3
\$\begingroup\$

I want to measure the temperature of IC/Microcontroller. There is no On-die temperature sensor to measure the heat. I need to check how much hot the controller gets. The LM35 or DHT11 Temperature will not work as they sense ambient temperature not the surface temperature. How about thermocouples or Thermistor NTC?

Any other device for measuring the temp of IC?

This is the device OP is referring to:

enter image description here

\$\endgroup\$
1
  • \$\begingroup\$ any comments on my suggestion? \$\endgroup\$ Dec 8, 2016 at 19:33

3 Answers 3

6
\$\begingroup\$

If you want to measure the actual die temperature precisely you may be able to use an internal diode anyway. Such parasitic diodes usually exist, for example from an input to V- on a bipolar chip. You could calibrate the diode in an oven or environmental chamber on a particular chip (no power). To measure the actual temperature you would remove the power and quickly (before the temperature changes much) check the diode.

Alternately, a small very fine wire bead or fine ribbon thermocouple to the leadframe will measure some approximate temperature, and an IR thermometer looking at the encapsulation an approximation of the package temperature. If you use a thermocouple, double back the wires and heatsink to other parts of the leadframe to prevent excessive errors due to heat loss down the wires (while maintaining electrical isolation).

\$\endgroup\$
4
  • \$\begingroup\$ I have to measure the temp of this particular module. adafruit.com/product/2491 \$\endgroup\$
    – MICRO
    Nov 1, 2016 at 18:17
  • \$\begingroup\$ The chip itself is under the shield. You should be able to use the method I mentioned, possibly even during operation (though I would not recommend it unless you feel lucky) if you can calibrate the diode. Just pull an input below ground (say with a 100K resistor and -5V) and measure the voltage vs. temperature characteristic. That will give you the actual die temperature. \$\endgroup\$ Nov 1, 2016 at 18:54
  • \$\begingroup\$ If the chip has an open-drain output that can be left high-Z during the tests, that'd be ideal, basically... \$\endgroup\$ Dec 7, 2016 at 23:22
  • 1
    \$\begingroup\$ Using a parasitic diode on an IO pin like this is common practice for measuring accurate die temperatures in the semiconductor industry. Even for a device that is all MOS, there is often a useable parasitic diode between any pin and the substrate due to ESD protection devices. Experiment by sinking a very small current out of an IO pin - preferably one that would otherwise be logic low during operation. \$\endgroup\$
    – user49628
    Dec 8, 2016 at 0:13
0
\$\begingroup\$

What IC/Microcontroller would you like to measure?

Like Spehro Pefhany said, didoes are commonly used to measure the temperature of computer processors: http://www.ti.com/lit/an/sboa019/sboa019.pdf

Thermocouples are a nice idea if you want to actively maintain a certain temperature. Unless if the surrounding temperature is consistant, I wouldn't use a thermocouple to measure the temperature of a chip. The thermocouple would give you the temperature difference between the chip and the surroundings, this could be very high in a cold environment and very low in a hot environment, or even worse, you might get a negative signal when moving from a cold to a hot environment...

\$\endgroup\$
1
  • \$\begingroup\$ I have to measure temp of WiFi chip, ESP8266. There is a more modular version of that called ESP12E. Here is the link adafruit.com/product/2491 \$\endgroup\$
    – MICRO
    Nov 1, 2016 at 18:14
0
\$\begingroup\$

use any thermometer, thermistor, thermocouple or temp IC

  • This becomes a thermodynamic problem

Requirements:

  • decide also if you also want to improve thermal stability and specify
  • identify the thermal resistance of various case packages and thickness
  • define if you want a permanent bonded temp. sensor or temporary.
  • identify thermal resistance of various adhesives per mm, hotglue, epoxy, cyanoacrylate, polyurethane, silicone ( get familiar , generally thinnest is best )
  • design your connection to be >5x thermal insulating to ambient, relative to contact with hotspot to reduce ambient effect.
  • understand the junction temperature rise above case temperature for various thermal insulators like epoxy on SOT23 and FR4 PCB vs thickness and metal cladding
  • characterize RF interference with dielectric or conductors near critical locations on board and effect on impedance of transmission lines
    • use your fingertip to detect changes in Performance . eg. RSSI, Tx level
    • locate the hotspot and choose Temp IC and attach.
    • avoid RF interference with measurement sensor, (See if Tx causes false readings and determine if any fix can be done or ignore during Transmit.)
    • minimize RF coupling to Sense wires to avoid damaging sensor from excess power
    • .. e.g. use suitable ferrite sleeve over wire pair and ferrite beads on signal for both Common Mode and Differential Mode RF coupling.

suggestion

  • use appropriate size wiring for sensor, such as AWG 30 magnet wire or wirewrap wire and bond with a toothpick dot of cycnoacrylate and coat with a few mm of silicone for insulation from ambient.
    • if you have thermocouples, wisely choose location considering above interference to RF and electrical shorts and bond with Kapton Tape to board.
  • use tape for strain relief on fine wires or bond with epoxy dots to sensor and dots spaced along wire over board.
    • keep in mind adding thermal ambient insulation to sensor is counter-productive to keeping board cool with free ambient air flow.

It all depends on the reason you are measuring T.

  • To design a better heatsink?
  • to integrate into another product and measure hotspot rise?
  • to compute the Thermal resistance of the Rjc for junction to case

inside photo of older version

enter image description here

Improved version FCC qualified.

enter image description here

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.