# Help me understand triac maximum current and when to use a heatsink

Datasheets list , under Absolute Maximum Ratings RMS on-state current = 16 A, for example.

1)How do I know if this is free air maximum current or properly heatsinked?

2)How can two triacs in identical packages have different maximum current?

3)If the answer to 1 is "with heatsink", how do I determine the maximum free-air current?

• Can you provide the device number, or a link to a datasheet? Different manufacturers put information into sheets in slightly different ways, so the best information will come back if you provide this info. – Scott Seidman Jan 23 '13 at 17:09
• @Scott Seidman, I would rather not, for two reasons. First, not citing a part number leaves the question general. Second, I need this information in selecting a device. It has to operate at 1.5A at 40C ambient. Currently I am using BT136, with a limiting current of 4A, and it generates ~80 degrees temperature drop. – Vorac Jan 23 '13 at 17:14
• Could please someone enlighten me on point 2) up there. – Vorac Jan 23 '13 at 17:38
• @Vorac how is it that we can have thousands of transistors in TO-92 packages, each with different characteristics? The package is just a box. What's inside? – Phil Frost Jan 23 '13 at 17:51

Without reading the datasheet, you can't know for sure, but if this is a device designed to accept a heatsink, it's safest to assume that any maximum power or current ratings assume a huge heatsink.

If you need to calculate the maximum current with a smaller heatsink, work backwards. First calculate the power dissipated at a given current. This will vary from device to device based on how it was manufactured. What semiconductor was used? What's it's geometry? How was it doped? The datasheet won't answer these questions, but it will tell you the practical consequences that result.

Once you know the power you must handle as heat, find all the thermal resistances from the junction inside the device to the ambient air. The datasheet should give a junction-to-case thermal resistance, and your heatsink will specify its own resistance, given natural convection or forced air. There is some additional resistance from the case to the heatsink; if the datasheet doesn't give a value, then $0.5 ^\circ C / W$ is a good estimate for a TO-220 with thermal grease.

Add all the thermal resistances together, and calculate the temperature rise above ambient by multiplying that total thermal resistance by the previously calculated power. Add this temperature rise to your maximum ambient temperature and compare this against the maximum junction temperature, also specified in the datasheet.

Exceeding the Absolute Maximum rating is likely to destroy the device. Therefore, operating the device at that point is not a good idea. The datasheet should also provide a normal operating current. That's the figure you want to design with.

Heat comes from power, and power being voltage * current, you have to include the voltage drop across the device to find out how much heat is generated. That's one reason two devices could dissipate the same amount of power at two different operating currents.

• Not to contradict gbarry, just to drill this in yet another time: unless you know very very well what you are doing and what Absolute Maximum Ratings means (it is sort of a legal term) you must IGNORE that section of a datasheet. – Wouter van Ooijen Jan 23 '13 at 19:58
• @ Wouter van Ooijen , why should I ignore it? – Vorac Mar 13 '13 at 8:53