Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Is there thermometer for IC junction temperature?

Status
Not open for further replies.

plouf

Junior Member level 3
Joined
Feb 9, 2012
Messages
31
Helped
1
Reputation
2
Reaction score
3
Trophy points
1,288
Location
Athens,Greece
Activity points
1,597
is there any instrument, who can measure temperature of IC itself
"normal" cheap IR thermometer need a very big surface to scan for correct temperature
 

Hi,

With one power IC I did a trick.
It had no onchip temperature sensor, but it gad some digital inputs with protection diodes.
I "misused" one protection diode to measure the chip temperature.
For the function the input should be low.
But for temperature measurement I introduced a negative current to this input. Maybe 50uA.
Thus the resulting voltage was about -550mV at room temperature. This voltage I monitored.
Now the temperature drift is about -2mV/°C, and because the protection diode is on the silicon, it gives the temperature if the silicon.
This means an increase of chip temperature of 20°C will cause the voltage to be about 510mV.

Klaus
 

hi thanks for reply

i actually wondering about external measurement instrument, (like multimeter)
just place probe in ic case and tell you 100celcius
 

Many multimeters have thermocouple inputs. The sensor is only the tiny junction at the tip of the probe wires. Placing a blob of heat sink compound on the IC and placing the probe tip inside it will give a reasonably accurate indication of temperature.

Brian.
 

Hi,

You can't tell the temperature of inside (silicon, junction) by outside (case) temperature measurement.
Otherwise the value "r_th_jc" often given in a datasheet makes no sense.

If you want to measure silicon tempeeature, then you need a sensor inside the case.
If you want to measure case temperature, then you may use a sensor outside the case.

Klaus
 

Otherwise the value "r_th_jc" often given in a datasheet makes no sense....

You are right, of course, but the same information can be turned on its head.

We can have, for example, for any given chip temp a corresponding case temp. These two are related by the temp gradient and is the thermal resistance for the heat to flow out...

If the case temp is measured accurately (the metal tab is the best candidate) with a thermocouple (even a thermistor), then we can estimate the junction temp fairly accurately.

In the world of statisticians, all measurements are nothing but exercises in estimations. If the case temp is measured accurately, the estimate for the junction temp can be fairly reliable (for all practical purposes).
 

Hi,

There are a lot of informations we can't know.

But we know that the OP asked for junction temperature. If he wanted to know case temperature I assume he asked for it.
And he wrote "place probe in IC case" (not outside IC case)

We don't know
* chip case and it's size
* ambient temperature
* air flow or other influence in case temp (PCB mount, existing heatsink...)
* whether this is a single measurement in laboratory or a permanent monitoring in the field
* if one can expect constant power dissipation or there are short heavy peaks
* the expected accuracy and timing

Without knowing all this....I persinally can only recommend a true junction temperature measurement via a diode.

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top