The Fundamentals of Semiconductor Junction Thermometers

Temperature sensors can be fabricated with semiconductor processing technology by employing the temperature characteristics of the pn junction.  The batch processing and well-defined manufacturing processes associated with semiconductor technology can provide low cost and consistent quality temperature sensors.

Most semiconductor junction temperature sensors utilize a diode-connected bipolar transistor (short-circuited collector-base junction). A constant current passed through the base-emitter junction produces a junction voltage between the base and emitter (V_be) that is a linear function of the absolute temperature. The overall forward voltage drop has a temperature coefficient of approximately 2 mV °C^-1.

Fig: Bipolar transistor configured as a temperature sensor

In the above figure, the base of the transistor is shorted to the collector. A constant current flowing in the remaining pn (base to emitter) junction produces a forward voltage drop V_F proportional to temperature.

The temperature coefficient of a semiconductor sensor is larger but still quite small when compared to a thermocouple or resistive temperature device/detector (RTD). Furthermore, the semiconductor sensor’s forward voltage has an offset that varies significantly from unit to unit. Nonetheless, the semiconductor junction voltage versus temperature is much more linear than that of a thermocouple or RTD. Also, the temperature-sensing element, circuitry is easily integrated to produce a monolithic temperature sensor with an output that can be easily interfaced to a microcontroller and to provide features that are useful in particular applications. For instance, by using an embedded temperature sensor with additional circuitry, protection features can be added to integrated circuits (ICs). A temperature sensor becomes an embedded item in a semiconductor product when it has a secondary or supplemental purpose instead of the primary function.

The Operation of Linear Variable Differential Transformer (LVDT)

 Linear variable differential transformer (LVDT) is an inductive transducer that is commonly used to translate linear motion into electrical signals.

An illustration of LVDT circuit is shown below:

Fig 1.0 LVDT connection circuit

The transformer consists of a single primary winding P and two secondary windings S₁ and S₂ wound on a cylindrical former. A sinusoidal voltage of amplitude 3 to 15 volt and frequency 50 to 20 kHz is employed to excite the primary winding. The two secondary windings have equal number of turns and are identically placed on either side of the primary winding.

The primary winding is connected to an alternating current source. A movable soft-iron core is placed inside the former. The displacement to be measured is applied to the arm attached to the soft iron core. The core is usually made of high permeability, nickel iron. This is slotted longitudinally to reduce eddy current losses. The assembly is placed in a stainless steel housing to provide electrostatic and electromagnetic shielding. The frequency of ac signal applied to primary winding can be between 50 Hz and 20 kHz.

As the primary winding is excited by an alternating current source, it produces an alternating magnetic field which in turn induces alternating voltages in the two secondary windings.

The output voltage of secondary S₁ is ES₁ and that of secondary S₂ is E_S2. In order to convert the outputs from S₁ and S₂ into a single voltage, the two secondary S₁ and S₂ are connected in series opposition. The differential output voltage is:

E₀ = ES₁ – ES₂                                                                        

Operation of LVDT

When the core is at its normal (NULL) position, the flux linking with both the secondary windings is equal and hence equal voltages are induced in them. Therefore at null position: E_S1 = E_S2. Thus, the output voltage E₀ is zero at null position.

If the core is moved to the left of the null position, more flux links with S₁ and less with winding S₂. Correspondingly, output voltages E_S1 is greater than E_S2. The magnitude of output voltage is thus,

 E₀ = E_S1 – E_S2 and we can say, it is in phase with primary voltage.

In the same way, when the core is moved to the right of the null position E_S2 will be more than E_S1. Therefore the output voltage 

E₀ = E_S1 – E_S2 and 180° out of phase with primary voltage.

The amount of voltage change in either secondary winding is proportional to the amount of movement of the core. Thus, we have an indication of amount of linear motion. By noticing whether output voltage is increased or decreased, we can determine the direction of motion.

 Related: Transducers and Sensors

Merits of LVDT

• Output is quite high. Hence, immediate amplification is not necessary.
• Output voltage is step-less and hence the resolution is very good.
• The sensitivity is high (about 40 V/mm).
• It does not load the measured mechanically.
• Linearity is good up to 5 mm of displacement.
• It consumes low power and low hysteresis loss.

The Limitations of LVDT

• It is affected by stray electromagnetic fields. Thus, proper shielding of the device is required.
• LVDT has large threshold.
• The ac inputs generate noise.

Also read: Signal Converters commonly used in Industrial Instrumentation

Don't miss out on key updates, join our newsletter list here.