# Heat dissipation in a DIP Solid State Relay

I am working on an application that switches a high power (15 W typical, 100 W maximum) resistive load at 230 VAC (RMS) from a 5 VDC control line. For this I plan to use a Solid State Relay. I am currently planning to use the AQH3213A from Panasonic, but I am concerned about the heat dissipation of the part. The load will be switched on regularly anywhere from a few seconds to a few minutes. I performed the following calculation for power dissipation under maximum load conditions:

$$P = I_{Load}\bullet V_{TM}$$ $$= 0.435 A \bullet 2.5 V$$ $$=1.09 W$$

However the datasheet (or any other information I've been able to find from Panasonic) does not list any thermal resistance values for the DIP package the SSR uses.

How can I figure out how hot the SSR will get under maximum conditions, and whether I need a heatsink? Furthermore how could I effectively heatsink a DIP 8?

• figure 1 in reference data says you're golden. it doesn't say how hot it will get, but implies that the temperature will be less than 125C (max storage temperature) I would expect well less than 80C at 25C ambient. – Jasen Jun 26 '17 at 8:09
• heatsinks don't wore well on plastic covered parts. – Jasen Jun 26 '17 at 8:14
• @Jasen Why not? There's a ton of heatsinks available for single-board computers which typically have the CPU packaged in a plastic BGA. – Dmitry Grigoryev Jun 26 '17 at 8:24

## 2 Answers

"Peak ON-state voltage" is not necessarily something that will apply to your SSR the whole time it is switched on. It is listed in the datasheet so that you can account for the voltage loss your load can see, not to calculate the dissipated power.

There's a figure in the datasheet which shows you how much current is OK for a given ambient temperature. This is what you should take into account.

DIP8 has 100 C/W Junction to Ambient according to linear. Thus it will raise 100 degrees at 1 Watt.
At 40 ambient with 1 Watt, the junction will be 40 + 100 = 140 degree.

The real numbers of this part might be slightly better then the 2.5V specified, keeping it below 125 at 40 ambient.
Note that only the maximum Vtm specified, which look right for the 0.3A part, the 1.2A parts will be better. But no real numbers are given.

Which explains the derating in Fig 1, they try to keep the junction below 125C: