EVALUATION KIT AVAILABLE

MAX31856 Precision Thermocouple to Digital Converter

with Linearization

General Description Benefits and Features

The MAX31856 performs cold-junction compensation ●● Provides High-Accuracy Thermocouple Temperature
and digitizes the signal from any type of thermocouple. Readings
The output data is formatted in degrees Celsius. This • Includes Automatic Linearization Correction for 8
converter resolves temperatures to 0.0078125°C, allows Thermocouple Types
readings as high as +1800°C and as low as -210°C • ±0.15% (max, -20°C to +85°C) Thermocouple Full-
(depending on thermocouple type), and exhibits thermo- Scale and Linearity Error
couple voltage measurement accuracy of ±0.15%. The • 19-Bit, 0.0078125°C Thermocouple Temperature
thermocouple inputs are protected against overvoltage Resolution
conditions up to ±45V.
●● Internal Cold-Junction Compensation Minimizes
A lookup table (LUT) stores linearity correction data for System Components
several types of thermocouples (K, J, N, R, S, T, E, and • ±0.7°C (max, -20°C to +85°C) Cold-Junction
B). Line frequency filtering of 50Hz and 60Hz is included, Accuracy
as is thermocouple fault detection. A SPI-compatible inter-
●● ±45V Input Protection Provides Robust System
face allows selection of thermocouple type and setup of
Performance
the conversion and fault detection processes.
●● Simplifies System Fault Management and
Applications Troubleshooting
●● Temperature Controllers • Detects Open Thermocouples
●● Industrial Ovens, Furnaces, and Environmental • Over- and Undertemperature Fault Detection
Chambers ●● 50Hz/60Hz Noise Rejection Filtering Improves
●● Industrial Equipment System Performance
Ordering Information appears at end of data sheet. ●● 14-Pin TSSOP Package

Typical Application Circuit

AGND DGND

BIAS FAULT
0.01µF

SDI
T-
0.1µF MAX31856

T+ SDO
0.01µF
TO MICROCONTROLLER
3.3V AVDD SCK
0.1µF

DNC CS

DRDY DVDD 3.3V

0.1µF

19-7534; Rev 0; 2/15

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Absolute Maximum Ratings

AVDD, DVDD........................................................-0.3V to +4.0V Operating Temperature Range.......................... -55°C to +125°C
T+, T-, Bias...........................................................................±45V Junction Temperature.......................................................+150°C
T+, T-, Bias........................................................................±20mA Storage Temperature Range ............................ -65°C to +150°C
All Other Pins........................................-0.3V to (VDVDD + 0.3V) Lead Temperature (soldering, 10s).................................. +300°C
Continuous Power Dissipation (TA = +70°C) Soldering Temperature
TSSOP (derate 9.1mW/°C above +70°C)...................727.3mW (reflow) .................See IPC/JEDEC J-STD-020A Specification
ESD Protection (All pins, Human Body Model)..................2000V
Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these
or any other conditions beyond those indicated in the operational sections of the specifications is not implied. Exposure to absolute maximum rating conditions for extended periods may affect
device reliability.

Package Thermal Characteristics (Note 1)

TSSOP
Junction-to-Ambient Thermal Resistance (θJA).........110°C/W Junction-to-Case Thermal Resistance (θJC)................30°C/W

Note 1: Package thermal resistances were obtained using the method described in JEDEC specification JESD51-7, using a four-layer
board. For detailed information on package thermal considerations, refer to www.maximintegrated.com/thermal-tutorial.

Recommended DC Operating Conditions

(TA = -55°C to +125°C, unless otherwise noted.)(Notes 2 and 4)

PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS

VAVDD,
Power-Supply Voltage 3.0 3.3 3.6 V
VDVDD
AVDD-DVDD -100 +100 mV
Cable Resistance RCABLE Per lead 40 kΩ
Input Logic 0 VIL 0.8 V
Input Logic 1 VIH 2.1 V

Electrical Characteristics
(3.0V ≤ VDD ≤ 3.6V, TA = -55°C to +125°C, unless otherwise noted.)(Notes 2, 3, and 4)

PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS

Standby 5.25 10 µA
Supply Current IDD
Active conversion 1.2 2 mA
Thermocouple Temperature 19 Bits
Resolution 0.0078125 °C
Cold-Junction Temperature Data
0.015625 °C
Resolution
TA = +25°C -10 +10
TA = -40°C to +85°C -10 +65
Thermocouple Input Bias Current ITCBIAS nA
TA = -55°C to +105°C -20 +110
TA = -55°C to +125°C -20 +400

www.maximintegrated.com Maxim Integrated │  2

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Electrical Characteristics (continued)

(3.0V ≤ VDD ≤ 3.6V, TA = -55°C to +125°C, unless otherwise noted.)(Notes 2, 3, and 4)

PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS

TA = +25°C ±0.2

Thermocouple Input Differential TA = -40°C to +85°C -4 +4

ITCIDBIAS nA
Bias Current (Note 4) TA = -55°C to +105°C -5.5 +5.5
TA = -55°C to +125°C -10 +10
AV = 8 1.3
Input-Referred Noise VN µVRMS
AV = 32 0.4
Power-Supply Rejection PSR Cold-junction sensor 0.15 °C/V
Power-On-Reset Voltage
VPOR 2.7 2.85 V
Threshold
Power-On-Reset Voltage
VHYST 0.1 V
Hysteresis
Bias Voltage VBIAS 0.735 V
BIAS Output Resistance RBIAS 2 kΩ
Input Common-Mode Range 0.5 1.4 V
TA = +25°C -0.05 +0.05
TA = -20°C to +85°C -0.15 +0.15
Full-Scale and INL Error (Note 6) TA = -40°C to +105°C -0.2 +0.2 %FS
TA = -40°C to +125°C -0.3 +0.3
TA = -55°C to +125°C -0.35 +0.35
TA = +25°C -0.01 +0.01
TA = -20°C to +85°C -0.015 +0.015
Input Offset Voltage (Note 7) %FS
TA = -40°C to +105°C -0.017 +0.017
TA = -55°C to +125°C -0.02 +0.02
TA = +25°C -7.8 +7.8
TA = -20°C to +85°C -11.7 +11.7
AV = 8
TA = -40°C to +105°C -13.3 +13.3
TA = -55°C to +125°C -15.6 +15.6
Input Offset Voltage µV
TA = +25°C -2.0 +2.0
TA = -20°C to +85°C -2.9 +2.9
AV = 32
TA = -40°C to +105°C -3.3 +3.3
TA = -55°C to +125°C -3.9 +3.9
TA = -20°C to +85°C -0.7 +0.7
Cold-Junction Temperature Error TA = -40°C to +105°C -1 +1 °C
TA = -55°C to +125°C -2 +2
Overvoltage Rising Threshold VAVDD – VAVDD + VAVDD
V
(Note 8) 0.1 0.17 + 0.35
Overvoltage Hysteresis 0.09 V

www.maximintegrated.com Maxim Integrated │  3

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Electrical Characteristics (continued)

(3.0V ≤ VDD ≤ 3.6V, TA = -55°C to +125°C, unless otherwise noted.)(Notes 2, 3, and 4)

PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS

Undervoltage Falling Edge
-0.3 -0.17 0 V
Threshold (Note 8)
Undervoltage Hysteresis 0.09 V
Type B,
TA = 0 to 125°C, -0.24 +0.25
TTC = 95°C to +1798°C
Type E,
TA = -55°C to +125°C -0.14 +0.06
TTC = -200°C to +1000°C
Type J,
TA = -55°C to +125°C -0.11 +0.10
TTC = -210°C to +1200°C
Type K,
TA = -55°C to +125°C -0.13 +0.12
Thermocouple Linearity TTC = -200°C to +1372°C
°C
Correction Error Type N,
TA = -55°C to +125°C -0.09 +0.08
TTC = -200°C to +1300°C
Type R,
TA = -50°C to +125°C -0.19 +0.17
TTC = -50°C to +1768°C
Type S,
TA = -50°C to +125°C -0.16 +0.20
TTC = -50°C to +1768°C
Type T,
TA = -55°C to +125°C -0.07 +0.07
TTC = -200°C to +400°C

1-Shot conversion or first

conversion in auto-conversion 143 155
mode (60Hz)

1-Shot conversion or first

Temperature Conversion Time conversion in auto-conversion 169 185
tCONV mode (50Hz) ms
(Thermocouple + Cold Junction)

Auto conversion mode,

82 90
conversions 2 through n (60Hz)

Auto conversion mode,

98 110
conversions 2 through n (50Hz)

www.maximintegrated.com Maxim Integrated │  4

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Electrical Characteristics (continued)

(3.0V ≤ VDD ≤ 3.6V, TA = -55°C to +125°C, unless otherwise noted.)(Notes 2, 3, and 4)

PARAMETER SYMBOL CONDITIONS MIN TYP MAX UNITS

Common-Mode Rejection CMR 0.5V ≤ VCM ≤ 1.4V 70 dB
50/60Hz Noise Rejection Fundamental and harmonics 91 dB
SERIAL INTERFACE
Input Leakage Current ILEAK (Note 5) -1 +1 µA
Output High Voltage VOH IOUT = -1.6mA VCC - 0.4 V
Output Low Voltage VOL IOUT = 1.6mA 0.4 V
Input Capacitance CIN 8 pF
Serial Clock Frequency fSCL 5 MHz
SCK Pulse High Width tCH 100 ns
SCK Pulse Low Width tCL 100 ns
SCK Rise and Fall Time tR, tF CL = 10pF 200 ns
CS Fall to SCK Rise tCC CL = 10pF 100 ns
SCK to CS Hold tCCH CL = 10pF 100 ns
CS Rise to Output Disable tCDZ CL = 10pF 40 ns
Data to SCLK Setup tDC 35 ns
SCLK to Data Hold tCDH 35 ns
SCK Fall to Output Data Valid tCDD CL = 10pF 80 ns
CS Inactive Time tCWH (Note 3) 400 ns

Note 2: All voltages are referenced to GND. Currents entering the IC are specified positive, and currents exiting the IC are negative.
Note 3: All Serial Interface timing specifications are guaranteed by design.
Note 4: Specification is 100% tested at TA = +25°C. Specification limits over temperature (TA = TMIN to TMAX) are guaranteed by
design and characterization; not production tested.
Note 5: For all pins except T+ and T- (see the Thermocouple Input Bias Current parameter in the Electrical Characteristics table.
Note 6: Using a common-mode voltage other than VBIAS will change this specification. See the Typical Operating Characteristics
for details.
Note 7: Input-referred full-scale voltage is 78.125mV when AV = 8 and is 19.531mV when AV = 32.
Note 8: Overvoltage and undervoltage limits apply to T+, T-, and BIAS pins.

www.maximintegrated.com Maxim Integrated │  5

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

CS

tCC

SCLK

tCDD tCDD
tCDH
tDC

SDI A7 A6 A0

tCDZ
SDO
D7 D6 D1 D0

WRITE ADDRESS BYTE READ DATA BYTE

NOTE: SCLK CAN BE EITHER POLARITY, TIMING SHOWN FOR CPOL = 1.

Figure 1. Timing Diagram: SPI Read Data Transfer

CS

tCWH
tCC
tR
tCL tF tCCH

SCLK
tCDH
tCH
tCDH
tDC

SDI A7 A6 A0 D7 D0

WRITE ADDRESS BYTE WRITE DATA BYTE

NOTE: SCLK CAN BE EITHER POLARITY, TIMING SHOWN FOR CPOL = 1.

Figure 2. Timing Diagram: SPI Write Data Transfer

www.maximintegrated.com Maxim Integrated │  6

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Typical Operating Characteristics

(VCC = 3.3V and TA = +25°C, unless otherwise noted.)

ACTIVE SUPPLY CURRENT STANDBY SUPPLY CURRENT THERMOCOUPLE INPUT BIAS CURRENT
vs. TEMPERATURE toc01
vs. TEMPERATURE toc02
vs. TEMPERATURE toc03
1.5 10 250
1.45 VDD = 3.0V, 3.3V, 3.6V
VDD = 3.6V 9 200
1.4
ACTIVE CURRENT (mA)

INPUT BIAS CURRENT (nA)

STANDBY CURRENT (μA)

1.35 VDD = 3.6V

8 150
1.3
1.25 7 VDD = 3.3V 100
1.2
VDD = 3.3V 6 50
1.15
1.1
5 0
1.05 VDD = 3.0V
VDD = 3.0V
1 4 -50
-55 -35 -15 5 25 45 65 85 105 125 -55 -35 -15 5 25 45 65 85 105 125 -55 -35 -15 5 25 45 65 85 105 125
TEMPERATURE (°C) TEMPERATURE (°C) TEMPERATURE (°C)

THERMOCOUPLE INPUT THERMOCOUPLE INPUT

DIFFERENTIAL BIAS CURRENT DIFFERENTIAL BIAS CURRENT
vs. TEMPERATURE vs. TEMPERATURE toc05
toc04
5 5
Differential Input Voltage = 0V Differential Input Voltage = 0.1V
4 4 VDD = 3.0V
DIFFERENTIAL BIAS CURRENT (nA)
DIFFERENTIAL BIAS CURRENT (nA)

VDD = 3.0V, 3.3V, 3.6V

3 3 VDD = 3.3V

VDD = 3.6V
2 2

1 1

0 0

-1 -1
-55 -35 -15 5 25 45 65 85 105 125 -55 -35 -15 5 25 45 65 85 105 125
TEMPERATURE (°C) TEMPERATURE (°C)

AV = 8 FULL-SCALE ERROR AV = 32 FULL-SCALE ERROR

vs. TEMPERATURE toc06
vs. TEMPERATURE toc07
0.15 0.15

0.1 VDD = 3.0V 0.1

VDD = 3.0V
VDD = 3.3V
FULL-SCALE ERROR (%)

FULL-SCALE ERROR (%)

0.05 0.05

0 0

-0.05 VDD = 3.6V -0.05 VDD = 3.3V

VDD = 3.6V

-0.1 -0.1

-0.15 -0.15
-55 -35 -15 5 25 45 65 85 105 125 -55 -35 -15 5 25 45 65 85 105 125
TEMPERATURE (°C) TEMPERATURE (°C)

www.maximintegrated.com Maxim Integrated │  7

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Typical Operating Characteristics (continued)

(VCC = 3.3V and TA = +25°C, unless otherwise noted.)

AV = 8 OFFSET ERROR AV = 32 OFFSET ERROR

vs. TEMPERATURE toc08
vs. TEMPERATURE toc09
10 10
8 8 VDD = 3.0V, 3.3V, 3.6V

6 VDD = 3.6V 6
4
OFFSET ERROR (μV)

VDD = 3.3V 4

OFFSET ERROR (μV)

2 2
0 0
-2 -2
-4 -4
-6 -6
VDD = 3.0V
-8 -8
-10 -10
-55 -35 -15 5 25 45 65 85 105 125 -55 -35 -15 5 25 45 65 85 105 125
TEMPERATURE (°C) TEMPERATURE (°C)

FULL-SCALE ERROR COLD-JUNCTION TEMPERATURE

vs. COMMON-MODE VOLTAGE ERROR vs. TEMPERATURE
toc10 toc11
0.08 0.2

0.06 0.15
VDD = 3.6V
COLD-JUNCTION ERROR (°C)

0.04 0.1
FULL-SCALE ERROR (%)

0.02 0.05
VDD = 3.6V
0 0
VDD = 3.3V
-0.02 -0.05
VDD = 3.0V
-0.04 -0.1 VDD = 3.3V

-0.06 -0.15
VDD = 3.0V
-0.08 -0.2
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 -50 0 50 100
COMMON-MODE VOLTAGE (V) TEMPERATURE (°C)

B-TYPE LINEARIZATION ERROR E-TYPE LINEARIZATION ERROR

vs. THERMOCOUPLE TEMPERATURE vs. THERMOCOUPLE TEMPERATURE
toc12 toc13
0.5 0.5
CJ Temp = 0°C, 25°C, 85°C,125°C CJ Temp = -40°C, 25°C, 85°C,125°C
0.4 0.4
0.3 0.3
LINEARIZATION ERROR (°C)
LINEARIZATION ERROR (°C)

0.2 0.2
0.1 0.1
0 0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-200 200 600 1000 1400 1800 -200 200 600 1000 1400 1800
THERMOCOUPLE TEMPERATURE (°C) THERMOCOUPLE TEMPERATURE (°C)

www.maximintegrated.com Maxim Integrated │  8

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Typical Operating Characteristics (continued)

(VCC = 3.3V and TA = +25°C, unless otherwise noted.)

J-TYPE LINEARIZATION ERROR K-TYPE LINEARIZATION ERROR

vs. THERMOCOUPLE TEMPERATURE vs. THERMOCOUPLE TEMPERATURE
toc14 toc15
0.5 0.5
CJ Temp = -40°C, 25°C, 85°C,125°C CJ Temp = -40°C, 25°C, 85°C,125°C
0.4 0.4

0.3 0.3

LINEARIZATION ERROR (°C)

LINEARIZATION ERROR (°C)

0.2 0.2

0.1 0.1

0 0

-0.1 -0.1

-0.2 -0.2

-0.3 -0.3

-0.4 -0.4

-0.5 -0.5
-200 200 600 1000 1400 1800 -200 200 600 1000 1400 1800
THERMOCOUPLE TEMPERATURE (°C) THERMOCOUPLE TEMPERATURE (°C)

N-TYPE LINEARIZATION ERROR R-TYPE LINEARIZATION ERROR

vs. THERMOCOUPLE TEMPERATURE vs. THERMOCOUPLE TEMPERATURE
toc16 toc17
0.5 0.5
CJ Temp = -40°C, 25°C, 85°C,125°C CJ Temp = -40°C, 25°C, 85°C,125°C
0.4 0.4
0.3 0.3
LINEARIZATION ERROR (°C)
LINEARIZATION ERROR (°C)

0.2 0.2
0.1 0.1
0 0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-200 200 600 1000 1400 1800 -200 200 600 1000 1400 1800
THERMOCOUPLE TEMPERATURE (°C) THERMOCOUPLE TEMPERATURE (°C)

S-TYPE LINEARIZATION ERROR T-TYPE LINEARIZATION ERROR

vs. THERMOCOUPLE TEMPERATURE vs. THERMOCOUPLE TEMPERATURE
toc18 toc19
0.5 0.5
CJ Temp = -40°C, 25°C, 85°C,125°C CJ Temp = -40°C, 25°C, 85°C,125°C
0.4 0.4
0.3 0.3
LINEARIZATION ERROR (°C)
LINEARIZATION ERROR (°C)

0.2 0.2
0.1 0.1
0 0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-200 200 600 1000 1400 1800 -200 200 600 1000 1400 1800
THERMOCOUPLE TEMPERATURE (°C) THERMOCOUPLE TEMPERATURE (°C)

www.maximintegrated.com Maxim Integrated │  9

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Pin Configuration

TOP VIEW
+
AGND 1 14 DGND

BIAS 2 13 FAULT

T- 3 12 SDI
MAX31856
T+ 4 11 SDO

AVDD 5 10 SCK

DNC 6 9 CS

DRDY 7 8 DVDD

TSSOP

Pin Description
PIN NAME FUNCTION
1 AGND Analog Ground
2 BIAS Bias Voltage Source. Nominally 0.735V. This pin is floating when no conversions are taking place.
3 T- Thermocouple Negative Input. See Table 1.
4 T+ Thermocouple Positive Input. See Table 1.
5 AVDD Analog Positive Supply. Bypass with a 0.1µF capacitor to AGND.
6 DNC Do Not Connect
7 DRDY Data Ready Output
8 DVDD Digital Positive Supply. Bypass with a 0.1µF capacitor to DGND.
9 CS Chip Select. Set CS low to enable the serial interface.
10 SCK Serial Clock Input
11 SDO Serial Data Output
12 SDI Serial Data Input
13 FAULT Cable, thermocouple, or temperature fault output
14 DGND Digital Ground

www.maximintegrated.com Maxim Integrated │  10

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Block Diagram

T- INPUT
LINEARIZATION AND CONTROL
PROTECTION 19-BIT
PGA COLD-JUNCTION AND
AND FAULT ADC
T+ COMPENSATION INTERFACE
DETECTION

TEMPERATURE
SENSOR

Detailed Description tion code are summed to produce the code corresponding
The MAX31856 is a sophisticated thermocouple-to-digital to the cold-junction compensated thermocouple tempera-
converter with a built-in 19-bit analog-to-digital converter ture. Finally, the LUT is used to produce a cold-junction
(ADC). Internal functions include correction for thermo- compensated output code in units of °C.
couple nonlinearity, input protection, cold-junction com- Thermocouple Voltage Conversion
pensation sensing and correction, a digital controller, a
T+ and T- are the thermocouple inputs. T- is biased to
SPI-compatible interface, and associated control logic.
approximately 0.735V by the BIAS output. The amplifier
In the simplest configuration, the thermocouple wires con- provides gain to the μV- and mV-level thermocouple sig-
nect directly to inputs T- and T+, with a common-mode nals to make the amplitude appropriate for the ADC’s full-
bias voltage provided by the BIAS output. Additional scale input range. Two amplifier gains provide full-scale
filtering and/or protection components may be added if input ranges of ±78.125mV and ±19.531mV to accommo-
needed, as discussed in the Applications Information date higher- and lower-sensitivity thermocouples.
section. Operation is controlled by two configuration bytes
Because long thermocouple wires may pick up noise
and four bytes that contain over- and undertemperature
from a variety of sources, including AC power cables, the
detection thresholds.
amplified signal is lowpass filtered before being applied
Temperature Conversion to the ADC. The ADC provides further digital lowpass and
The temperature conversion process consists of five notch filtering to attenuate input noise. The notch frequen-
steps as described in the sections below. The input ampli- cies are either 50Hz and its harmonics or 60Hz and its
fier and ADC amplify and digitize the thermocouple’s volt- harmonics, selectable using bit 0 of the Configuration 0
age output. The internal temperature sensor measures register (00h). In addition, bits D6:4 of the Configuration
the cold-junction temperature. Using the internal lookup 1 register (01h) enable an averaging mode that provides
table (LUT), the ADC code corresponding to the cold- additional filtering with an associated increase in conver-
junction temperature for the selected thermocouple type sion time. 2, 4, 8, or 16 samples may be averaged using
is determined. The thermocouple code and the cold-junc- this mode.

www.maximintegrated.com Maxim Integrated │  11

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

The conversion mode can be either continuous or “nor- the cold junction must be measured. This is done with the
mally off”, as selected by bit 7 of the Configuration 0 internal precision temperature sensor, which has accuracy
register (00h). When in the normally off mode, a single better than ±0.7°C from -20°C to +85°C. By placing the
“1-shot” conversion may be selected using bit 6 of the MAX31856 near the cold junction, the cold-junction tem-
Configuration 0 register (00h). perature can be measured and used to compensate for
Thermocouple type is user-selectable using bits D3:0 of cold-junction effects.
the Configuration 1 register (01h). Thermocouple types The MAX31856 stores the cold-junction temperature data
K, J, N, R, S, T, B, and E are supported by automatic in registers 0Ah and 0Bh. When the cold-junction tem-
cold-junction compensation and linearization. (To use perature sensor is enabled, these registers are read-only
a different thermocouple type, use bits D3:0 to select a and contain the measured cold-junction temperature plus
gain of either 8 or 32. The linearization and cold-junction the value in the Cold-Junction Offset register. Reading
compensation calculations may then be done externally the register with the cold-junction temperature sensor
using the cold-junction temperature and thermocouple enabled will reset the DRDY pin high. Both bytes of this
voltage data.) register should be read as a multibyte transfer to ensure
both bytes are from the same temperature update. When
Cold-Junction Temperature Sensing the cold-junction temperature sensor is disabled, these
The function of the thermocouple is to sense a difference in registers become read-write registers that contain the
temperature between two ends of the thermocouple wires. most recent measured temperature value. If desired, data
The thermocouple‘s sensing junction (often called the “hot” from an external temperature sensor may be written to
junction regardless of its temperature) can be measured these registers when the internal cold-junction sensor
across its rated operating temperature range (see Table 1 is disabled. The maximum cold-junction temperature is
for supported thermocouple temperature ranges). clamped at 128°C and the minimum is clamped at -64°C.
Additional thermocouples are created where the thermo- See Table 2 for the Reference Junction (Cold Junction)
couple wires make contact with different metals, usually Temperature Data Format.
at a connector or at the point where they are soldered to If desired, a temperature offset may be written to the
a PCB (the “cold junction”). To compensate for the errors Cold-Junction Offset register (09h). The value stored in
due to these additional thermocouples, the temperature at registers 0Ah and 0Bh will then be equal to the measured

Table 1. Supported Thermocouples and Temperature Ranges

NOMINAL COLD-JUNCTION
TYPE T-WIRE T+ WIRE TEMP RANGE
SENSITIVITY (µV/°C) TEMP RANGE
10.086
B Platinum/Rhodium Platinum/Rhodium 250oC to 1820oC 0 to 125°C
(+500°C to +1500°C)
76.373
E Constantan Chromel -200°C to +1000°C -55°C to +125°C
(0°C to +1000°C)
57.953
J Constantan Iron -210°C to +1200°C -55°C to +125°C
(0°C to + 750°C)
41.276
K Alumel Chromel -200°C to +1372°C -55°C to +125°C
(0°C to + 1000°C)
36.256
N Nisil Nicrosil -200°C to +1300°C -55°C to +125°C
(0°C to +1000°C)
10.506
R Platinum Platinum/Rhodium -50°C to +1768°C -50°C to +125°C
(0°C to +1000°C)
9.587
S Platinum Platinum/Rhodium -50°C to +1768°C -50°C to +125°C
(0°C to +1000°C)
52.18
T Constantan Copper -200°C to +400°C -55°C to +125°C
(0°C to +400°C)

www.maximintegrated.com Maxim Integrated │  12

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

value plus the offset value. The MSB of the offset register Thermocouple Linearization and Conversion
is 4°C and the LSB is 0.0625°C. The resulting range of of Code to Temperature
the offset value applied to the measured CJ temperature Because all thermocouples are nonlinear, the raw cold-
is -8°C to +7.9375°C. The default offset value is 0°C junction-compensated value must be corrected for non-
(00h). linearity and converted to a temperature value. This
Optimal performance is achieved when the thermocouple is done using the LUT to produce the linearized and
cold junction and the cold-junction sensor are at the same cold-junction-compensated temperature value, which is
temperature. Avoid placing heat-generating devices or stored after every conversion as 19 bits in the Linearized
components near the cold junction because this may pro- Thermocouple Temperature registers (0Ch, 0Dh, and
duce cold-junction-related errors. When a significant tem- 0Eh). All three bytes should be read as a multibyte trans-
perature differential between the internal sensor and the fer to ensure all are from the same data update. See
cold junction is unavoidable, an external temperature sen- Table 3 for the Linearized Thermocouple Temperature
sor may be used instead. The temperature measured by Data Format.
the external sensor may be written to the cold-junction tem- Linearization accuracy varies by thermocouple type, “hot-
perature register and used for cold-junction compensation. junction” temperature, and cold-junction temperature, with
Bit 3 of Configuration 0 register (00h) disables the internal the largest errors typically occurring near the hot-junction
cold-junction temperature sensor and allows temperature and cold-junction extremes. Worst-case values for linear-
values from an external sensor to be written directly into ization errors are shown in the Electrical Characteristics
the Cold-Junction Temperature registers (0Ah and 0Bh). table.
Cold-Junction Temperature Translation and Over-/Undertemperature Fault Detection
Compensation Over- and undertemperature fault detection are available
Thermocouple temperature values and corresponding for both the cold-junction temperature and the linearized
ADC codes are stored in an internal lookup table. After and cold-junction-compensated temperature reading. Two
measuring the cold-junction temperature, the LUT is registers (03h and 04h) contain the high and low thresh-
used to convert the temperature value to the equivalent olds for the cold-junction temperature. The cold-junction
ADC code for the type of thermocouple being used. temperature value in registers 0Ah and 0Bh is compared
Values between LUT entries are interpolated. The cold- to the threshold values. If a threshold is exceeded, the
junction ADC code is added to the conversion result in corresponding bit is set in the Fault Status register (0Fh)
the thermocouple voltage register to yield a cold-junction- and, if not masked, the FAULT output will assert.
compensated value.
Table 2. Reference Junction Table 3. Linearized Thermocouple
(Cold-Junction) Temperature Data Format Temperature Data Format
TEMPERATURE (°C) DIGITAL OUTPUT TEMPERATURE (°C) DIGITAL OUTPUT
+127.984375 0111 1111 1111 1100 +1600.00 0110 0100 0000 0000 0000 0000
+127 0111 1111 0000 0000 +1000.00 0011 1110 1000 0000 0000 0000
+125 0111 1101 0000 0000 +100.9375 0000 0110 0100 1111 0000 0000
+64 0100 0000 0000 0000 +25.00 0000 0001 1001 0000 0000 0000
+25 0001 1001 0000 0000 +0.0625 0000 0000 0000 0001 0000 0000
+0.5 0000 0000 1000 0000 0.00 0000 0000 0000 0000 0000 0000
+0.015625 0000 0000 0000 0100 -0.0625 1111 1111 1111 1111 0000 0000
0 0000 0000 0000 0000 -0.25 1111 1111 1111 1100 0000 0000
-0.5 1111 1111 1000 0000 -1.00 1111 1111 1111 0000 0000 0000
-25 1110 0111 0000 0000 -250.00 1111 0000 0110 0000 0000 0000
-55 1100 1001 0000 0000 This format also applies to the High Fault and Low Fault
thresholds.
(Note that the practical temperature range varies with the ther-
mocouple type.)

www.maximintegrated.com Maxim Integrated │  13

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Four registers (05h through 08h) contain over- and under- current through the thermocouple wires. The time required
temperature thresholds for the linearized and cold-junc- to detect an open circuit depends on the values of the
tion-compensated temperature. These threshold register lead resistances and any filter capacitance at the thermo-
values are compared to the linearized temperature read- couple input and therefore, bits 4 and 5 also select the
ing found in registers 0Ch, 0Dh, and 0Eh. If a threshold is time allowed for open-circuit fault detection. A nominal
exceeded, the corresponding bit is set in the Fault Status detection time of either 10ms, 32ms, or 100ms can be
register (0Fh) and, if not masked, the FAULT output will selected. The Open-Circuit Detection Mode table (Table
assert. 4) shows the effect of these two bits on the conversion
time. When the device is in one-shot mode, open-circuit
Integrated Input Protection detection can be disabled or set to occur every one-shot
The internal circuitry is protected from excessive volt- conversion. When the device is in automatic conversion
ages applied to the thermocouple cables by integrated mode, open-circuit detection may be disabled, or it may
MOSFETs at the T+ and T- inputs, and the BIAS output. be set to automatically test for open circuits every 16 con-
These MOSFETs turn off when the input voltage is nega- version cycles. If on-demand detection is desired, select
tive or greater than VDD. The MOSFETs are capable of “detection disabled” (00), then select the setting for the
withstanding input voltages up to ±45V. If fault voltages desired time constant. An open-circuit detection test will
beyond the ±45V limits are expected, see the Applications be performed immediately after the current conversion is
Information section. completed. Disabling the open fault detection when in
When the absolute input voltage at T+ or T- is negative or comparator mode while there is an open fault present
greater than VDD, the Under-/Overvoltage Fault bit, Bit 1, will not clear the fault bit or FAULT pin. If this happens,
is set in the Fault Status register (0Fh) and the FAULT pin to subsequently clear the fault, the MAX31856 must
asserts if not masked. Conversions are suspended while be placed in interrupt mode and then the fault cleared.
the OVUV fault is present and will resume when the fault Note that, when cold-junction sensing is enabled, open-
is removed. circuit fault detection and cold-junction sensing occur
concurrently. Therefore, cold-junction temperature sens-
Open-Circuit Fault Detection ing has no effect on the overall cycle time when open-
Detection of open-circuit faults, such as those caused circuit fault detection is enabled. An open-circuit fault is
by broken thermocouple wires, can be enabled or dis- indicated by the Open Fault bit, Bit 0, in the Fault Status
abled using bits 4 and 5 in the Configuration 0 register register (0Fh) and the FAULT pin asserts if not masked.
(00h). Fault detection is accomplished by forcing a small

Table 4. Open-Circuit Detection Mode

BITS 5:4 FAULT TEST TIME (ms)
OCFAULT1:
FAULT TEST INPUT NETWORK CJ SENSE ENABLED CJ SENSE DISABLED
OCFAULT0 (Config
Byte 0) TYP MAX TYP MAX
00 Disabled N/A 0 0 0 0
Enabled
01 (Once every 16 RS < 5kΩ 13.3 15 40 44
conversions)
Enabled 40kΩ > RS > 5kΩ;
10 (Once every 16 Time constant < 33.4 37 60 66
conversions) 2ms
Enabled 40kΩ > RS > 5kΩ;
11 (Once every 16 Time constant > 113.4 125 140 154
conversions) 2ms

www.maximintegrated.com Maxim Integrated │  14

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Cold-Junction and Thermocouple edge (see Table 5 and Figure 3). There is one clock for
Out-of-Range Detection each bit transferred. Address and data bits are transferred
Thermocouple characteristics, the measurement circuitry, in groups of eight, MSB first.
and the linearization calculations limit the optimum tem- Address and Data Bytes
perature ranges for both the cold junction and the mea-
Address and data bytes are shifted MSB-first into the
surement junction (“hot junction”). Bit D7 of the Fault
serial-data input (SDI) and out of the serial-data output
Status register indicates when the cold-junction tempera-
(SDO). Any transfer requires the address of the byte to
ture falls outside of the optimum range, and bit D6 indi-
specify a write or a read, followed by one or more bytes of
cates when the hot-junction temperature is out of range.
data. Data is transferred out of the SDO for a read opera-
Table 1 shows the temperature limits that apply for the
tion and into the SDI for a write operation. The address
supported thermocouple types. These values are rounded
byte is always the first byte transferred after CS is driven
to the nearest °C. When the temperature falls outside of
low. The MSB (A7) of this byte determines whether the
the limit for a given measurement, the reported thermo-
following byte will be written or read. If A7 is 0, one or
couple temperature is clamped at the limit value. Note
more byte reads will follow the address byte. If A7 is 1,
that the FAULT pin never asserts for an out-of-range fault.
one or more byte writes will follow the address byte.
Serial Interface For a single-byte transfer, 1 byte is read or written and
Four pins are used for SPI-compatible communications: then CS is driven high (see Figure 4 and Figure 5). For
SDO (serial-data out), SDI (serial-data in), CS (chip a multiple-byte transfer, multiple bytes can be read or
select), and SCLK (serial clock). SDI and SDO are the written after the address has been written (see Figure 6).
serial-data input and output pins, respectively. The CS The address continues to increment through all memory
input initiates and terminates a data transfer. SCLK syn- locations as long as CS remains low. If data continues to
chronizes data movement between the master (microcon- be clocked in or out, the address will loop from 7Fh/FFh to
troller) and the slave (MAX31856). 00h/80h. Invalid memory addresses report an FFh value.
Attempting to write to a read-only register will result in no
The serial clock (SCLK), which is generated by the change to that register’s contents.
microcontroller, is active only when CS is low and dur-
ing address and data transfer to any device on the SPI DRDY
bus. The inactive clock polarity is programmable in some The DRDY output goes low when a new conversion result
microcontrollers. The MAX31856 automatically accom- is available in the Linearized Thermocouple Temperature
modates either clock polarity by sampling SCLK when CS register. When a read-operation of the Linearized
becomes active to determine the polarity of the inactive Thermocouple Temperature register or the Cold-Junction
clock. Input data (SDI) is latched on the internal strobe Temperature Register (if enabled) completes, DRDY
edge and output data (SDO) is shifted out on the shift returns high.

Table 5. Serial Interface Function

MODE CS SCLK SDI SDO
Disable Reset High Input Disabled Input disabled High impedance
CPOL = 1*, SCLK rising
Write Low Data bit latch High impedance
CPOL = 0, SCLK falling
CPOL = 1, SCLK falling
Read Low X Next data bit shift**
CPOL = 0, SCLK rising
Note: CPHA bit polarity must be set to 1.
*CPOL is the clock polarity bit that is set in the control register of the microcontroller.
**SDO remains at high impedance until 8 bits of data are ready to be shifted out during a read.

www.maximintegrated.com Maxim Integrated │  15

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

CS
SHIFT INTERNAL STROBE
CPOL = 1

SCLK

CS

SHIFT INTERNAL STROBE

CPOL = 0

SCLK

NOTE: CPOL IS A BIT THAT IS SET IN THE MICROCONTROLLER’S CONTROL REGISTER.

Figure 3. Serial Clock as a Function of Microcontroller Clock Polarity (CPOL)

CS

SCLK

SDI
A7 A6 A5 A4 A3 A2 A1 A0

SDO HIGH-Z
D7 D6 D5 D4 D3 D2 D1 D0

Figure 4. SPI Single-Byte Read

CS

SCLK

SDI
A7 A6 A5 A4 A3 A2 A1 A0 D7 D6 D5 D4 D3 D2 D1 D0

SDO HIGH-Z

Figure 5. SPI Single-Byte Write

www.maximintegrated.com Maxim Integrated │  16

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

CS

SCLK

ADDRESS DATA DATA DATA

WRITE SDI
BYTE BYTE 0 BYTE 1 BYTE N

ADDRESS
SDI
BYTE

READ

DATA DATA DATA

SDO
BYTE 0 BYTE 1 BYTE N

Figure 6. SPI Multibyte Transfer

DRDY

REGISTER
CONVERSION n CONVERSION n+1 CONVERSION n+2
CONTENTS

CS
DATA

SDO
ADDRESS
DATA

SDI

Figure 7. DRDY Operation

www.maximintegrated.com Maxim Integrated │  17

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Internal Registers
Communication with the MAX31856 is accomplished through 16 8-bit registers that contain conversion, status, and
configuration data. All programming is done by selecting the appropriate address of the desired register location. The
Register Memory Map (Table 6) illustrates the addresses for the temperature, status, and configuration registers.
The registers are accessed using the 0Xh addresses for reads and the 8Xh addresses for writes. Data is read from or
written to the registers MSB first. Attempts to write to a read-only register results in no change in the data.

Table 6. Register Memory Map

FACTORY
ADDRESS READ/WRITE NAME FUNCTION
DEFAULT
00h/80h Read/Write CR0 00h Configuration 0 Register
01h/81h Read/Write CR1 03h Configuration 1 Register
02h/82h Read/Write MASK FFh Fault Mask Register
03h/83h Read/Write CJHF 7Fh Cold-Junction High Fault Threshold
04h/84h Read/Write CJLF C0h Cold-Junction Low Fault Threshold
05h/85h Read/Write LTHFTH 7Fh Linearized Temperature High Fault Threshold MSB
06h/86h Read/Write LTHFTL FFh Linearized Temperature High Fault Threshold LSB
07h/87h Read/Write LTLFTH 80h Linearized Temperature Low Fault Threshold MSB
08h/88h Read/Write LTLFTL 00h Linearized Temperature Low Fault Threshold LSB
09h/89h Read/Write CJTO 00h Cold-Junction Temperature Offset Register
0Ah/8Ah Read/Write CJTH 00h Cold-Junction Temperature Register, MSB
0Bh/8Bh Read/Write CJTL 00h Cold-Junction Temperature Register, LSB
0Ch Read Only LTCBH 00h Linearized TC Temperature, Byte 2
0Dh Read Only LTCBM 00h Linearized TC Temperature, Byte 1
0Eh Read Only LTCBL 00h Linearized TC Temperature, Byte 0
0Fh Read Only SR 00h Fault Status Register

Register 00h/80h: Configuration 0 Register (CR0)

The Configuration 0 register selects the conversion mode (automatic or triggered by the 1-shot command), selects open-
circuit fault detection timing, enables the cold-junction sensor, clears the fault status register, and selects the filter notch
frequencies. The effects of the configuration bits are described below.

Default Value: 00h

MEMORY AC-
R/W R/W R/W R/W R/W R/W R/W R/W
CESS
00h/80h CMODE 1SHOT OCFAULT1 OCFAULT0 CJ FAULT FAULTCLR 50/60Hz
Bit 7 Bit 0

www.maximintegrated.com Maxim Integrated │  18

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 00h/80h: Configuration 0 Register (CR0) (continued)

BIT NAME DESCRIPTION
Conversion Mode
7 CMODE 0 = Normally Off mode (default)
1 = Automatic Conversion mode. Conversions occur continuously every 100ms (nominal).

One-Shot Mode
0 = No conversions requested (default)
1 = This causes a single cold-junction and thermocouple conversion to take place when Conversion
6 1SHOT Mode bit =0 (normally off mode). The conversion is triggered when CS goes high after writing a 1 to
this bit. Note that if a multi-byte write is performed, the conversion is triggered when CS goes high
at the end of the transaction. A single conversion requires approximately 143ms in 60Hz filter mode
or 169ms in 50Hz filter mode to complete. This bit self clears to 0.

These bits enable/disable open-circuit fault detection and select fault detection timing.
5:4 OCFAULT[1:0]
See Open-Circuit Fault Detection section and Table 4 for operation of these bits.

Cold-Junction Sensor Disable

0 = Cold-junction temperature sensor enabled (default)
1 = Cold-junction temperature sensor disabled. Data from an external temperature sensor may be
3 CJ written to the cold-junction temperature register. When this bit changes from 0 to 1, the most recent
cold-junction temperature value will remain in the cold-junction temperature register until the internal
sensor is enabled or until a new value is written to the register. The overall temperature conversion
time is reduced by 25ms (typ) when this bit is set to 1.

Fault Mode
0 = Comparator Mode. The FAULT output and respective fault bit reflects the state of any non-
masked faults by asserting when the fault condition is true, and deasserting when the fault condition
is no longer true. There is a 2°C hysteresis when in comparator mode for threshold fault conditions.
2 FAULT (default)
1 = Interrupt Mode. The FAULT output and respective fault bit asserts when a non-masked fault
condition is true and remain asserted until a 1 is written to the Fault Status Clear bit. This deasserts
FAULT and respective fault bit until a new fault is detected (note that this may occur immediately if
the fault condition is still in place).

Fault Status Clear

0 = Default
1 = When in interrupt mode, returns all Fault Status bits [7:0] in the Fault Status Register (0Fh) to 0
1 FAULTCLR
and deasserts the FAULT output. This bit has no effect in comparator mode. Note that the FAULT
output and the fault bit may reassert immediately if the fault persists. To prevent the FAULT output
from reasserting, first set the Fault Mask bits. The fault status clear bit self-clears to 0.

50Hz/60Hz Noise Rejection Filter Selection

0= Selects rejection of 60Hz and its harmonics (default)
0 50/60Hz 1= Selects rejection of 50Hz and its harmonics
Note: Change the notch frequency only while in the “Normally Off” mode – not in the Automatic
Conversion mode.

www.maximintegrated.com Maxim Integrated │  19

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 01h/81h: Configuration 1 Register (CR1)

The Configuration 1 register selects the averaging time for the thermocouple voltage conversion averaging mode and
also selects the thermocouple type being monitored.

Default Value: 03h

MEMORY
N/A R/W R/W R/W R/W R/W R/W R/W
ACCESS
01h/81h Reserved AVGSEL2 AVGSEL1 AVGSEL0 TC TYPE3 TC TYPE2 TC TYPE1 TC TYPE0
Bit 7 Bit 0

BIT NAME DESCRIPTION

7 Reserved Reserved.
Thermocouple Voltage Conversion Averaging Mode
000 = 1 sample (default)
001 = 2 samples averaged
010 = 4 samples averaged
011 = 8 samples averaged
1xx = 16 samples averaged
Adding samples increases the conversion time and reduces noise.
Typical conversion times:
6:4 AVGSEL[2:0] 1-shot or first conversion in Auto mode:
= tCONV + (samples -1) x 33.33mS (60Hz rejection)
= tCONV + (samples -1) x 40mS (50Hz rejection)
2 thru n conversions in Auto mode
= tCONV + (samples -1) x 16.67mS (60Hz rejection)
= tCONV + (samples -1) x 20mS (50Hz rejection)

The Thermocouple Voltage Conversion Averaging Mode settings should not be changed while
conversions are taking place.

Thermocouple Type
0000 = B Type
0001 = E Type
0010 = J Type
0011 = K Type (default)
0100 = N Type
3:0 TC TYPE[3:0]
0101 = R Type
0110 = S Type
0111 = T Type
10xx = Voltage Mode, Gain = 8. Code = 8 x 1.6 x 217 x VIN
11xx = Voltage Mode, Gain = 32. Code = 32 x 1.6 x 217 x VIN
Where Code is 19 bit signed number from TC registers and VIN is thermocouple input voltage

www.maximintegrated.com Maxim Integrated │  20

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 02h/82h: Fault Mask Register (MASK)

The Fault Mask Register allows the user to mask faults from causing the FAULT output from asserting. Masked faults
will still result in fault bits being set in the Fault Status register (0Fh). Note that the FAULT output is never asserted by
thermocouple and cold-junction out-of-range status.

Default Value: FFh

MEMORY
N/A N/A R/W R/W R/W R/W R/W R/W
ACCESS
CJ High CJ Low TC High TC Low OV/UV Open
02h/82h Reserved Reserved FAULT FAULT FAULT FAULT FAULT FAULT
Mask Mask Mask Mask Mask Mask
Bit 7 Bit 0

BIT NAME DESCRIPTION

7:6 Reserved Reserved.
Cold-Junction High Fault Threshold Mask
CJ High 0 = FAULT output asserted when the Cold-Junction Temperature rises above the Cold-Junction
5
FAULT Mask Temperature high threshold limit value
1 = FAULT output masked (default)

Cold-Junction Low Fault Threshold Mask

CJ Low 0 = FAULT output asserted when the Cold-Junction Temperature falls below the Cold-Junction
4
FAULT Mask Temperature low threshold limit value
1 = FAULT output masked (default)

Thermocouple Temperature High Fault Threshold Mask

TC High 0 = FAULT output asserted when the Thermocouple Temperature rises above the Thermocouple
3
FAULT Mask Temperature high threshold limit value
1 = FAULT output masked (default)

Thermocouple Temperature Low Fault Threshold Mask

TC Low 0 = FAULT output asserted when the Thermocouple Temperature falls below the Thermocouple
2
FAULT Mask Temperature low threshold limit value
1 = FAULT output masked (default)

Over-voltage or Undervoltage Input Fault Mask

OV/UV FAULT
1 0 = FAULT output asserted when an over- or undervoltage condition is detected
Mask
1 = FAULT output masked (default)
Thermocouple Open-Circuit Fault Mask
Open FAULT
0 0 = FAULT output asserted when a thermocouple open condition is detected
Mask
1 = FAULT output masked (default)

www.maximintegrated.com Maxim Integrated │  21

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 03h/83h: Cold-Junction High Fault Threshold Register (CJHF)

Write a temperature limit value to this register. When the measured cold-junction temperature is greater than this value,
the CJ High fault status bit will be set and (if not masked) the FAULT output will assert.

Default Value: 7Fh

MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
03h/83h CJHF7 CJHF6 CJHF5 CJHF4 CJHF3 CJHF2 CJHF1 CJHF0
Sign 26 25 24 23 22 21 20
Bit 7 Bit 0

Register 04h/84h: Cold-Junction Low Fault Threshold Register (CJLF)

Write a temperature limit value to this register. When the measured cold-junction temperature is less than this value, the
CJ Low fault status bit will be set and (if not masked) the FAULT output will assert.
Default Value: C0h
MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
04h/84h CJLF7 CJLF6 CJLF5 CJLF4 CJLF3 CJLF2 CJLF1 CJLF0
Sign 26 25 24 23 22 21 20
Bit 7 Bit 0

Register 05h/85h: Linearized Temperature High Fault Threshold Register, MSB (LTHFTH)
Write the MSB of the two-byte temperature limit value to this register. When the linearized thermocouple temperature is
greater than the two-byte (05h and 06h) limit value, the TC High fault status bit will be set and (if not masked) the FAULT
output will assert.

Default Value: 7Fh

MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
05h/85h LTHFTH7 LTHFTH6 LTHFTH5 LTHFTH4 LTHFTH3 LTHFTH2 LTHFTH1 LTHFTH0
Sign 210 29 28 27 26 25 24
Bit 7 Bit 0

Register 06h/86h: Linearized Temperature High Fault Threshold Register, LSB (LTHFTL)
Write the LSB of the two-byte temperature limit value to this register. When the linearized thermocouple temperature is
greater than the two-byte (05h and 06h) limit value, the TC High fault status bit will be set and (if not masked) the FAULT
output will assert.

Default Value: FFh

MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
06h/86h LTHFTL7 LTHFTL6 LTHFTL5 LTHFTL4 LTHFTL3 LTHFTL2 LTHFTL1 LTHFTL0
23 22 21 20 2-1 2-2 2-3 2-4
Bit 7 Bit 0

www.maximintegrated.com Maxim Integrated │  22

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 07h/87h: Linearized Temperature Low Fault Threshold Register, MSB (LTLFTH)
Write the MSB of the two-byte temperature limit value to this register. When the linearized thermocouple temperature is
less than the two-byte (07h and 08h) limit value, the TC Low fault status bit will be set and (if not masked) the FAULT
output will assert.
Default Value: 80h
MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
07h/87h LTLFTH7 LTLFTH6 LTLFTH5 LTLFTH4 LTLFTH3 LTLFTH2 LTLFTH1 LTLFTH0
Sign 210 29 28 27 26 25 24
Bit 7 Bit 0

Register 08h/88h: Linearized Temperature Low Fault Threshold Register, LSB (LTLFTL)
Write the LSB of the two-byte temperature limit value to this register. When the linearized thermocouple temperature is
less than the two-byte (07h and 08h) limit value, the TC Low fault status bit will be set and (if not masked) the FAULT
output will assert.

Default Value: 00h

MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
08h/88h LTLFTL7 LTLFTL6 LTLFTL5 LTLFTL4 LTLFTL3 LTLFTL2 LTLFTL1 LTLFTL0
23 22 21 20 2-1 2-2 2-3 2-4
Bit 7 Bit 0

Register 09h/89h: Cold-Junction Temperature Offset Register (CJTO)

When the cold-junction temperature sensor is enabled, this register allows an offset temperature to be applied to the
measured value. See the Cold-Junction Temperature Sensing section of this data sheet for additional information.

Default Value: 00h

MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
09h/89h CJTO7 CJTO6 CJTO5 CJTO4 CJTO3 CJTO2 CJTO1 CJTO0
Sign 22 21 20 2-1 2-2 2-3 2-4
Bit 7 Bit 0

www.maximintegrated.com Maxim Integrated │  23

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 0Ah/8Ah: Cold-Junction Temperature Register, MSB (CJTH)

This register contains the MSB of the two-byte (0Ah and 0Bh) value used for cold-junction compensation of the ther-
mocouple measurement. When the cold-junction temperature sensor is enabled, this register is read-only and contains
the MSB of the measured cold-junction temperature plus the value in the Cold-Junction Offset register. Also when the
cold-junction temperature sensor is enabled, a read of this register will reset the DRDY pin high. When the cold-junction
temperature sensor is disabled, this register becomes a read-write register that contains the MSB of the most recent
cold-junction conversion result until a new value is written into it. This allows writing the results from an external tem-
perature sensor, if desired. The maximum contained in the two cold-junction temperature bytes is clamped at 128°C and
the minimum is clamped at -64°C.

Default Value: 00h

MEMORY
R/W R/W R/W R/W R/W R/W R/W R/W
ACCESS
0Ah/8Ah CJTH7 CJTH6 CJTH5 CJTH4 CJTH3 CJTH2 CJTH1 CJTH0
Sign 26 25 24 23 22 21 20
Bit 7 Bit 0

Register 0Bh/8Bh: Cold-Junction Temperature Register, LSB (CJTL)

This register contains LSB of the two-byte (0Ah and 0Bh) value used for cold-junction compensation of the thermocouple
measurement. When the cold-junction temperature sensor is enabled, this register is read-only and contains the LSB of
the measured cold-junction temperature plus the value in the Cold-Junction Offset register. Also when the cold-junction
temperature sensor is enabled, a read of this register will reset the DRDY pin high. When the cold-junction temperature
sensor is disabled, this register becomes a read-write register that contains the LSB of the most recent cold-junction
conversion result until a new value is written into it.

Default Value: 00h

MEMORY
R/W R/W R/W R/W R/W R/W R R
ACCESS
0Bh/8Bh CJTL7 CJTL6 CJTL5 CJTL4 CJTL3 CJTL2 CJTL1 CJTL0
2-1 2-2 2-3 2-4 2-5 2-6 0 0
Bit 7 Bit 0

Register 0Ch: Linearized TC Temperature, Byte 2 (LTCBH)

This is the high byte of the 19-bit register that contains the linearized and cold-junction-compensated thermocouple
temperature value.
Default Value: 00h
MEMORY
R R R R R R R R
ACCESS
0Ch LTCBH7 LTCBH6 LTCBH5 LTCBH4 LTCBH3 LTCBH2 LTCBH1 LTCBH0
Sign 210 29 28 27 26 25 24
Bit 7 Bit 0

www.maximintegrated.com Maxim Integrated │  24

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 0Dh: Linearized TC Temperature, Byte 1 (LTCBM)

This is the middle byte of the 19-bit register that contains the linearized and cold-junction-compensated thermocouple
temperature value.

Default Value: 00h

MEMORY
R R R R R R R R
ACCESS
0Dh LTCBM7 LTCBM6 LTCBM5 LTCBM4 LTCBM3 LTCBM2 LTCBM1 LTCBM0
23 22 21 20 2-1 2-2 2-3 2-4
Bit 7 Bit 0

Register 0Eh: Linearized TC Temperature, Byte 0 (LTCBL)

This is the low byte of the 19-bit register that contains the linearized and cold-junction-compensated thermocouple tem-
perature value.

Default Value: 00h

MEMORY
R R R R R R R R
ACCESS
0Eh LTCBL7 LTCBL6 LTCBL5 LTCBL4 LTCBL3 LTCBL2 LTCBL1 LTCBL0
2-5 2-6 2-7 X X X X X
Bit 7 Bit 0

Register 0Fh: Fault Status Register (SR)

The Fault Status Register contains eight bits that indicate the fault conditions (Thermocouple Out-of-Range, Cold
Junction Out-of-Range, Cold Junction High, Cold Junction Low, Thermocouple High Temperature, Thermocouple Low
Temperature, Over-Under Voltage, or Open Thermocouple) that have been detected.

Default Value: 00h

MEMORY
R R R R R R R R
ACCESS
0Fh CJ Range TC Range CJHIGH CJLOW TCHIGH TCLOW OVUV OPEN
Bit 7 Bit 0

Note: When the MAX31856 is set to operate in “comparator” fault mode (set with bit 2 of Configuration 0 register (00h)), the fault
status bits simply reflect the state of any faults by asserting when the fault condition is true, and deasserting when the fault condition
is no longer true.
When in “interrupt” fault mode, the fault status bits assert when a fault condition is true. The bits remain asserted until a 1 is written
to the Fault Status Clear bit. This deasserts the fault bits until a new fault is detected (note that this may occur immediately if the
fault condition is still in place).

www.maximintegrated.com Maxim Integrated │  25

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Register 0Fh: Fault Status Register (SR) (continued)

BIT NAME DESCRIPTION

Cold Junction Out-of-Range
0 = The Cold-Junction temperature is within the normal operating range (-55°C to +125°C for types E,
7 CJ Range
J, K, N, and T; -50°C to +125°C for types R and S; 0 to 125°C for type B).
1 = The Cold-Junction temperature is outside of the normal operating range.

Thermocouple Out-of-Range
0 = The Thermocouple Hot Junction temperature is within the normal operating range (see Table 1).
6 TC Range
1 = The Thermocouple Hot Junction temperature is outside of the normal operating range.
Note: The TC Range bit should be ignored in voltage mode.

Cold-Junction High Fault

0 = The Cold-Junction temperature is at or lower than the cold-junction temperature high threshold
5 CJHIGH (default).
1 = The Cold-Junction temperature is higher than the cold-junction temperature high threshold. The
FAULT output is asserted unless masked.

Cold-Junction Low Fault

0 = The Cold-Junction temperature is at or higher than the cold-junction temperature low threshold
4 CJLOW (default).
1 = The Cold-Junction temperature is lower than the cold-junction temperature low threshold. The
FAULT output is asserted unless masked.

Thermocouple Temperature High Fault

0 = The Thermocouple Temperature is at or lower than the thermocouple temperature high threshold
3 TCHIGH (default).
1 = The Thermocouple Temperature is higher than the thermocouple temperature high threshold. The
FAULT output is asserted unless masked.

Thermocouple Temperature Low Fault

0 = Thermocouple temperature is at or higher than the thermocouple temperature low threshold
2 TCLOW (default).
1 = Thermocouple temperature is lower than the thermocouple temperature low threshold. The FAULT
output is asserted unless masked.

Overvoltage or Undervoltage Input Fault

0 = The input voltage is positive and less than VDD (default).
1 OVUV 1 = The input voltage is negative or greater than VDD. The FAULT output is asserted unless masked.
Note: The presence of the OVUV fault will suspend conversions and the ability of the MAX31856 to
detect other faults (or clear faults when in comparator mode) until the fault is no longer present.

Thermocouple Open-Circuit Fault

0 = No open circuit or broken thermocouple wires are detected (default)
0 OPEN
1 = An open circuit such as broken thermocouple wires has been detected. The FAULT output is
asserted unless masked.

www.maximintegrated.com Maxim Integrated │  26

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Applications Information These values may need to be modified depending on the

nature of the noise pickup. Other techniques, such as
Thermocouple Temperature Sensing Guide- adding series resistance and shielding the thermocouple
lines wires and circuit board, may also be necessary in the
Follow these guidelines to get the best results when sens- presence of larger noise sources. Figure 8 shows the
ing temperature. The Typical Application Circuit shows a typical application circuit with input capacitors and input
basic MAX31856 schematic. Connect the thermocouple resistors added.
wires to inputs T+ and T-; be sure that the wires are con-
nected to the correct input as shown in Figure 8. Connect Input Protection
the BIAS output to T-. This biases the thermocouple within The ±45V input protection circuitry prevents damage to
the common-mode range of the inputs. the IC caused by overvoltage conditions at T+, T-, or
BIAS. If larger input faults are possible, external protec-
Noise Considerations tion should be added. Resistors in series with T+, T-, and
Because of the small signal levels involved, thermocouple BIAS can increase the acceptable fault voltages. For
temperature measurement is susceptible to power-sup- example, adding 2kΩ in series with these inputs allows
ply-coupled noise. The effects of power-supply noise can an additional ±40V of overdrive before the 20mA input
be minimized by placing 0.1µF ceramic bypass capacitors current limit is reached. Note, however, that if the input
close to the VDD pins and to GND. has 45V across it and 20mA flowing into it, the power dis-
The input amplifier is a low-noise amplifier designed to sipation will be 900mW due to the overdrive at that input.
enable high-precision input sensing. Keep the thermocou- Overdriving other inputs at the same time will further
ple and connecting wires away from electrical noise sourc- increase the power dissipation. Always ensure that if a
es. It is strongly recommended to add a 100nF ceramic continuous overdrive voltage greater than ±45V is expect-
surface-mount differential capacitor, placed across the T+ ed, any current-limiting resistors are large enough to keep
and T- pins, to filter noise on the thermocouple lines. In total power dissipation well under the IC’s absolute maxi-
environments with high noise levels, especially significant mum power dissipation. Note also that added resistance
RF fields, a 100nF capacitor between T+ and T- should in series with T+ and T- can increase offset voltage, as
be supplemented with a 10nF capacitor between T+ and mentioned in the Effect of Series Resistance section.
GND, and another 10nF capacitor between T- and GND.

AGND DGND

BIAS FAULT
0.01µF

SDI
T-
100Ω 0.1µF MAX31856

T+ SDO
100Ω 0.01µF
TO MICROCONTROLLER
3.3V AVDD SCK
0.1µF

DNC CS

DRDY DVDD 3.3V

0.1µF

Figure 8. Typical Connection to Reduce the Effect of Noise Pickup in the Thermocouple Cable

www.maximintegrated.com Maxim Integrated │  27

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

AGND DGND

2kΩ
BIAS FAULT
0.01µF
2kΩ
SDI
T-
0.1µF
2kΩ
MAX31856
T+ SDO
0.01µF
TO MICROCONTROLLER
3.3V AVDD SCK
0.1µF

DNC CS

DRDY DVDD 3.3V

0.1µF

Figure 9. When Thermocouple Inputs May Be Exposed to Fault Voltages Greater than ±45V, Resistors Can be Added to Limit
Current into the MAX31856.

Effect of Series Resistance To minimize the effect of input resistance on accuracy:

Bias and leakage current at the thermocouple inputs will ●● Minimize the values of any external resistors
flow through input resistors and cable resistance, gen- ●● When the cable resistance is very low, match the val-
erating input offset voltage. For the circuits in Figure 8 ues of the external resistors as closely as possible.
and Figure 9, assuming that the thermocouple’s source
resistance is negligible, the offset voltage due to series ●● If the cable resistance is known, increase the value
resistance will be: of the resistor connected to T- by the value of the
cable resistance. This will minimize the total mis-
IB x ∆RS + ∆IB x RS match between the two inputs.
where: If the cable resistance is excessive, consider using larger-
●● RS is the series resistance between each input and gauge thermocouple wire.
the bias point
MAX31856 Location
●● ∆RS is the difference between the two RS values.
Because the MAX31856 includes an internal cold-junction
This will generally be equal to the tolerance of any
temperature sensor, place it in a location whose tempera-
discrete series resistors plus any cable resistance.
ture is as close as possible to that of the cold junction. If
●● IB is the input bias and leakage current the thermocouple leads are directly soldered to the PCB,
●● ∆IB is the differential input bias and leakage current the MAX31856 should be as close as possible to the
thermocouple lead connections and thermal gradients
As an example, assume that the circuit in Figure 8 will be
between the IC and the thermocouple connections should
used up to a temperature of 85°C, the mismatch between
be minimized. If the thermocouple leads terminate in a
the 100Ω input resistors is 1Ω, and the external cable
connector, mount the IC as close as possible to the con-
resistance is 50Ω. This yields a worst-case offset voltage
nector, and again minimize thermal gradients between the
due to the external resistances of:
connector and the IC.
65nA x (50Ω + 1Ω) + 4nA x 100Ω = 3.7μV

www.maximintegrated.com Maxim Integrated │  28

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Using “Unsupported” Thermocouple Types the transfer functions from the Configuration 1 Register
To use a thermocouple type other than B, E, J, K, N, table. When voltage mode is selected, no linearization is
R, S, or T, select one of the voltage mode options in performed on the conversion data. Use the voltage data
Configuration 1. Selecting “Gain = 8” results in a full-scale and the cold-junction temperature to calculate the thermo-
input voltage range of ±78.125mV. “Gain = 32” results couple’s hot-junction temperature.
in a full-scale input voltage range of ±19.531mV. See

Ordering Information Package Information

PART TEMP RANGE PIN-PACKAGE For the latest package outline information and land patterns
(footprints), go to www.maximintegrated.com/packages. Note
MAX31856MUD+ -55°C to +125°C 14 TSSOP that a “+”, “#”, or “-” in the package code indicates RoHS status
MAX31856MUD+T -55°C to +125°C 14 TSSOP only. Package drawings may show a different suffix character, but
the drawing pertains to the package regardless of RoHS status.
+Denotes a lead(Pb)-free/RoHS-compliant package.
T = Tape and reel. PACKAGE PACKAGE OUTLINE LAND
TYPE CODE NO. PATTERN NO.
14 TSSOP U14+2 21-0066 90-0113

www.maximintegrated.com Maxim Integrated │  29

MAX31856 Precision Thermocouple to Digital Converter
with Linearization

Revision History
REVISION REVISION PAGES
DESCRIPTION
NUMBER DATE CHANGED
0 2/15 Initial release —

For pricing, delivery, and ordering information, please contact Maxim Direct at 1-888-629-4642, or visit Maxim Integrated’s website at www.maximintegrated.com.

Maxim Integrated cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim Integrated product. No circuit patent licenses
are implied. Maxim Integrated reserves the right to change the circuitry and specifications without notice at any time. The parametric values (min and max limits)
shown in the Electrical Characteristics table are guaranteed. Other parametric values quoted in this data sheet are provided for guidance.

Maxim Integrated and the Maxim Integrated logo are trademarks of Maxim Integrated Products, Inc. © 2015 Maxim Integrated Products, Inc. │  30