RobustPCBWhere Robust Design Starts
الحاسباتEnglishUpgradeتسجيل الدخولإنشاء حساب

ما الجديد

سجل التغييرات الكاملالمحرّك 1.18.3

Junction Temperature by Sense Diode

Thermal

Junction temperature from a diode VF reading, and the measured RθJA of your assembly.

هذه الصفحة غير مترجمة بالكامل بعد. الأجزاء غير المترجمة معروضة بالإنجليزية.

المدخلات

Typically 600–750 mV at 1 mA
V
Usually −1.6 to −2.0 mV/°C. Measure it rather than assume it.
mV/°C
V
A
1 mA
W
°C

النتائج

1 ناجحة · 1 للمراجعة
Junction temperature TJ
97.2°C
ناجح

المعيار: TJ ≤ 125 °C; 150 °C is the usual absolute limit for silicon
Measured RθJA

(TJ − TA)/PD at 1 W
72.2°C/W
Resolution
0.56°C/mV
كل النتائج (3)
Rise above ambient
+ 72.2°C
Calibration slope
-1.800mV/°C
VF at 0 °C (intercept)
695mV
Your assembly measures 72.2 °C/W at 1 W. Compare it with the datasheet RθJA.
Cut the heating power to measure, and read VF within a few hundred microseconds, before the die cools.

Calibration line: VF = 0.6950 V − 1.800 mV/°C × T. A 1 mV reading error is 0.56 °C.

The sense current dissipates 520 µW in the junction, about 0.04 °C of self-heating at 72.2 °C/W. Keep Imeas at 1 mA or less, and use the same current for calibration and measurement.

slope -1.80 mV/°C

المسح البارامتري

غيّر مدخلاً واحداً على مدى، وشاهد الإجابة والحكم والهامش عند كل خطوة، جدولاً ومنحنى.Pro

أسوأ الحالات

أضف التفاوتات إلى المدخلات لتحصل على النطاق المضمون حول الإجابة الاسمية، لا الحالة النموذجية وحدها.Pro

Circuit

المبدأ

Any silicon junction is a thermometer. At a fixed small current its forward voltage falls with temperature, roughly linearly, by about 2 mV per degree. Calibrate that line, read VF under load, and you have the junction temperature and the real Rθ of your assembly.

Diode thermometer
VF(T)=VF0+m (T−T0),m≈−1.6  to  −2.0  mV/°CV_F(T) = V_{F0} + m\,(T - T_{0}), \qquad m \approx -1.6 \;\t{to}\; -2.0\;\t{mV/°C}
Typically 600–750 mV at 25 °C and about 1 mA.
Junction temperature from a reading
TJ=T0+VF−VF0mT_J = T_0 + \frac{V_F - V_{F0}}{m}
Two-point calibration
m=VF2−VF1T2−T1VF0=VF1−m T1m = \frac{V_{F2}-V_{F1}}{T_2-T_1} \qquad V_{F0} = V_{F1} - m\,T_1
Extracting thermal resistance
RθJA=TJ−TAPDR_{θJA} = \frac{T_J - T_A}{P_D}
Apply a known power, let it settle, then read VF. This is the Rθ of your assembly, not of a JEDEC test board.
Why the slope exists
VF=nkTqln⁡ ⁣(IFIS(T))V_F = \frac{nkT}{q}\ln\!\left(\frac{I_F}{I_S(T)}\right)
IS rises so fast with temperature that it outweighs the kT/q term, so the slope is negative.
  • VFV_Fforward voltage at the fixed sense current
  • mmtemperature coefficient, negative, in V/°C
  • ImeasI_{meas}sense current, about 1 mA, small enough not to self-heat
  • PDP_Dheating power applied to produce the temperature rise
تفاصيل إضافية

Which diode

Any junction you can reach: a substrate diode, an input protection diode, a base-emitter junction or an output clamp. The die is close to isothermal, so any of them reads close to the temperature you want.

Calibration

  • Cover at least the junction temperature range you expect in service. Do not extrapolate.
  • Use a small, known current, about 1 mA: enough to measure cleanly, too little to heat the junction noticeably.
  • Step the chamber in 25 °C steps, let each settle, and record VF. Six points give a solid line and expose a bad reading.
  • Heat the die with a different structure, and read the diode. That keeps the thermometer separate from the heater.

Cautions

Give it time. A board takes 10 to 15 minutes to settle. Read early and you under-report the rise.
Avoid IR drop. The heating current shares your ground. Interrupt the heating power to read, and use four-wire (Kelvin) sensing.
Above about 2 W, spread the heating across several devices rather than one saturated output.

Where the datasheet gives ΨJT, a thermocouple on the package top is usually a faster route to TJ.

Method adapted from Allegro MicroSystems AN295014 Rev. 1, on measuring IC temperature with a sense diode. It reports about −1.8 mV/°C from 25 °C to 175 °C. The explanations are our own.

إصدار المحرك ⁨1.18.3⁩