Electronic – LDO selection based on power dissipation

First, a couple of the answers (at least on the first draft) seem to have confused SOT-223 with SOT-23. SOT-23 is a very small packaged designed more for small size than for heat dissipation. SOT-223 is also quite small, but does have a substantial thermal tab:

Sources differ on the actual thermal properties of SOT-223. The TI app note AN-1028 cited by Garrett gives a junction-to-ambient thermal resistance (\$\theta_{JC}\$) of 12 C/W. The Microchip app note AN792 also cited in Garrett's answer gives 57 C/W. Another TI datasheet, for the TLV1117, gives 104 C/W.

The main reason for this discrepency is that the thermal resistance depends not just on the package, but on the size of the copper pads available to serve as a heat sink for the part, as shown in this graph taken from the TI app note:

The 12 C/W number is apparently the asymptotic limit of this curve. Note that it requires 2 oz copper and probably 2 in² or more of copper area to achieve that value.

To finally get to your question, how to lay out the heat sink pad, in roughly decreasing importance:

• The larger the pad you can fit in your design the better.
• Heavier copper is better (e.g., 2 oz rather than 1 oz copper).
• When connecting through to a thermal pad on the opposite side of the board, use many vias. As a rule of thumb, I'd recommend spacing vias on a 50 mil grid or so, over the whole pad area.
• Use vias larger than the minimum size. As a rule of thumb I'd try to use at least 8 mil via diameter and use 10 to 18 mil by preference. Extremely large vias, of course, end up reducing the pad area, so there's a limit to how large you want to go.
• Place the heat-generating part as close to the center of the thermal pad as possible.

Finally, in contrast to the suggestion in another answer, I would do my design this way:

• Determine the input and output voltages of your regulator, and the operating current. From this determine the power requirement.

• Determine the maximum ambient temperature where your circuit will operate.

• Determine the maximum junction temperature you can operate at. Typically this is 125 C in the datasheet, but you may want to de-rate by 25 C or more to give design margin and improve reliability.

• Now choose a package and design a layout that allows you to meet your maximum operating junction temperature.

In particular, it is not possible to determine the temperature rise until after you've chosen a package.

Cooling is a vital part of most computer system design.

Air at room temperature is the usual coolant, because it's available everywhere* and you can move it easily with a small fan.

Some enthusiasts and large systems are liquid cooled. All the plumbing and pumping systems can't be miniaturized, and add cost, so this is usually done only on "tower" size systems: http://www.pcworld.com/article/227964/pc_liquid_cooling_system_do_it_yourself.html

The extreme is liquid nitrogen cooling: http://www.corsair.com/blog/setting-up-your-pc-for-liquid-nitrogen-overclocking/ , which is even more expensive and impractical. It's nearly always easier to stick with conventional cooling and buy more computers to achieve the same computational throughput.

(There are also issues with hold timing when running extremely cold - signals may arrive too early if the chip has not been designed to run at that temperature. If you move the minimum design temperature down to -200 that may force you to move the maximum temperature down as well, leaving you with a product that requires a liquid air supply and thus a very limited market)

• Except in space, where ultimately all your cooling must be radiative