Suggestions for measuring temperature of SMT D2PAKs...

Apr 16, 2009 12 Replies

Hello, Does anyone have a suggestion for the best way to measure the heat given off by a D2PAK SMT package which is mounted to a PCB and active in the circuit? Here are the things I am considering....



1) Thermal imaging camera is waaaay too expensive and out of the question.
2) Maybe a hand-held laser thermometer? This link is only $30!!...or should I go to the local Sears/Lowes?
formatting link

3) I could epoxy a thermocouple tip directly under the package on the opposite side of the board where the thermal vias come through, but the laser thermometer seems so much nicer and quick and friendly compared to this...especially if I want to measure several different boards!!
4) Thermistor is an option, but very similar to #3 above.
5) A hand-held Fluke-style meter with a hand-held surface probe is another consideration, but I would need to buy it....does it outperform the laser thermometer for any reason?

Optional details for those bored people just wanting to harass me about something.... I have a pre-production pump controller module in my hands. It uses an Infineon half-bridge driver, but the darned things are very slow with respect to switching times, so I am generating more heat than I would like. To compensate for the slow switching, we designed the PCB to be an effective heatsink. Unfortunately the PCB house slacked off on two of our very important fab notes 1) copper fill the thermal vias, which they didnt do (air doesnt conduct heat as well as copper) and 2) we asked for 2oz copper but it really looks like they used



1/2oz copper. I would like to now quantitatively understand the heat issues and be able to compare it against a properly done PCB under full load.

thx alot! frenchy (Steve French)


World, One additional note....a buddy of mine is bringing over a "Thermistor Ring Terminal" here in a few hours...Advanced Thermal Products P/N: A1004RT24K16...not sure how to effectively adhere it to the PCB/D2PAK- device for optimal temperature readings. Man, that no-contact laser solution seems so much easier! I wonder if the laser thermometer has "Emissivity Concerns" that would make a thermistor better/more reliable/quicker/accurate-on-the-fly...hmmm.... thx, frenchy (Steve French)

Cheap IR thermometers have a very wide field of view and probably won't measure a small object like this very well.

Omega has some nice small foil-type thermocouples with an adhesive/fabric base, sort of like a tiny band-aid. Or just solder a small thermocouple onto the drain tab.

Solder conducts better than air, too.

It's hard to get real 1 oz copper, much less 2 oz. People start with

1/4 or 1/2 and plate it up, and they tend to quit too soon. One of my fab drawings says START WITH 2 OZ COPPER.

John

Yep. If you're willing to pay more (but don't have enough for a fancy thermal imager like you have, John :-) ) here's one that has a considerably smaller spot size:

formatting link

Same company that makes the cooking thermometers that Alton Brown recommends. :-)

---Joel

The IR thermometers work well, but spot size may be an issue. You need to worry about emissivity. You can warm up the board and take a few measurements and find out what materials have high emissivity. Any uncoated metal, like solder, will probably be low. Epoxy will probably be high. Matte Scotch Tape is high.

If you want to measure the device temperature, you'll probably need a thermocouple tacked to the tab.

Mark

Steve... Go to th Harbor Freight site you show and upper left corner type in item number 93984. That is one I use. Warren

I've done exactly this sucessfully and easily using Dallas/Maxim Thermochron iButtons. I held them in place with tape, leave then there for hours to capture the thermal data and then upload and plot. If I needed to correlate some circuit action with the data I'd simply synchronise my watch and note the time the action took place. Then it was easy to see how the temperature changed at that moment in time. Easy.

Dave.

Wow, that one is only $15 and it has a laser sight! Are you happy with the accuracy and spot size? I notice it has a "6:1 spot size to distance ratio"... have you been happy with using it on small spots like a D2PAK? That means you have to be approx 0.1" away from the device (almost touching it).

thx, frenchy (Steve French)

John, Thanks for your helpful reply! A few follow-ups below...

I just borrowed a hand-held Omega CL26 and I plan to solder a thermocouple to the D2PAK tab just as you suggest. For quick-turn, high accuracy, this seems better than shotty cheap laser thermometers with large spot size and emissivity issues.

Thanks for the comment...I see you said the same thing back in 2003. I think it would be best for me to start a new thread dedicated to PCB copper weight.

thx, frenchy (Steve French)

I use a small tube in front of the sensor on a typical laser IR thermometer to limit diameter. This way I use to locate overheating parts. Else I use my finger, which can get too hot sometimes.

greg

limit diameter.

So now you're measuring the temperature of the walls of the tube.

John

Black whiteboard marker is excellent, and wipes off easily.

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