Thermal management is of critical importance in LED lighting systems. LEDs themselves give off heat when lit due to their own inefficiencies, and knock on effects of that heat can rapidly reduce an LEDs’ lifespan if it isn’t managed carefully. When deciding what cooling approach to take, there are a few considerations. Heat management techniques for LED systems can be divided into those that use passive cooling only and those that need both passive and active cooling. Passive cooling usually involves a heat sink applied directly onto the LED assembly. If the heat sink can’t be made big enough to get rid of all the heat without compromising the design of the rest of the system, then active cooling is required. Active cooling means adding a fan to create forced air flow, usually in combination with as big a heat sink as possible.
A typical approach to adding thermal management to an LED assembly would be to start with looking at the heat sinks available for that assembly, to see whether there is one that can dissipate the heat quickly enough, while allowing the LED to be run at the brightness levels required. The passive-only approach is generally favoured because a fan can cause additional issues. Fans are often the weakest link in terms of reliability in the system; ingress from dust, dirt and insects is particularly common. They are also audibly noisy in most cases, which may or may not suit your application. Space considerations also play a part in how much airflow is available to cool the LED and whether a fan is a better choice.
Heat sink manufacturers often design and make heat sinks especially for certain LED assemblies, so check with the manufacturer’s authorised distributor to see what is available. If there is no specific heat sink available, you still have the option to select a general purpose LED heat sink. However, this involves careful consideration of system conditions such as temperatures, thermal resistances and the amount of power to be dissipated. Let’s take a look at how this is done.
If we consider a typical LED system, let’s say it’s an LED COB module assembled on a heat sink, the main components are the LED chip itself, the module, the heat sink, and some kind of gap filler between the module and the heat sink, such as a thermal pad or some grease. Aavid’s SuperThermal range, for example, uses a proprietary fibre orientation technology that delivers high conductivity, flexibility and adhesion. Every part of the system has its own thermal resistance, which represents how much heat can be dissipated from each part (formally, it’s the temperature difference when a unit of heat flows through that junction in unit time). The thermal resistance of the whole system may be represented by the sum of the thermal resistance of the individual parts, which is equal to the temperature difference between the heat source and ambient divided by the amount of power dissipated.
From this we get the familiar formula for heat sink calculations:
T represents temperature. LEDs must operate below their recommended maximum junction temperature Tj in order to prevent damage (usually Tj is stated on the LED’s data sheet, or you may find it on a graph of lifetime vs Tj in some cases). However, it is often prudent to keep the LED’s temperature below, say, 90% of the maximum Tj, so bear that in mind. Ta is the temperature you expect the ambient air around the heat sink to be. For open air mounted spotlights this might be 30°C, for a recessed ceiling downlighter it might be closer to 50 or 55°C.
θ represents thermal resistance. θjc is the thermal resistance of the LED COB module (a property of the LED module; check the datasheets).θca is the thermal resistance between the LED case and the ambient air, which can be broken down into θb and θs. θb is the thermal resistance between the LED case and the heat sink, that is, the thermal resistance of the gap filler. This is a property of the thermal pad or grease. θs is the thermal resistance of the heat sink, which we are trying to calculate.
P represents power, with Pd meaning the power to be dissipated from the LED module. This can usually be estimated as 80% of the electrical power drawn by the module (the rest goes to light).
Once the calculation is complete, defining θs, or the range in which θs must fall, will allow you to make an informed choice about which heat sink to choose.
For example, let’s say we are selecting a heat sink for an Osram PrevaLED Core Z3 1100 LED COB module, which is 1100 lumens. Suitable candidates include the Mechatronix LPF6050-ZHC. This compact heat sink is 60mm in diameter, and 50mm tall, and it has a thermal resistance θs of 4.0°C/W. If a lower (better) thermal resistance is required, select a bigger heat sink or one with more surface area. For example, the LPF6768-ZHP has a diameter of 67mm and a height of 68mm. Its thermal resistance θs is 2.1°C/W.
If the figures don’t add up, select a bigger heat sink, or if there isn’t space, consider active cooling such as a fan or a fanless air mover such as Aavid’s SynJet range. If all else fails it may be time to run the LED below rated power (though this will reduce the brightness) or choose another LED module altogether.
For more on LED lighting visit Engineers’ Insight, a blog from Avnet Abacus.
