7.1 Overview

The LM26LV and LM26LV-Q1 are precision, dual-output, temperature switches with analog temperature sensor output. The trip temperature (T_TRIP) is factory selected by the order number. The V_TEMP class AB analog output provides a voltage that is proportional to temperature. The LM26LV and LM26LV-Q1 include an internal reference DAC (TEMP THRESHOLD), analog temperature sensor and analog comparator. The reference DAC is connected to one of the comparator inputs. The reference DAC output voltage (V_TRIP) is preprogrammed by TI. The result of the reference DAC voltage and the temperature sensor output comparison is provided on two output pins OVERTEMP and OVERTEMP.

The V_TEMP output has a programmable gain. The output gain has 4 possible settings as described in Table 1. The gain setting is dependent on the temperature trip point selected.

Built-in temperature hysteresis (T_HYST) prevents the digital outputs from oscillating. The OVERTEMP and OVERTEMP activates when the die temperature exceeds T_TRIP and releases when the temperature falls below a temperature equal to T_TRIP minus T_HYST. OVERTEMP is active-high with a push-pull structure. OVERTEMP, is active-low with an open-drain structure. The comparator hysteresis is fixed at 5°C.

Driving the TRIP-TEST high activates the digital outputs. A processor can check the logic level of the OVERTEMP or OVERTEMP, confirming that they changed to their active state. This allows for system production testing verification that the comparator and output circuitry are functional after system assembly. When the TRIP-TEST pin is high, the trip-level reference voltage appears at the V_TEMP pin. Tying OVERTEMP to TRIP-TEST latches the output after it trips. It can be cleared by forcing TRIP-TEST low or powering off the LM26LV or LM26LV-Q1.

7.2 Functional Block Diagram

7.3 Feature Description

7.3.2 V_TEMP vs Die Temperature Approximations

The LM26LV's and LM26LV-Q1's V_TEMP analog temperature output is very linear. Table 1 and the equation in The Second-Order Equation (Parabolic) represent the most accurate typical performance of the V_TEMP voltage output versus temperature.

7.3.2.1 The Second-Order Equation (Parabolic)

The data from Table 1, or Equation 1, when plotted, has an umbrella-shaped parabolic curve. V_TEMP is in mV.

Equation 1.

7.3.2.2 The First-Order Approximation (Linear)

For a quicker approximation, although less accurate than the second-order, over the full operating temperature range the linear formula below can be used. Using Equation 2, with the constant and slope in the following set of equations, the best-fit V_TEMP versus die temperature performance can be calculated with an approximation error less than 18 mV. V_TEMP is in mV.

Equation 2.

7.3.2.3 First-Order Approximation (Linear) Over Small Temperature Range

For a linear approximation, a line can easily be calculated over the desired temperature range from Table 1 using the two-point equation:

Equation 3.

where

• V is in mV
• T is in °C
• T₁ and V₁ are the coordinates of the lowest temperature
• T₂ and V₂ are the coordinates of the highest temperature

For example, to determine the equation of a line with GAIN4, with a temperature from 20°C to 50°C, proceed using Equation 4, Equation 5, and Equation 6:

Equation 4.

Equation 5. V – 2396 mV = –12.8 mV/°C × (T – 20°C)

Equation 6. V = –12.8 mV/°C × (T – 20°C) + 2396 mV

Using this method of linear approximation, the transfer function can be approximated for one or more temperature ranges of interest.

7.3.3 OVERTEMP and OVERTEMP Digital Outputs

The OVERTEMP active high, push-pull output and the OVERTEMP active low, open-drain output both assert at the same time whenever the die temperature reaches the factory preset temperature trip point. They also assert simultaneously whenever the TRIP_TEST pin is set high. Both outputs deassert when the die temperature goes below the temperature trip point hysteresis. These two types of digital outputs enable the user the flexibility to choose the type of output that is most suitable for his design.

Either the OVERTEMP or the OVERTEMP digital output pins can be left open if not used.

7.3.3.1 OVERTEMP Open-Drain Digital Output

The OVERTEMP active low, open-drain digital output, if used, requires a pullup resistor between this pin and V_DD. The following section shows how to determine the pullup resistor value.

7.3.3.1.1 Determining the Pullup Resistor Value

The pullup resistor value is calculated at the condition of maximum total current, I_T, through the resistor. The total current is:

Equation 7. I_T = I_L + I_SINK

where

• I_T is the maximum total current through the pullup resistor at V_OL.
• I_L is the load current, which is very low for typical digital inputs.

The pullup resistor maximum value can be found by using Equation 8.

Equation 8.

where

• V_DD(MAX) is the maximum power supply voltage to be used in the customer's system.
• V_OUT is the Voltage at the OVERTEMP pin. Use V_OL for calculating the pullup resistor.

7.3.3.1.1.1 Example Calculation

Suppose, for this example, a V_DD of 3.3 V ± 0.3 V, a CMOS digital input as a load, a V_OL of 0.2 V.

• For V_OL of 0.2 V the electrical specification for OVERTEMP shows a maximum I_SINK of 385 µA.
• Let I_L = 1 µA, then I_T is about 386 µA maximum. If 35 µA is selected as the current limit then I_T for the calculation becomes 35 µA.
• V_DD(MAX) is 3.3 V + 0.3 V = 3.6 V, then calculate the pullup resistor as R_PULLUP = (3.6 – 0.2) / 35 µA = 97 kΩ.
• Based on this calculated value, select the closest resistor value in the tolerance family used.

In this example, if 5% resistor values are used, then the next closest value is 100 kΩ.

7.3.5 V_TEMP Analog Temperature Sensor Output

The V_TEMP push-pull output provides the ability to sink and source significant current. This is beneficial when, for example, driving dynamic loads like an input stage on an analog-to-digital converter (ADC). In these applications the source current is required to quickly charge the input capacitor of the ADC. See Application and Implementation for more discussion of this topic. The LM26LV and LM26LV-Q1 are ideal for applications which require strong source or sink current.

7.3.5.1 Noise Considerations

The LM26LV's and LM26LV-Q1's supply-noise rejection (the ratio of the AC signal on V_TEMP to the AC signal on V_DD) was measured during bench tests. The device's typical attenuation is shown in Typical Characteristics. A load capacitor on the output can help to filter noise.

For operation in very noisy environments, some bypass capacitance must be present on the supply within approximately 2 inches of the LM26LV or LM26LV-Q1.

7.3.5.2 Capacitive Loads

The V_TEMP Output handles capacitive loading well. In an extremely noisy environment, or when driving a switched sampling input on an ADC, it may be necessary to add some filtering to minimize noise coupling. Without any precautions, the V_TEMP can drive a capacitive load less than or equal to 1100 pF as shown in Figure 20. For capacitive loads greater than 1100 pF, a series resistor is required on the output, as shown in Figure 21, to maintain stable conditions.

Figure 20. LM26LV or LM26LV-Q1 No Decoupling Required for Capacitive Loads Less Than 1100 pF.

Figure 21. LM26LV or LM26LV-Q1 With Series Resistor for Capacitive Loading Greater Than 1100 pF

Table 2. Minimum Series Resistence for Capacitive Loads

CLOAD	MINIMUM RS
1.1 nF to 99 nF	3 kΩ
100 nF to 999 nF	1.5 kΩ
1 µF	800 Ω

7.3.5.3 Voltage Shift

The LM26LV and LM26LV-Q1 are very linear over temperature and supply voltage range. Due to the intrinsic behavior of an NMOS/PMOS rail-to-rail buffer, a slight shift in the output can occur when the supply voltage is ramped over the operating range of the device. The location of the shift is determined by the relative levels of V_DD and V_TEMP. The shift typically occurs when V_DD – V_TEMP = 1 V.

This slight shift (a few millivolts) takes place over a wide change (approximately 200 mV) in V_DD or V_TEMP. Because the shift takes place over a wide temperature change of 5°C to 20°C, V_TEMP is always monotonic. The accuracy specifications Accuracy Characteristics already includes this possible shift.

7.4 Device Functional Modes

The LM26LV and LM26LV-Q1 have several modes of operation as detailed in the following drawings.

Figure 22. Temperature Switch Using Push-Pull Output

Figure 24. TRIP_TEST Digital Output Test Circuit

Figure 23. Temperature Switch Using Open-Drain Output

The TRIP_TEST pin, normally used to check the operation of the OVERTEMP and OVERTEMP pins, may be used to latch the outputs whenever the temperature exceeds the programmed limit and causes the digital outputs to assert. As shown in Figure 25, when OVERTEMP goes high the TRIP_TEST input is also pulled high and causes OVERTEMP output to latch high and the OVERTEMP output to latch low. The latch can be released by either momentarily pulling the TRIP_TEST pin low (GND), or by toggling the power supply to the device. The resistor limits the current out of the OVERTEMP output pin.

Figure 25. Latch Circuit Using OVERTEMP Output

Alternately, the circuit in Figure 25 the 100 kΩ can be replaced with a short and the momentary reset switch may be removed. In this configuration, when OVERTEMP goes active high, it drives TRIP_TEST high. THRIP TEST high causes OVERTEMP to stay high. It is therefore latched. To release the latch, power down, then power up the LM26LV or LM26LV-Q1. The LM26LV and LM26LV-Q1 always come up in a released condition.