1206 thermals

Dec 20, 2010 6 Replies

We were thinking about adding a feature to one of our boards, sort of a programmable circuit breaker thing. It would just be firmware, but it could add some thermal stress to some 1206 resistors. So I did a quick measurement:



A 1206 resistor, mounted normally on layer 6 of a 6-layer board, with a solid ground plane on layer 5, about 8 mils away. Hot-spot theta, measured with an IR viewer, was about 80 K/W. So this would be comfortable at, say, half a watt dissipation.



John


What's your maximum allowable ambient temperature near that resistor?

Those thermal imagers are the bees knees. They're still a bit expensive, but very worth it (ours was about $2.5k).

Bob

== All google group posts are automatically deleted due to spam ==

60C maybe. It's a VME module, so we can expect pretty good air flow across the board. I did the resistor measurement in still air.

What kind is that? In knew prices were coming down, but that's impressive.

Our FLIR is a couple of years old now. It cost $12K. But it's pretty good. It can measure the hot-spot temperature of an 0603 resistor.

Here's a tiny NEC dual transistor with power being dissipated in one side.

ftp://jjlarkin.lmi.net/UPA800_80mW_one-side.jpg

So much for any hopes of thermal matching.

John

Ours is made by FLIR. I didn't buy it so I'm just going by what I was told. I was showing it to our receptionist and took a moment to image her. It's amazing what it reveals.

So, your 1206 will be running at 100C? Is it rated for that?

Bob

== All google group posts are automatically deleted due to spam ==

These resistors, common Panasonic thickfilms, are rated for 1/4 watt or 1/8 watt dissipation, depending on which specs you prefer. Whatever that means.

I don't trust any of these specs. The hot-spot temperature is wildly dependent on how well you get the heat out of the end caps. An 0603 can dissipate a watt if you solder it to really big copper pours. This

1206 has the standard pcb pads, but the nearby ground plane sucks a lot of heat out. I figure a hot-spot of 100C is conservative.

I tried pulse testing these resisors, too. They die at about 60 watts for 1 millisecond.

John

Looks like 2 dies in one package.

Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------

Yes. This, like many dual transistors, is an isolated pair that shares only epoxy. Thermally, it's hardly better than two SOT23 transistors side-by-side.

ADI did recently revive their monolithic duals.

John

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required