or more LEDs will cause the entire circuit to be switched off.

    In order to eliminate or better said to minimize the consequence mentioned above, a series of balancing (balancing) resistors must be connected in series with each LED string: R B1 , R B2  & R B3, which will help to compensate for the current variations caused by the typical V F  LED differences. Small imbalances of V F  within a series of LEDs connected in series could cause a significant variation of the current I F through the string. The typical value of the ballast resistance is less than 20 Ohms. In other situations, a better current generator with transistor or a current mirror with transistors is used to better balance the currents connected in parallel instead of a common resistor. The ballast or balancing resistors in the structure of the constant current generators or the current mirrors will be used continuously to compensate for the small Vbe variations. To keep the current constant regardless of temperature variations, the transistors used in such a manner will have to be "compensated / thermally connected". Mounting them on the same radiator is a common method to do this.

    Advantages of parallel LED connections:

    • the possibility of supplying a large number of LEDs.

    Disadvantages of parallel LED connections:

    • Low efficiency;
    • Increasing circuit complexities;
    • Low reliability.

    The low reliability is caused by a considerable risk of current variations occurring. A shorted LED will increase the current I F  through the remaining series LEDs. An increased current will cause the other LEDs in the string to fail. If the junction of an LED breaks, this will cause all LEDs in a row to stop working. So, both possible defects are the main cause of low reliability for such an LED connection.

    8.4 Matrix connection

    To help improve the reliability of parallel connections, the matrix connection can be used where the LEDs are connected on horizontal and vertical supply sides. This mode of connection, called in specialty literature and cross connection, is nothing more than the connection of LEDs in series and in parallel. In this connection, the required output voltage and current is the same as when connecting LEDs in parallel, where the number of LEDs that can be powered without exceeding the maximum voltage allowed by the driver is much higher than in the case of series of LEDs.

    However, the matrix connection is somewhat more error tolerant, no balancing resistors are used for parallel operation and the efficiency of the connections is much improved. However, current distribution on the sides of the matrix remains a problem. An unequal distribution of current (caused by component tolerances) can lead to visible differences in brightness. For example, differences in thermal characteristics may cause a variation of current that will inevitably lead to the problem previously reported over time.

    A shorted LED will cause a vertical power supply to be turned off and the remaining LEDs will continue to operate normally. If an LED is not open, only the remaining LEDs on that row will be able to operate. Therefore, the matrix connection allows individual control of a large number of LEDs using a driver with a lower output voltage than when using serial or parallel connections.

    8.5 Multichannel connection

    Usually, it is recommended to use the LED serial connection whenever possible because this connection mode avoids the thermal distribution problems encountered in the case of matrix and parallel connections. The most robust connection is to use a separate driver for each LED string or a single multi-channel driver. This connection mode combines the reliability advantages of serial connections with a high current capacity characteristic of matrix and parallel connections. The obvious disadvantage of such an approach is the increased costs and complexity.

    9. Protective circuits

    LEDs are extremely reliable devices, with average lifespans approaching 50,000 hours. By far the most common failure is the gradual degradation of the light intensity to 50% of the nominal value.
    However, failures also occur due to mechanical stress or temperatures, misuse, packaging defects etc. The most common and "catastrophic" failure for LEDs is to stop them. When this happens, as we have seen, in the case of the serial connection, all the LEDs in the string are interrupted. A frequent cause that leads to this type of defect ("interrupted led") is the application of excessive excessive voltage.

    The use of a constant current source will protect the LEDs from the defect described above. However, the components may be subjected to under / over voltage which may be induced by external circuits or events. Thus, in order to prevent and eliminate the above phenomenon, a protective device (PDX) is connected in parallel with each LED. These devices are nothing more than switches, which if one of the LEDs fails to open, the PDX shuts off to ensure continuity of operation for the other LEDs in the string. Once the LED is replaced, the PDX will reset automatically. Usually, to keep costs to a minimum, a PDX connects in parallel on two serial LEDs.

    10. LED mountings

    Regardless of type, color, size or power, all LEDs work best when powered by a constant power source. The manufacturers mention the characteristics of the LEDs, such as: light efficiency, color, etc., at a certain current (denoted by IF) and at a certain operating voltage (denoted by VF). Most integrated circuits designed to pilot LEDs are designed to deliver good output voltage stability across a range of currents. Therefore, it can be quite difficult to determine which LED method is best for a particular application. From the beginning we try to avoid that solution by which each LED is supplied by a constant current source, because the method, In applications that require a large number of LEDs, it becomes economically inefficient. As a result, during the development of LED applications, it was tried to adopt that power solution that would lead to: a minimum number of constant current sources and a maximum number of LEDs, in conjunction with an arrangement through which to the best luminous efficiency is obtained. Thus have appeared various solutions of serial / parallel connection of LEDs, with the disadvantages and the advantages of rigor, solutions that I will list below. in conjunction with an arrangement to obtain the best luminous efficiency. Thus have appeared various solutions of serial / parallel connection of LEDs, with the disadvantages and advantages of rigor, solutions that I will list below. in conjunction with an arrangement to obtain the best luminous efficiency. Thus have appeared various solutions of serial / parallel connection of LEDs, with the disadvantages and advantages of rigor, solutions that I will list below.

    10.1 Power supply of LEDs through constant current generator with transistors
    One of the oldest methods of supplying one or more LEDs is shown in Fig. 8 below.



    10.2 Power supply of LEDs through constant current generator with LM317


    LM317HV adjusts ~ 1.23V between ADJ and V OUT terminals . The current through LEDs is given by the relation: I LED  = 1.23 / R. The advantage of this circuit is that by maintaining a constant current with the help of voltage control on the resistor R, the voltage at the LED terminals is also maintained constant.
    In the case of the circuit in Fig. 15, the current through each of the three LED branches connected in series is given by the relation: ILED = 0.6 / Rsense.

    10.3 LED driver with LM2941
    Another driver similar to LM317 is LM2941. LM2941 is a voltage regulator that allows up to 26V input voltage. LM2941 regulates 1.275 V between ADJ and GND terminals. Figure 16 is an LED driver with LM2941 capable of delivering 354 mA.


    10.4 Led driver with LT3021 The
    LT3021 is another linear voltage regulator with a maximum admitted voltage of 10V input at a current of 0.5A (fig.17). The LT3021 maintains constant 0.2V between the ADJ and GND terminals. The LED current is given by the 0.2 / R ratio. If the nominal voltage of the LED is 3.6 V, the number of LEDs connected in series is equal to two.

    10.5 LED driver with TLE4242G
    An integrated circuit that can be used to power LEDs is TLE4242 (fig. 18), where VREF is 177 mV between ADJ and GND pins. The maximum input voltage at the input is 42 V. Using a 5.1 Ohm resistor the LED current will be 34.7mA.

    Switched LED drivers

    A switched LED driver is linked to how the switching sources work and are built. The switching voltage regulator maintains a constant voltage at different current loads. Therefore, switching sources intended for powering LEDs can maintain a constant current through LEDs at any voltage drop on the LED, provided the overvoltage protection and thermal packing of the source are operable or, in other words, exist. in the construction of the source.
    Some examples of switching power source topologies are:

    • BUCK  - where the output voltage is generally lower than the input voltage;
    • BOOST  - where the output voltage is generally higher than the input voltage;
    • BUCK  - BOOST, is an up / down voltage structure, with inverted output;
    • SEPIC  (single-ended primary inductor converter). This converter borrows the functions of the buck and boost converters, increasing or decreasing the output voltage, even though at one point the input voltage may be lower than the output voltage. The control strategy allows to obtain a much lower noise source, in conjunction with a minimum number of external components.
    • FLYBACK.  A topology based on the functions of buck and boost converters, but instead of inductors contains a transformer that galvanically isolates the output input.

    Before choosing a particular topology to build or power an LED driver, I recommend that you study the technical specifications of each integrated circuit. In applications that use switching sources as a voltage source, in order to power one or more LEDs in series, it is recommended that the LEDs be connected to the output of the source before mounting power because the source will have to operate in empty an output voltage greater than the nominal operating voltage of the LEDs.

    10.6 LED driver with L6902 (fig.14)
    L4902 is a "buck" switching LED driver. Table 6 below shows the values of the various components required to obtain a certain LED current.


    Resistors R1 and R3 have the role of providing overvoltage protection (typically 23.3V) in case it is interrupted, for example, due to an LED failure, supplying all LEDs. If the operating voltage of an LED is 3.6V then 6 LEDs can be connected in series.


    10.7 LED driver with L4973
    Another buck topology driver topology is shown in Fig. 15, where the input voltage is 48Vdc. In this case, up to 12 LEDs with Vf = 3.6V can be connected in series. Resistors R1, R2 and the internal voltage reference of 5.1 V, reduce the voltage response Vfb to 0.5 V.
    10.8 LED driver with LTC3783 (fig.16)
    This LED flyback driver can provide 150 mA to several LEDs -s connected in series. Surge protection triggers at 130 V and deactivates at 120 V. The number of series LEDs that can be connected is: 120 / 3.6 = 33 LEDs. The PWM terminal can be used to control the brightness of LEDs connected in series.


    10.9 LED driver with MAX5035 (fig.17)
    This buck topology supports input voltages between 7.5 and 30V. LED current is typically 350mA and the output voltage for LED power supply is 12V. 3 3.6V LEDs can be connected in series.
    V CONTROL  is a linear voltage for controlling the brightness of the LED, by which the LED current is modified according to the following equation:

    I LED  = [V REF  x (R 1  + R 5 ) - V CONTROL  x R 1 ] / (R 5  x R P )


    • R P  is the parallel equivalent resistance of resistors R2, R3 and R4.
    • V REF  is typical 1.22 V.
    • V CONTROL  is a voltage for controlling the brightness, by default of the LED current, which starts from 0 V for the maximum LED current and can reach the maximum output voltage.

    - http://www.energystar.gov - http://en.wikipedia.org/wiki/LED - http://www.cree.com/ - http: //www.talkingel...plifier- P2.html - Datasheets: LM317HV, LM2941, LT3021, TLE4242G, L6902D, L4973, LTC4930, LTC3783, MAX5035. - Maxim AN3668 High-Efficiency Current Drive for High-Brightness LEDs. - Maxim AN3639 Design of a Nonisolated, Flyback LED Driver Circuit.  

    Sign in to follow this  

    User Feedback

    Recommended Comments

    There are no comments to display.

  • Member Statistics

    Total Members
    Most Online
    Newest Member
  • Recently Browsing   0 members

    No registered users viewing this page.

  • Create New...