Selection of heat sinks in LED thermal management
Auto Loader good features as following:
1.host machine separated from the hopper,easy and safe to operate
2.Independent filter for easy cleaning
3.Stainless hopper
4.Microcomputer controlled
5.Alarm for overload and lack of material
6.High-speed rectificative motor for small footpoint and strong suction
force for new material for RAL-300
7.Induction high-pressure pump of low noise and long service time for
RAL-303 and RAL-503
8.Reverse cleaning for RAL-303 and RAL-503,RAL-300(Optional purchase)
9.With CE standard(Optional purchase)
Feeding And Conveying,Plastic Auto Loading Machine,Plastic Loader For Auto Loader,Automatic Loader With Infrared Hopper Ningbo Beilun Rhong Machinery Manufacturing Co., Ltd. , http://www.rhongmachinery.com
As the forward current of high-brightness LEDs increases and the package size decreases, the potential for thermal runaway and catastrophic failure increases. In many LED applications, higher levels of protection are required due to extreme high temperature environments.
Thermal foldback is a common method of reducing LED failure and avoiding LED life shortening due to overheating. This control method uses a signal that is inversely proportional to temperature and reduces the LED current after setting the temperature breakpoint.
Here are two examples: a 100W streetlight application and a 12W military flashlight application. These two examples illustrate the differences between the more complex systems and the simpler systems and their respective design flows.
BACKGROUND In conventional lighting applications that use high power LEDs, large heat sinks are required to dissipate the heat released by the LEDs. The LEDs themselves do not dissipate heat, instead they conduct heat through the semiconductor junctions. This conduction power (PD) is equal to the product of the forward voltage (VF) and the forward current (IF).
PD=VFIF
In order to maintain a safe LED junction temperature, this conduction power must be eliminated. Thermal impedance in the system needs to be analyzed to customize a heat sink at rated power to ensure the desired thermal characteristics.
A typical high-power LED will consume most of its power through its components, solder joints, printed circuit boards, and heat sinks. As shown in Figure 1. With this simple model, the calculations are fairly simple. The power dissipation (PD) of the LED junction must be distributed by the junction-environment total thermal resistance (JA), which is very similar to when the current passes through the electrical resistor.
The resulting temperature difference (TJ-TA) between the junction temperature (TJ) and the ambient temperature (TA) is equal to an electrical voltage (the thermal equivalent of Ohm's law):
TJ-TA=PDJA
JA refers to the sum of the following values.
JS: junction to solder joint thermal resistance;
SH: solder joint to heat sink thermal resistance;
HA: Solder joint to ambient thermal resistance.
JS stands for internal LED thermal resistance, while SH stands for printed circuit board (PCB) dielectric and junction thermal resistance. Finally, HA represents the heat sink resistance, JS is the value specified in the LED manufacturer data sheet, and is a simple LED package function. It can vary from 2 to 15 ° C / W. If the solder joints are well connected to the heat sink (including: multiple thermal vias, proper copper, good soldering and possibly thermal paste), SH is basically negligible. This will result in a very low SH value of less than 2 ° C / W.
HA remains the same because it depends more on the surface area of ​​the heat sink and its thermal conductivity. On a standard FR4 printed circuit board (similar to the size of the LED), there is no external heat sink, only the bottom copper layer, the HA value can be very large, exceeding 100 ° C / W. Through the external heat sink shown in Figure 1, the thermal resistance can be reduced to maintain the desired temperature difference (TJ-TA). Thermal design requires the selection of a suitable heat sink based on the following HA equation:
It is easy to calculate by this equation that if the power is increased or the allowable temperature difference is reduced, the necessary thermal resistance will be reduced, which is equivalent to the need for a larger heat sink.
In practical applications, the output LED power will increase by 5~10 forward due to the forward voltage and other electronic deviations during the service life. The range of possible temperature rises is calculated based on the worst-case expected TA value. In addition, in the specifications specified by the manufacturer, the maximum allowable TJ value is usually reduced to ensure that the LED life and efficiency are not reduced. These tolerances force us to improve the worst-case thermal design criteria, which is 25 to 50 more than the calibration.
LED Drivers These LEDs are just one component of a dynamic system with LED driver mastering mechanisms. High-brightness LED drivers typically support their operation through a switching converter. The converter adjusts the system to provide an approximately constant LED luminous flux output. The drive adapts to changing dynamic conditions and provides continuous regulation to ensure electrical stability of the system. In the most common LED drivers, the output current needs to be adjusted because it has a close relationship with the output flux and is easy to adjust.
Although electrical stability is fundamental to the control scheme, thermal equilibrium is a function of controllable variables (LED current) and uncontrollable variables (environment). As the ambient temperature increases from a room temperature of 25 ° C, the forward voltage of the LED decreases. Because the current is constantly adjusted, the power is reduced, eventually achieving the goal of achieving thermal equilibrium at the junction. However, the eventual increase in ambient temperature will cause the junction temperature to exceed the safe operating range of the LED. At this point, the performance of various components within the LED is degraded, degraded, ultimately leading to thermal runaway and catastrophic LED failure.
Each LED manufacturer provides a characteristic curve of the maximum forward current corresponding to changes in ambient temperature. The CreeXRE series curve, shown in Figure 2, identifies the recommended LED over-temperature safe operating range (SOA). This fast reference design material provides multiple JA graphics. Since JS is specified in the datasheet and SH can be ignored in a well-functioning system, HA is a controllable variable. Maintaining a LED drive current within a defined range for a given HA prevents thermal runaway and/or substantial lifetime degradation that occurs when the LED is operating in an unsafe state.
It is not difficult to see from Figure 2 that a large heat sink will expand the range of LEDs. However, in some LED applications, the high cost of the radiator and the larger heat sink volume are prohibitive. For such applications, a good solution is needed to achieve heat dissipation.
Designers prefer a generic approach to designing a thermal management solution with a large tolerance range for each specification. This makes the application of LED drivers possible. Since the drive regulates current and power, it is only necessary to detect non-safe operating conditions and allow the drive to react accordingly.
Thermal Folding Considering the forward current drop specified by the manufacturer, designers can now rely on LED drivers to provide a helpful control mechanism to provide thermal protection for the LEDs. Since most new LED drivers have dimming inputs, there is almost always a simple way to reduce the output current to the LEDs. In view of this, a circuit can be designed to detect the temperature near the LED. If the system has good thermal resistance characteristics, the junction temperature of the LED can be interpolated by measurement.
Therefore, the LED driver can maintain or reduce the regulated current as required by the arrangement shown in FIG. The map can be changed and basically conforms to the manufacturer's data sheet specifications, and it can be drawn more conservatively. Regardless of the method used, the LED must be protected from current overcurrent and overheating. In particular, the heat sink requirements can be reduced as needed, as the thermal runaway caused by the worst conditions can be removed.
Thermal foldback can be applied in many ways. The most common and simple method is to measure the temperature near the LED using an NTC (negative temperature coefficient) thermistor, as shown in Figure 4. The NTC thermistor is a resistor that increases with decreasing temperature and decreases with increasing temperature. If the resistor divider setpoint deviates from the reference voltage and the bottom resistor is a thermistor, the divided voltage will decrease as the temperature increases. If the voltage is clamped to a maximum voltage below the reference voltage, then for some temperature range that rises to the maximum temperature breakpoint (TBK), the voltage is fixed to the clamp voltage. However, for temperatures above TBK, the voltage will drop, as shown in Figure 3. This voltage can be used to control the analog dimming input of the LED driver to perform basic thermal foldback.
When the LED is dimmed, the foldback pattern will be different. Since the nominal LED current level (ILED-NOM) is reduced to the dimming current level (ILED-DIM), the foldback map can be modified to accommodate the new temperature breakpoint (TBK-DIM). This expands the temperature range in which the LEDs are used, as shown in Figure 3. It can be done step by step or continuously according to different devices.
Another variation is an additional minimum LED current (ILED-MIN) clamp to prevent the LED current from being zero, as shown in Figure 3. In many applications, end users do not want a set of thermal foldbacks for security reasons. With this feature, the minimum demand current clamp allows the system to be unconstrained by the safe operating range. However, in this regard, users are willing to trade short-term functions at the expense of shortening their service life.
Street Lights An example of a standard street light is exposed to harsh environmental conditions and the performance of the mechanical heat sink may be degraded for various reasons throughout its useful life. This reduction in performance greatly increases the total thermal resistance JA and will eventually result in higher LED junction temperatures and thus reduced lifetime. In order to meet the requirements of municipal facilities for service life, it is almost always necessary to have thermal reentry in street lamps.
Figure 5 shows a 100W streetlight application. The front-end AC-DC converter takes a 120V AC input and converts it to a 35V DC input. Second order
It is a LM3409 constant current step-down LED driver with a load of 6 series and parallel, in which 8 LEDs are connected in series. Each string has a drive current of 700mA.
The LM3409 regulates current with a simple hysteresis control method. During the main switch (Q1) is turned on, the inductor current ramps up to the peak current threshold set by the IADJ pin. Once this threshold is reached, Q1 turns off and the inductor current ramps down until the programmed shutdown timer stops. The programming of the shutdown timer is implemented by the RC from the output voltage. This causes the timer to be proportional to the output voltage, resulting in inductor current ripple and subsequent, constant LED current ripple that exceeds the operating range.
By reducing the voltage on the IADJ (from 1.24V to 0V), continuous analog dimming of the average LED current can be easily implemented. If the voltage of the IADJ reaches or exceeds 1.24V, then the maximum nominal current of the LED should be adjusted. When the IADJ pin voltage drops to 1.24V, the current begins to dim, providing an excellent way to perform thermal foldback.
The thermal foldback circuit in this application is more fundamental than previously described, using only one IADJ add-on NTC thermistor. The resistance of the NTC thermistor will be higher than 250k (IADJ is greater than 1.24V) until the temperature reaches the required breakpoint. Then as a function of the NTC, the resistance is reduced while reducing the voltage and LED current of the IADJ, respectively.
It should be noted that the NTC's resistance to temperature conversion function is non-linear. This nonlinearity extends the boundary point temperature (TEND) at which true zero current occurs. In streetlight applications, the linearity of thermal foldback is not at the highest level. In fact, the end-of-life time of a streetlight is usually specified as 70 degrees when the brightness is reduced to the initial brightness; therefore, an accurate thermal foldback map is meaningless to the streetlight designer. That is, a sophisticated temperature sensor can be easily used for more linear thermal foldback mapping if needed.
An example of a flashlight is shown in Figure 6 for a more complex thermal foldback device using the LM3424. This application is a 15W dimming military flashlight consisting of LM3424. The LM3424 controls 6 series LEDs with a drive current of 700mA and a battery voltage of 9V. Because the string voltage changes during dimming, from 24V to less than 9V, the multi-topology LM3424 acts as a buck-boost controller. LED analog dimming is required to evaluate its simplicity, size and cost.
The LM3424 uses a conventional error amplifier to regulate the output current in a closed loop. The LED average current difference detection is detected at the top of the LED assembly. The duty cycle of the main switch (Q1) is dynamically changed to ensure that adjustments are made at any time.
The LM3424 features a fully programmable thermal foldback circuit integrated on the chip. The foldback breakpoint is set by the resistor divider in accordance with TREF, and the internal reference voltage is 3V (VS). Temperature sensing is performed using a sensor or NTC voltage divider in the case of TSENSE. When the TSENSE voltage drops to a predetermined TREF voltage, the circuit begins dimming the LED as shown in FIG. The slope of the thermal foldback can be set by the resistor (RGAIN) installed between TGAIN and GND. If you use a sophisticated temperature sensor, such as the LM94022, you can get an advanced linear plot.
A reference voltage VS external Zener clamp can be added to set the minimum required current, as shown in Figure 3. This highly controllable thermal foldback also maximizes the life of the flashlight while maximizing the brightness output of a particular LED for a given temperature value.
Another useful feature in flashlight applications is the combination of dimming and thermal foldback. Because both use thermal foldback