Freescale Semiconductor
Application Note
Document Name: AN3785
Rev 6, 6/2012
How to Implement the Freescale MPL115A
Digital Barometer
by: John Young
INTRODUCTION
MPL115A is a simple barometer with digital output and high performance targeting low cost commercial applications. The
device employs a MEMS PRT pressure sensor with a conditioning IC to provide accurate pressure data. The sensor output is
accurate to ±1 kPa over the 50 kPa to 115 kPa pressure range. An integrated ADC provides digitized temperature and pressure
sensor outputs via an I2C or SPI bus. The part’s operating voltage is from 2.375 V to 5.5 V. The part is available as either the
MPL115A1 (SPI) or MPL115A2 (I2C) part. The customer can order the device in either bus protocol depending on which is more
convenient for their end application.
Calibration data is housed in on-board ROM. This data is used by a host microcontroller to apply a compensation algorithm to
the raw ADC data from the pressure sensor and may be accessed at any time. The Calibration data is stored as a series of
coefficients that are applied to the raw data to compensate for temperature and pressure variation in the raw output.
The following application note will describe the SPI and I2C protocol needed to communicate to the digital pressure sensor.
Following communication and extraction of the stored coefficients, raw data from the sensor’s PRT can be compensated on the
host microcontroller for pressure and temperature. The application note details this process and the end result of the calculation
is Compensated Pressure.
Updates to this Application Note:
This application note has been updated to reflect changes in the pressure compensation algorithm - the 2nd order coefficients
have now been removed. This modifies the compensated Pressure equation to be simpler than the previous version. Both are
detailed so a comparsion can be made. Particular details have also migrated to the datasheet for MPL115A1 and MPL115A2 to
expand on content in those documents.
Sample Boards:
Freescale has two sample eval boards available in either SPI or I2C MPL115A device variations; KITMPL115A1SPI or
KITMPL115A2I2C. These are simple PCB boards with sensor parts, associated pullup resistors and decoupling capacitors
soldered onboard for customer evaluation. This provides a quick sample tool to run evaluations of the small device given its LGA
footprint.
© Freescale Semiconductor, Inc., 2008, 2009, 2011, 2012. All rights reserved.
1
U1
1.0UF
1
2
4
CAP
SHDN
SHDN
SCLK
CS
8
5
7
DIN
DOUT
6
GND
VDD1
VDD
C1
3
C2
GND1
MPL115A1
1.0UF
1
VDD
A
B
4
3
2
1
1
3
0.1UF
2
C3
SW1
J4
Figure 1. KITMPL115A1SPI: MPL115A1 (SPI) Interface Board
VDD
VDD
R1
4.70K
R2
4.70K
1
R3
4.70K
U1
2
VDD1
1.0UF
4
CAP
SHDN
7
SDA
GND
1
VDD
C1
C2
3
GND1
8
J1
SCL
RST
5
1
2
DOUT
6
HDR_1X2
MPL115A2
1
1.0UF
VDD
A
1
2
B
0.1UF
3
C3
SW1
Figure 2. KITMPL115A2I2C: MPL115A2 (I2C) Interface Board
PRODUCT FEATURES
The MPL115A is an absolute pressure sensor with digital output for low cost applications. A miniature 5 x 3 x 1.2 mm LGA
package ideally suits it for portable electronics and space constrained applications. Low current consumptions of 5 A during
Active mode and 1 A during Shutdown (Sleep) mode target battery and other low-power applications. A wide operating
temperature range from -40°C to +105°C fits demanding environmental requirements.
MPL115A employs a MEMS pressure sensor with a conditioning IC to provide accurate pressure measurement from 50 to
115 kPa. An integrated ADC provides digitized temperature and pressure sensor outputs via an I2C or SPI port. Calibration Data
is stored in internal ROM. Utilizing raw sensor output, the host microcontroller executes a compensation algorithm to render
Compensated Absolute Pressure with 1 kPa accuracy.
The MPL115A pressure sensor’s small form factor, low power capability, precision, and digital output optimize it for barometric
measurement applications
AN3785
Sensors
Freescale Semiconductor
2
1
Mechanical and Electrical Specifications
1.1
Maximum Ratings
Voltage (with respect to GND unless otherwise noted)
VDD ..................................................................................................................... -0.3 V to +5.5 V
SHDN, SCLK, CS, DIN, DOUT ..................................................................... -0.3 V to VDD+0.3 V
Operating Temperature Range ..........................................................................-40°C to +105°C
Storage Temperature Range..............................................................................-40°C to +125°C
Overpressure ................................................................................................................1000 kPa
1.2
Operating Characteristics
VDD = 2.375 V to 5.5 V, TA = -40°C to +105°C, unless otherwise noted. Typical values are at VDD = 3.3 V, TA = +25°C.
Ref
Parameters
Symbol
1
Operating Supply Voltage
VDD
2
Supply Current
IDD
Conditions
Min
Typ
Max
Units
2.375
3.3
5.5
V
Shutdown (SHDN = GND)
—
—
1
A
Standby
—
3.5
10
A
Average – at one measurement per second
—
5
6
A
Pressure Sensor
3
Range
50
—
115
kPa
4
Resolution
—
0.15
—
kPa
5
Accuracy
-20ºC to 85ºC
—
—
±1
kPa
6
Power Supply Rejection
Typical operating circuit at DC
0.1
—
kPa/V
100 mV p-p 217 Hz square wave plus 100 mV
pseudo random noise with 10 MHz bandwidth
0.1
—
kPa
7
Conversion Time
(Start Pressure and Temperature
Conversion)
tc
Time between start convert command and
data available in the Pressure and
Temperature registers
—
1.6
3
ms
8
Wakeup Time
tw
Time between leaving Shutdown mode
(SHDN goes high) and communicating with
the device to issue a command or read data.
—
3
5
ms
I2C I/O Stages: SCL, SDA
9
SCL Clock Frequency
fSCL
—
—
400
kHz
10
Low Level Input Voltage
VIL
—
—
0.3VDD
V
11
High Level Input Voltage
VIH
0.7VDD
—
—
V
0
—
0.4
s
I2C Outputs: SDA
12
Data Setup Time
tSU
Setup time from command receipt to ready to
transmit
I2C Addressing
MPL115A2 uses 7-bit addressing, does not acknowledge the general call address 0000000. Slave address has been set to 0x60 or 1100000.
AN3785
Sensors
Freescale Semiconductor
3
1.3
Pin Connections SPI
Top View
VDD
1
8
SCLK
CAP
2
7
DIN
GND
3
6
DOUT
SHDN
4
5
CS
Figure 3. Pin Connections SPI
1.4
Pin Connections I2C
Top View
VDD
1
8
SCL
CAP
2
7
SDA
GND
3
6
NC
SHDN
4
5
RST
Figure 4. Pin Connections I2C
AN3785
Sensors
Freescale Semiconductor
4
2
Overview of Functions/Operation
Initial
powerup
Reading
coefficient data
Data conversion
Compensated
pressure reading
Shutdown
Figure 5. Sequence Flow Chart
The MPL115A interfaces to a host (or system) microcontroller in the user’s application. All communications are via SPI or I2C. A
typical usage sequence is as follows:
Initial Power-up
All circuit elements are active. SPI or I2C port pins are high impedance and associated registers are cleared. The device then
enters standby mode.
Reading Coefficient Data
The user then typically accesses the part and reads the coefficient data. The main circuits within the slave device are disabled
during read activity. The coefficients are usually stored in the host microcontoller local memory but can be re-read at any time.
It is suggested to read the coefficients once and store the values in the host microcontroller for speed. It is not necessary to read
this multiple times because the coefficients within a device are constant and do not change. However, note that the coefficients
will be different from device to device, and cannot be used for another part.
Data Conversion
This is the first step that is performed each time a new pressure reading is required which is initiated by the host sending the
CONVERT command. The main system circuits are activated (wake) in response to the command and after the conversion
completes, the result is placed into the Pressure and Temperature ADC output registers.
The conversion completes within the maximum conversion time, tc of 1.6ms typical, or 3ms max. The device then enters standby
mode.
Compensated Pressure Reading
After the conversion has been given sufficient time to complete, the host microcontroller reads the result from the ADC output
registers and calculates the Compensated Pressure, a barometric/atmospheric pressure value which is compensated for
changes in temperature and pressure sensor linearity. This is done using the coefficient data from the MPL115A and the raw
sampled pressure and temperature ADC output values, in a compensation equation (detailed later). Note that this is an absolute
pressure measurement with a vacuum as a reference.
From this step the host controller may either wait and then return to the Data Conversion step to obtain the next pressure reading
or it may go to the Shutdown step with the full supply voltage range.
AN3785
Sensors
Freescale Semiconductor
5
I2C Communication
The I2C protocol is explained below along with how to implement this pressure sensor. For additional information on I2C,
please see AN4481 Sensors I2C Setup and FAQ. I2C is compliant to the Philips I2C Bus specification, version 2.1.
The MPL115A2 uses a 7 bit device address such that in an 8 bit representation, the device address is followed by a least
significant bit that toggles a I2C read/write value. This leads to a I2C MPL115A2 address of 0xC0 for a write command, and 0xC1
for a read command.
MPL115A2 Commands
Address
Write Command
0xC0
Read Command
0xC1
Figure 6. MPL115A2 Commands
Write Mode
I2C writes to the device are conducted using normal I2C protocol. An I2C start condition is followed by the 7-bit slave address
and Write bit (0). The slave acknowledges the write request and latches the following data byte from the master. Further bytes
are ignored when in normal (user) mode. A stop or repeat start condition must be sent by the master to complete the sequence.
User mode write commands include CONVERT to Start Pressure and Temperature conversion. These commands are shown in
Table 2. An example of the user mode I2C write sequence is shown in Figure .
7 bit slave address plus
WRITE bit
S
D
D
D
D
Start
D
8 bit command
D D W
Ack
0
X
X
X
X
X
Stop bit (or repeated start)
X
X
Acknowledge
Ack
P
End of Transmission
Figure 7. I²C Write Sequence – User Mode
Conversion Time
Note that when sending a write command to Start Conversion, there is a conversion time of 1.6ms typical for the part to
internally sample and output results. After this conversion time, the data will be ready to be read. Before entering a repeated start
or new start, ensure this delay is inserted before reading new data.
Read Mode
I2C reads from the device are conducted using normal I2C protocol.
To read from the device the master must first send a data address or command that points to the required data. To do this the
master sends the 7-bit slave address + Write bit followed by data address byte. Transmission is initiated by an I2C start condition
and ended by either a Repeated Start or a Stop condition. The master then resends the 7-bit slave address, this time with a Read
bit, either directly following the Repeated Start condition or, if the previous transmission ended with a Stop condition, preceded
by an I2C start condition.
The slave acknowledges each byte received and proceeds to send the requested data following receipt of the second slave
address. The slave continues to send data, incrementing the address pointer at the end of each byte. The master acknowledges
receipt of each byte sent by the slave and asserts a Stop (or Repeated Start) condition at the conclusion of the transmission.
Examples of I2C reads for pressure data (2 bytes) and coefficient data (up to 12 bytes) are shown in Figure 7 and Figure 9.
10-bit pressure and temperature data is transmitted MSB first with trailing zeros. This allows the user to read only the first byte
in systems that do not require the full resolution of the part and may only have limited processing capability. The device continues
to send data until either the transmission is complete or the master either issues a stop condition or fails to assert an acknowledge
bit i.e. the slave treats a “Nak” (not acknowledge) as a stop bit. The master thus controls the amount of data it receives and
discards any data that is not needed.
AN3785
Sensors
Freescale Semiconductor
6
I²C Read Pressure Sequence
7 bit slave address
plus WRITE bit
S
D
D
D
D
D
D
“Read Pressure MSB”
command
D W
Ack
Start
0
X
X
X
X
X
X
7 bit slave address
plus READ bit
X
Ack
D
D
D
D
D
D
R
Ack
Acknowledge
10 bit pressure data with trailing zeros
P
D
P
P
P
P
P
P
P
Ack
P
P
0
0
Stop bit
0
0
0
0
Nak(1)
P
Signifies End of Transmission
Slave stops sending data immediately
1. The standard termination sequence is “Nak” followed by the stop condition. An “Ack” followed by the stop condition may also be used.
Figure 8. I²C Read Pressure Sequence
I²C Read Coefficients Sequence
7 bit slave address
plus WRITE bit
S
D
D
D
D
D
D
Start
“Read Coefficient 1”
command
D W
Ack
X
0
0
0
0
1
0
7 bit slave address
plus READ bit
0 Ack
Coefficient data
second byte
D D D D D D R Ack
Repeated
Start
Acknowledge
Coefficient data
first byte
Sr
Coefficient data
third byte
A A A A A A A A Ack B B B B B B B B Ack C C C C C C C C Ack
Coefficient data
final byte
Ack Z Z Z Z Z Z Z Z Nak(1)
P
Ack(2)
1. The standard termination sequence is “Nak” followed by the stop condition. An “Ack” followed by the stop condition may also be used.
2. Stop bit signifies end of transmission. Slave stops sending data immediately.
Figure 9. I²C Read Coefficients Sequence
AN3785
Sensors
Freescale Semiconductor
7
I2C RST
The I2C RST function that can be seen in the pin connection, prevents the I2C circuits from drawing power while the I2C inputs
are active. Ideally this is accomplished by removing power from the I2C I/O circuits. The part ignores I2C communications when
RST is asserted and does not draw additional power due to transitions on the I2C connections. Drive line low to disable I2C
communications. I2C pins are high impedance and communications are ignored. If RST is asserted during an I2C transmission
then the transmission is aborted (and lost). All other internal functions operate normally. No other functions or registers are reset
as a result of asserting this function.
SPI Interface
In SPI mode MPL115A operates as a half duplex 4-wire SPI slave capable of bus speeds up to 8 Mb/sec.
SPI 4-Wire Mode
4 lines make up the SPI interface: SCLK, DIN, DOUT and CS. Data is read from the port in 2 byte sequence: address byte
plus Read/Write bit received on DIN followed by data byte transmitted on DOUT. Write commands are a 2 byte sequence:
address byte plus Read/Write bit followed by data byte, both received on DIN. Exceptions to this is the “action” address “Start
Conversion.” The interface is half duplex and so cannot receive and transmit simultaneously. Minimum data setup time (interval
between Start Conversions receipt and data ready on DOUT) is typically 1.6 ms.
SPI transfers to and from the part are controlled by the CS line. CS is driven low by the master at the start of communication
and held low for the duration of the two byte transfer.
SPI data transfers occur on the rising edges of SCLK. Data on the DIN and DOUT lines is changed on the falling edges of
SCLK and clocked on the rising edges of SCLK: the rising edges provide the “data valid” condition.
Driving CS high places DOUT in a high impedance state. DOUT remains high impedance while CS is high. DOUT is held low
while CS is low and there is no transmission activity on DOUT.
The next data transfer commences on the next CS high to low transition.
4-Wire Mode
CS
CS
SCLK
SCLK
DIN
DIN
SDI
R /W A[5] A[4] A[3] A[2] A[1] A[0]
DOUT
DOUT
SDO
D[7] D[6] D[5] D[4] D [3] D [2] D[1] D[0]
Figure 10. SPI Read Operation
CS
CS
SCLK
SCLK
DIN
DIN
SDI
R/W A[5] A[4] A[3] A[2] A[1] A[0]
D[7] D[6] D[5] D[4] D [3] D [2] D[1] D[0]
Figure 11. SPI Write Operation
DEVICE READ/WRITE OPERATIONS
All device read/write operations are memory mapped. Device actions “Start Conversions” are controlled by writing to the
appropriate memory address location. All memory address locations are 6-bit (see Table 1). A read/write bit (SPI) and a “don’t
care” bit(s) are added to make up the total number of bits to 8.
AN3785
Sensors
Freescale Semiconductor
8
For I2C devices the MSB (bit 7) is “don’t care” and bits 6 to 0 are available for addressing e.g. bit 6 is always 0 with the
remaining bits (5:0) making up the 6-bit address in MPL115A.
For SPI devices the MSB is R/W and the LSB is “don’t care” with the remaining bits (6:1) making up the 6-bit address. The
appropriate memory location is written to/read from in response to the condition of the R/W bit which is ‘1’ for read operations
and ‘0’ for write operations.
The basic device write commands are shown in Table 1 (SPI versions) and Table 2 (I2C versions). The availability of particular
commands depends on the operating mode of the device. There are three basic operating modes: Test, Debug and User
(normal).
Table 1. I2C Write Commands
Binary
HEX(1)
Devices Address + Write bit
11000000
0xC0
Devices Address + Write bit
Start Conversions
X0010010
0x12
Start Conversions
Command
Command
1. The command byte needs to be paired with a 0x00 as part of the SPI exchange to complete the passing of Start Conversions.
Table 2. I2C Write Commands description
Command
Start Conversions
Action Taken
Wake main circuits. Start clock. Allow supply stabilization time. Select pressure sensor input. Apply positive sensor
excitation and perform A to D conversion. Select temperature input. Perform A to D conversion. Load the Pressure and
Temperature registers with the result. Shut down main circuits and clock. This action takes a typical time of 1.6ms to
complete and have data available in the Pressure and Temperature registers.
Table 3. I2C Read Command Description
Binary
HEX(1)
Device Address + Read bit
11000001
0xC1
Read Pressure MSB
X0000000
0x00
Read Pressure LSB
X0000001
0x01
Read Temperature MSB
X0000010
0x02
Read Temperature LSB
X0000011
0x03
Read Coefficient data byte 1
X0000100
0x04
Command
X = don’t care
1. The command byte needs to be paired with a 0x00 as part of the SPI exchange to complete the passing of Start Conversions.
Table 4. SPI Write Commands
Command
Start Conversions
Binary
HEX(1)
0010010X
0x24
X = don’t care
1. The command byte needs to be paired with a 0x00 as part of the SPI exchange to complete the passing of Start Conversions.
AN3785
Sensors
Freescale Semiconductor
9
Table 5. SPI Write Command Description
Command
Action Taken
Wake main circuits. Start clock. Allow supply stabilization time. Select pressure sensor input. Apply
positive sensor excitation and perform A to D conversion. Select temperature input. Perform A to D
conversion. Load the Pressure and Temperature registers with the result. Shut down main circuits and
clock. This action takes a typical time of 1.6ms to complete and have data available in the Pressure
and Temperature registers.
Start Conversions
Table 6. Example SPI Read Commands
Binary
HEX(1)
Read Pressure MSB
1000000X
0x80
Read Pressure LSB
1000001X
0x82
Read Temperature MSB
1000010X
0x84
Read Temperature LSB
1000011X
0x86
Read Coefficient data byte 1
1000100X
0x88
Command
X = don’t care
1. The command byte needs to be paired with a 0x00 as part of the SPI exchange to complete the passing of stated command.
DEVICE MEMORY MAP
Table 1. Device Memory Map
Address
Name
Description
Size (bits)
0x00
Padc_MSB
10-bit Pressure output value MSB
8
0x01
Padc_LSB
10-bit Pressure output value LSB
2
0x02
Tadc_MSB
10-bit Temperature output value MSB
8
0x03
Tadc_LSB
10-bit Temperature output value LSB
2
0x04
a0MSB
a0 coefficient MSB
8
0x05
a0LSB
a0 coefficient LSB
8
0x06
b1MSB
b1 coefficient MSB
8
0x07
b1LSB
b1 coefficient LSB
8
0x08
b2MSB
b2 coefficient MSB
8
0x09
b2LSB
b2 coefficient LSB
8
0x0A
c12MSB
c12 coefficient MSB
8
0x0B
c12LSB
c12 coefficient LSB
8
0x0C
Reserved*
—
—
0x0D
Reserved*
—
—
0x0E
Reserved*
—
—
0x0F
Reserved*
—
—
0x10
Reserved
—
—
0x11
Reserved
—
—
0x12
CONVERT
Start both Pressure and Temperature Conversion
—
*These registers are set to 0x00. These are reserved, and were previously utilized as Coefficient values, now
set to a value to 0x00 for simplifying compensation.
2.1
Pressure, Temperature and Coefficient Bit-Width Specifications
The table below specifies the initial coefficient bit-width specifications for the compensation algorithm and the specifications for
Pressure and Temperature ADC values.
AN3785
Sensors
Freescale Semiconductor
10
Pressure, Temperature and Compensation Coefficient Specifications
a0
b1
b2
c12
Padc
Tadc
Total Bits
16
16
16
14
10
10
Sign Bits
1
1
1
1
0
0
Integer Bits
12
2
1
0
10
10
Fractional Bits
3
13
14
13
0
0
dec pt zero pad
0
0
0
9
0
0
AN3785
Sensors
Freescale Semiconductor
11
Example Binary Format Definitions:
a0 Signed, Integer Bits = 12, Fractional Bits = 3 :
Coeff a0 = S I11 I10 I9 I8 I7 I6 I5 I4 I3 I2 I1 I0 . F2 F1 F0
b1 Signed, Integer Bits = 2, Fractional Bits = 7 :
Coeff b1 = S I1 I0 . F12 F10 F9 F8 F7 F6 F5 F4 F3 F2 F1 F0
b2 Signed, Integer Bits = 1, Fractional Bits = 14 :
Coeff b2 = S I0 . F13 F12 F10 F9 F8 F7 F6 F5 F4 F3 F2 F1 F0
c12 Signed, Integer Bits = 0, Fractional Bits = 13, dec pt zero pad = 9 :
Coeff c12 = S 0 . 000 000 000 F12 F10 F9 F8 F7 F6 F5 F4 F3 F2 F1 F0
Padc Unsigned, Integer Bits = 10 :
Padc U = I9 I8 I7 I6 I5 I4 I3 I2 I1 I0
Tadc Unsigned, Integer Bits =10 :
Tadc U = I9 I8 I7 I6 I5 I4 I3 I2 I1 I0
NOTE: Negative coefficients are coded in 2’s complement notation.
2.2
Compensation
The 10-bit compensated pressure output, Pcomp, is calculated as follows:
Pcomp = a0 + b1 + c12 Tadc Padc + b2 Tadc
Eqn. 1
Where:
Padc is the 10-bit pressure ADC output of the MPL115A
Tadc is the 10-bit temperature ADC output of the MPL115A
a0 is the pressure offset coefficient
b1 is the pressure sensitivity coefficient
b2 is the 1st order temperature offset coefficient (TCO)
c12 is the coefficient for temperature sensitivity coefficient (TCS)
Pcomp will produce a value of 0 with an input pressure of 50 kPa and will produce a full-scale value of 1023 with an input pressure
of 115 kPa.
115 – 50
Pressure (kPa) = P comp ---------------------- + 50
1023
2.3
Eqn. 2
Evaluation Sequence, Arithmetic Circuits
The following is an example of the calculation for Pcomp, the compensated pressure output. Input values are in bold.
c12x2 = c12 * Tadc
a1 = b1 + c12x2
a1x1 = a1 * Padc
y1 = a0 + a1x1
a2x2 = b2 * Tadc
Pcomp = y1 + a2x2
This can be calculated as a succession of Multiply Accumulates (MACs) operations of the form y = a + b * x:
a
b
+
y
X
x
AN3785
Sensors
Freescale Semiconductor
12
The polynomial can be evaluated (Equation 1) as a sequence of 3 MACs:
Pcomp = a0 + b1 + c12 Tadc Padc + b2 Tadc
b1
a1
c12
Tadc
y1
a0
Padc
y
b2
PComp
Tadc
I2C COMMUNICATION AND EXAMPLE
The actions taken by the part in response to each command are as follows:
Table 2. I2C Write Command Description
Command
Start Conversions
Action Taken
Wake main circuits. Start clock. Allow supply stabilization time. Select pressure sensor input. Apply positive sensor
excitation and perform A to D conversion. Select temperature input. Perform A to D conversion. Load the Pressure and
Temperature registers with the result. Shut down main circuits and clock.
Table 3. I2C Read Command Description
Binary
HEX(1)
Device Address + Read bit
1100 0001
0xC1
Read Pressure MSB
X000 0000
0x00
Read Pressure LSB
X000 0001
0x01
Read Temperature MSB
X000 0010
0x02
Read Temperature LSB
X000 0011
0x03
Read Coefficient data byte 1
X000 0100
0x04
Command
X = don’t care
These are MPL115A2 I2C commands to read coefficients, execute Pressure and Temperature conversions, and to read Pressure
and Temperature data. The sequence of the commands for the interaction is given as an example to operate the MPL115A2.
Utilizing this gathered data, an example of the calculating the Compensated Pressure reading is given in floating point notation.
I2C Commands (simplified for communication)
Device Address + write bit “To Write” = 0xC0
Device Address + read bit “To Read” = 0xC1
Command to Write “Convert Pressure and Temperature” = 0x12
Command to Read “Pressure ADC High byte” = 0x00
AN3785
Sensors
Freescale Semiconductor
13
Command to Read “Pressure ADC Low byte” = 0x01
Command to Read “Temperature ADC High byte” = 0x02
Command to Read “Temperature ADC Low byte” = 0x03
Command to Read “Coefficient data byte 1 High byte” = 0x04
Read Coefficients:
[0xC0], [0x04], [0xC1], [0x3E], [0xCE], [0xB3], [0xF9], [0xC5], [0x17], [0x33], [0xC8]
Figure 12. I2C Read Coefficient Datagram
a0 coefficient MSB
= 0x3E
a0 coefficient LSB
= 0xCE a0 coefficient
b1 coefficient MSB
= 0xB3
b1 coefficient LSB
= 0xF9 b1 coefficient
b2 coefficient MSB
= 0xC5
b2 coefficient LSB
= 0x17
b2 coefficient
= 0x3ECE = 2009.75
= 0xB3F9
= -2.37585
= 0xC517
=
-0.92047
c12 coefficient MSB = 0x33
c12 coefficient LSB
= 0xC8 c12 coefficient = 0x33C8
= 0.000790
Figure 13. I2C Start Conversion Datagram
Command to I2C Start Conversion, 0x12
AN3785
Sensors
Freescale Semiconductor
14
Figure 14. I2C Read Results Datagram
Pressure MSB
= 0x66
Pressure LSB
= 0x80
Pressure
= 0x6680
= 0110 0110 1100 0000
= 410 ADC counts
Temperature MSB = 0x7E
Temperature LSB
= 0xC0
Temperature
= 0x7EC0 = 0111 1110 1100 0000
= 507 ADC counts
2.4
Example of Pressure Compensated Calculation in Floating-point Notation
a0 coefficient
=
2009.75
b1 coefficient
=
-2.37585
b2 coefficient
=
-0.92047
c12 coefficient =
0.000790
Pressure
=
410 ADC counts
Temperature
=
507 ADC counts
Pressure Compensation:
Pcomp = a0 + b1 + c12 Tadc Padc + b2 Tadc
Using the evaluation sequence shown in Section 3.3:
c12x2
= c12 * Tadc
a1
= b1 + c12x2 = -2.37585 + 0.40053
= -1.97532
a1x1
= a1 * Padc
= -1.97532 * 410
= -809.8812
y1
= a0 + a1x1
= 2009.75 + (-809.8812)
= 1199.8688
a2x2
= b2 * Tadc
= -0.92047 * 507
= -466.67829
= 1199.8688 + (-466.67829)
= 733.19051
PComp = y1 + a2x2
= 0.000790 * 507
= 0.40053
AN3785
Sensors
Freescale Semiconductor
15
2.5
SPI Device Read/Write Operations
All device read/write operations are memory mapped. Device actions e.g. “Start Pressure Conversion” are controlled by writing
to the appropriate memory address location. All memory address locations are 6-bit (see Table 1).
The 8-bit command word comprises:
• the most significant bit which is the Read/Write identifier which is '1' for read operations and '0' for writes.
• the 6-bit address (from Table 1);
• the least significant bit which is not used and is don't care (X).
The device write commands are shown in Table 1.
Table 4. SPI Write Command
Command
Start Conversions
Binary
HEX(1)
0010010X
0x24
X = don’t care
1. The command byte needs to be paired with a 0x00 as part of the SPI exchange to complete the passing of Start
Conversions.
The actions taken by the part in response to each command are as follows:
Table 5. SPI Write Command Description
Command
Start Conversions
Action Taken
Wake main circuits. Start clock. Allow supply stabilization time. Select pressure sensor input.
Apply positive sensor excitation and perform A to D conversion. Select temperature input.
Perform A to D conversion. Load the Pressure and Temperature registers with the result. Shut
down main circuits and clock.
AN3785
Sensors
Freescale Semiconductor
16
SPI Read operations are performed by sending the required address with a leading Read bit set to ‘1’. SPI operations require
that each byte be addressed individually. All data is transmitted most significant bit first.
Table 6. Example SPI Read Commands
Binary
HEX(1)
Read Pressure MSB
1000000X
0x80
Read Pressure LSB
1000001X
0x82
Read Temperature MSB
1000010X
0x84
Read Temperature LSB
1000011X
0x86
Read Coefficient data byte 1
1000100X
0x88
Command
X = don’t care
1. The command byte needs to be paired with a 0x00 as part of the SPI exchange to complete the passing of stated
command.
2.6
SPI Timing
Table 7 and Figure describe the timing requirements for the SPI system.
Table 7. SPI Timing
Ref
Function
Symbol
Min
Max
Unit
Of
—
8
MHz
1
Operating Frequency
2
SCLK Period
tSCLK
125
—
ns
3
SCLK High time
tCLKH
62.5
—
ns
4
SCLK Low time
tCLKL
62.5
—
ns
5
Enable lead time
tSCS
125
—
ns
6
Enable lag time
tHCS
125
—
ns
7
Data setup time
tSET
30
—
ns
8
Data hold time
tHOLD
30
—
ns
9
Data valid (after SCLK low edge)
tDDLY
—
32
ns
10
Width CS High
tWCS
30
—
ns
CS
SCLK
DIN
DOUT
Figure 15. SPI Timing Diagram
AN3785
Sensors
Freescale Semiconductor
17
2.7
Example of SPI Reading of Coefficients
These are MPL115A2 SPI commands to read coefficients, execute Pressure and Temperature conversions, and to read Pressure
and Temperature data. The sequence of the commands for the interaction is given as an example to operate the MPL115A2.
Utilizing this gathered data, an example of the calculating the Compensated Pressure reading is given in floating point notation.
SPI Commands (simplified for communication)
Command to Write “Convert Pressure and Temperature” = 0x24
Command to Read “Pressure ADC High byte” = 0x80
Command to Read “Pressure ADC Low byte” = 0x82
Command to Read “Temperature ADC High byte” = 0x84
Command to Read “Temperature ADC High byte” = 0x86
Command to Read “Coefficient data byte 1 High byte” = 0x88
Read Coefficients:
[CS=0], [0x88], [0x00], [0x8A], [0x00], [0x8C], [0x00], [0x8E], [0x00], [0x90], [0x00], [0x92], [0x00], [0x94], [0x00], [0x96], [0x00],
[0x00], [CS=1]
Start Pressure and Temperature Conversion, Read raw Pressure:
[CS=0], [0x24], [0x00], [CS=1], [13 ms Delay]
[CS=0], [0x80], [0x00], [0x82], [0x00], [0x84], [0x00,] [0x86], [0x00], [0x00], [CS=1]
NOTE: Extra [0x00] at the end of each sequence to output the last data byte on the slave side of the SPI.
Figure 16. SPI Read Coefficient Datagram
a0 coefficient MSB
= 0x41
a0 coefficient LSB
= 0xDF a0 coefficient
b1 coefficient MSB
= 0xB0
b1 coefficient LSB
= 0x28
b2 coefficient MSB
= 0xBE
b2 coefficient LSB
= 0xAD b2 coefficient
b1 coefficient
= 0x41DF
= 2107.875
= 0xB028
= -2.495117188
= 0xBEAD =
-1.02069
c12 coefficient MSB = 0x38
c12 coefficient LSB
= 0xCC c12 coefficient = 0x38CC = 0.000867
AN3785
Sensors
Freescale Semiconductor
18
Figure 17. SPI Start Conversion Datagram
Command to Start Pressure and Temperature Conversion, 0x24
Figure 18. SPI Read Results Datagram
Pressure MSB
= 0x67
Pressure LSB
= 0xC0
Pressure
= 0x67C0
= 0110 0111 11 00 0000
= 415 ADC counts
Temperature MSB = 0x80
Temperature LSB
= 0x40
Temperature
= 0x8040
= 1000 0000 01 00 0000
= 513 ADC counts
2.8
Example of Pressure Compensated Calculation in Floating-point Notation
a0 coefficient
=
2107.875
b1 coefficient
=
-2.495117188
b2 coefficient
=
-1.02069
c12 coefficient =
0.000867
Pressure
=
415 ADC counts
Temperature
=
513 ADC counts
Pressure Compensation:
Pcomp = a0 + b1 + c12 Tadc Padc + b2 Tadc
AN3785
Sensors
Freescale Semiconductor
19
Using the evaluation sequence shown in Section 3.3:
c12x2
= c12 * Tadc
a1
= b1 + c12x2 = -2.49512 + 0.44459
= -2.05052
a1x1
= a1 * Padc
= -2.05052 * 415
= -850.96785
y1
= a0 + a1x1
= 2107.875 + (-850.96785)
= 1256.90715
a2x2
= b2 * Tadc
= -1.02069 * 513
= -523.61444
PComp = y1 + a2x2
= 0.00086665 * 513
= 0.44459
= 1256.90715 + (-523.61444) = 733.29270
115 – 50
Pressure (kPa) = P comp ---------------------- + 50
1023
115 – 50
= 733.29 ---------------------- + 50
1023
= 96.59kPa
Coding Section
/*********************************************************\
* Calculate the compensated pressure PComp from the last set
* of ADC and coefficient values.
\*********************************************************/
sint16 mpl115a1_CalculatePComp(void)
{ uint16 Padc, Tadc;
sint16 a0, b1, b2, c12;
sint16 PComp;
// extract adc outputs
Padc = (mpl115a1_regs[0x00] << 8) | mpl115a1_regs[0x01];
Tadc = (mpl115a1_regs[0x02] << 8) | mpl115a1_regs[0x03];
// extract coefficients
a0 = (mpl115a1_regs[0x04] << 8) | mpl115a1_regs[0x05];
b1 = (mpl115a1_regs[0x06] << 8) | mpl115a1_regs[0x07];
b2 = (mpl115a1_regs[0x08] << 8) | mpl115a1_regs[0x09];
c12 = (mpl115a1_regs[0x0A] << 8) | mpl115a1_regs[0x0B];
// calculate internally compenstated PComp value using either version
//PComp = calculatePCompLong(Padc, Tadc, a0, b1, b2, c12);
PComp = calculatePCompShort(Padc, Tadc, a0, b1, b2, c12);
return (sint16)PComp;
}
/*********************************************************\
* Calculate the pressure in 1/16 kPa from a compensated
* PComp value.
\*********************************************************/
uint16 mpl115a1_CalculatePressure(sint16 PComp)
{
sint32 Pressure;
// The final step is to convert the internal PComp value into units of kPa.
// Pressure = PComp ∙ ((115.0 ‐ 50.0) / 1023.0) + 50
//
// The use of a floating point divide can be eliminated using the following approximation:
// Pressure = ( ( PComp ∙ 1041 ) >> 14 ) + 50
//
AN3785
Sensors
Freescale Semiconductor
20
// Note that in this implementation the final pressure value is reported with a 4 bit fractional
// part. This may be eliminated by right shifting the result four additional bits.
Pressure = ((((sint32)PComp) * 1041) >> 14) + 800;
return (uint16)Pressure;
}
/*********************************************************\
* Calculate the compensated pressure PComp value using the detailed description.
\*********************************************************/
sint16 calculatePCompLong(uint16 Padc, uint16 Tadc, sint16 a0, sint16 b1, sint16 b2, sint16 c12)
{
// TEMPORARY DATA VARIABLES:
sint32 lt1, lt2, lt3;
sint32 c12x2, a1, a1x1, y1, a2x2, PComp;
// Pressure calculation (long)
//============================
// This version of the pressure calculation function has the long description showing exaclty
// how the bit widths of the coefficients align through the calculation.
//
// Variables used to do large calculation as 3 temp variables in the process below
// signed long (sint32) lt1, lt2, lt3;
//
// Variables used for Pressure and Temperature Raw.
// unsigned short (uint16) Padc, Tadc.
//
// In order to optimize the fixed point arithmetic, each value is annotated with a descriptor x(N,F).
// x is 's' for signed or 'u' for unsigned
// N is the number of significant digits in the value
// F is the number of fractional bits, right of the decimal point
//
// Each of the input values and coefficients are identified below, based upon the coefficient bit
// width table in the data sheet:
// Padc : u(10.0)
// Tadc : u(10,0)
// a0 : s(16,3)
// b1 : s(16,13)
// b2 : s(16,14)
// c12 : s(16,24) // s(14,13) + 9 zero pad = s(16,15+9) => s(16,24) left justified
// PComp : s(16,4) // compensated pressure value contains a 8 bit integer part and a 4 bit fractional part
//
// The compensation formula is:
// PComp = a0 + (b1 + c12 ∙ Tadc) ∙ Padc + b2 ∙ Tadc
// The calculation can be broken down into individual steps Padc>>=6; //Note that the Padc is the raw value from Pegasus >>6 since its 10 bit unsigned
Tadc>>=6; //Note that the Tadc is the raw value from Pegasus >>6 since its 10 bit unsigned
//******* STEP 1 : c12x2 = c12 * Tadc
lt1 = c12; // s(16,24) // c12 is s(14,13)+9 zero pad = s(16,15)+9 => s(16,24) left justified
lt2 = (sint16)Tadc; // u(10,0)
lt3 = lt1 * lt2; // s(26,24) = c12 * Tadc
c12x2 = lt3 >> 11; // s(15,13) ‐ EQ 3 = c12x2
//******* STEP 2 : a1 = b1 + c12x2
lt1 = (sint16)b1; // s(16,13)
lt2 = c12x2; // s(15,13)
lt3 = lt1 + lt2; // s(16,13) = b1 + c12x2
a1 = lt3; // s(16,13) ‐ EQ 4 = a1
//******* STEP 3 : a1x1 = a1 * Padc
lt1 = a1; // s(16,13)
lt2 = (sint16)Padc; // u(10,0)
lt3 = lt1 * lt2; // s(26,13) = a1 * Padc
AN3785
Sensors
Freescale Semiconductor
21
a1x1 = lt3; // s(26,13) ‐ EQ 5 = a1x1
//******* STEP 4 y1 = a0 + a1x1
lt1 = ((sint32)a0) << 10; // s(26,13) shifted to match a1x1 F value to add. So s(16,3)<<10 = s(26,13)
lt2 = a1x1; // s(26,13)
lt3 = lt1 + lt2; // s(26,13) = a0 + a1x1
y1 = lt3; // s(26,13) ‐ EQ 6 = y1
//******* STEP 5 : a2x2 = b2 * Tadc
lt1 = (sint32)b2; // s(16,14)
lt2 = (sint32)Tadc; // u(10,0)
lt3 = lt1 * lt2; // s(26,14) = b2 * Tadc
a2x2 = lt3 >> 1; // s(25,13) ‐ EQ 7 = a2x2
//******* STEP 6 : PComp = y1 + a2x2
lt1 = y1; // s(26,13)
lt2 = a2x2; // s(25,13)
lt3 = lt1 + lt2; // s(26,13) = y1 + a2x2
PComp = lt3 >> 9; // s(17,4) ‐ EQ 8 = PComp return (sint16)PComp; // By calibration this is less than 16 bits
}
/*********************************************************\
* Calculate the compensated pressure PComp value using the brief version
\*********************************************************/
sint16 calculatePCompShort(uint16 Padc, uint16 Tadc, sint16 a0, sint16 b1, sint16 b2, sint16 c12)
{
sint32 c12x2, a1, a1x1, y1, a2x2, PComp;
// Pressure calculation (short)
//=============================
// This version of the pressure calculation function has the same function as the long
// version and gets exaclty the same result, but is implemented more succinctly.
Padc >>= 6; //Note that the Padc is the raw value from Pegasus >>6 since its 10 bit unsigned
Tadc >>= 6; //Note that the Tadc is the raw value from Pegasus >>6 since its 10 bit unsigned
c12x2 = (((sint32)c12) * Tadc) >> 11; // c12x2 = c12 * Tadc
a1 = (sint32)b1 + c12x2; // a1 = b1 + c12x2
a1x1 = a1 * Padc; // a1x1 = a1 * Padc
y1 = (((sint32)a0) << 10) + a1x1; // y1 = a0 + a1x1
a2x2 = (((sint32)b2) * Tadc) >> 1; // a2x2 = b2 * Tadc
PComp = (y1 + a2x2) >> 9; // PComp = y1 + a2x2
return (sint16)PComp;
}
AN3785
Sensors
Freescale Semiconductor
22
Table 7. Revision History
Revision
number
Revision
date
Description of changes
6
06/2012
• In Section 1, added Operating Characteristics table and figures for SPI and I2C Pin Connections,
in Section 2, updated Figures 7, 8 and 9.
AN3785
Sensors
Freescale Semiconductor
23
How to Reach Us:
Information in this document is provided solely to enable system and software
implementers to use Freescale products. There are no express or implied copyright
Home Page:
www.freescale.com
licenses granted hereunder to design or fabricate any integrated circuits based on the
information in this document.
Web Support:
http://www.freescale.com/support
Freescale reserves the right to make changes without further notice to any products
herein. Freescale makes no warranty, representation, or guarantee regarding the
suitability of its products for any particular purpose, nor does Freescale assume any
liability arising out of the application or use of any product or circuit, and specifically
disclaims any and all liability, including without limitation consequential or incidental
damages. “Typical” parameters that may be provided in Freescale data sheets and/or
specifications can and do vary in different applications, and actual performance may
vary over time. All operating parameters, including “typicals,” must be validated for each
customer application by customer’s technical experts. Freescale does not convey any
license under its patent rights nor the rights of others. Freescale sells products pursuant
to standard terms and conditions of sale, which can be found at the following address:
http://www.reg.net/v2/webservices/Freescale/Docs/TermsandConditions.htm.
Freescale, and the Freescale logo, are trademarks of Freescale Semiconductor, Inc.,
Reg. U.S. Pat. & Tm. Off. Xtrinsic is a trademark of Freescale Semiconductor, Inc.
All other product or service names are the property of their respective owners.
© 6/21/12 Freescale Semiconductor, Inc. All rights reserved.
RoHS-compliant and/or Pb-free versions of Freescale products have the functionality and electrical
characteristics of their non-RoHS-compliant and/or non-Pb-free counterparts. For further information,
see http:/www.freescale.com or contact your Freescale sales representative.
For information on Freescale’s Environmental Products program, go to http://www.freescale.com/epp.
AN3785
Rev. 6
06/2012