TCA9539 Low Voltage 16-Bit I2C and SMBus Low-Power I/O Expander with Interrupt Output, Reset Pin, and Configuration Registers

8.3.1 I/O Port

When an I/O is configured as an input, FETs Q1 and Q2 are off, which creates a high-impedance input. The input voltage may be raised above V_CC to a maximum of 5.5 V.

If the I/O is configured as an output, Q1 or Q2 is enabled, depending on the state of the Output Port register. In this case, there are low-impedance paths between the I/O pin and either V_CC or GND. The external voltage applied to this I/O pin must not exceed the recommended levels for proper operation.

8.3.2 RESET Input

A reset can be accomplished by holding the RESET pin low for a minimum of t_W. The TCA9539 registers and I²C/SMBus state machine are held in their default states until RESET is once again high. This input requires a pull-up resistor to V_CC, if no active connection is used.

8.3.3 Interrupt (INT) Output

An interrupt is generated by any rising or falling edge of the port inputs in the input mode. After time t_iv, the signal INT is valid. Resetting the interrupt circuit is achieved when data on the port is changed to the original setting or data is read from the port that generated the interrupt. Resetting occurs in the read mode at the acknowledge (ACK) bit after the rising edge of the SCL signal. Note that the INT is reset at the ACK just before the byte of changed data is sent. Interrupts that occur during the ACK clock pulse can be lost (or be very short) because of the resetting of the interrupt during this pulse. Each change of the I/Os after resetting is detected and is transmitted as INT.

Reading from or writing to another device does not affect the interrupt circuit, and a pin configured as an output cannot cause an interrupt. Changing an I/O from an output to an input may cause a false interrupt to occur if the state of the pin does not match the contents of the Input Port register. Because each 8-bit port is read independently, the interrupt caused by port 0 is not cleared by a read of port 1, or vice versa.

INT has an open-drain structure and requires a pull-up resistor to V_CC.

8.4.1 Power-On Reset

When power (from 0 V) is applied to V_CC, an internal power-on reset holds the TCA9539 in a reset condition until V_CC has reached V_{POR R}. At that point, the reset condition is released and the TCA9539 registers and I²C-SMBus state machine initialize to their default states. After that, V_CC must be lowered to below V_PORF and then back up to the operating voltage for a power-reset cycle.

8.5.1 I²C Interface

The TCA9539 has a standard bidirectional I²C interface that is controlled by a master device in order to be configured or read the status of this device. Each slave on the I²C bus has a specific device address to differentiate between other slave devices that are on the same I²C bus. Many slave devices require configuration upon startup to set the behavior of the device. This is typically done when the master accesses internal register maps of the slave, which have unique register addresses. A device can have one or multiple registers where data is stored, written, or read. For more information see Understanding the I²C Bus, SLVA704.

The physical I²C interface consists of the serial clock (SCL) and serial data (SDA) lines. Both SDA and SCL lines must be connected to V_CC through a pull-up resistor. The size of the pull-up resistor is determined by the amount of capacitance on the I²C lines. For further details, see I²C Pull-up Resistor Calculation, SLVA689. Data transfer may be initiated only when the bus is idle. A bus is considered idle if both SDA and SCL lines are high after a STOP condition. See Table 1.

Figure 25 and Figure 26 show the general procedure for a master to access a slave device:

1. If a master wants to send data to a slave:
   • Master-transmitter sends a START condition and addresses the slave-receiver.
   • Master-transmitter sends data to slave-receiver.
   • Master-transmitter terminates the transfer with a STOP condition.
2. If a master wants to receive or read data from a slave:
   • Master-receiver sends a START condition and addresses the slave-transmitter.
   • Master-receiver sends the requested register to read to slave-transmitter.
   • Master-receiver receives data from the slave-transmitter.
   • Master-receiver terminates the transfer with a STOP condition.

Figure 25. Definition of Start and Stop Conditions

Figure 26. Bit Transfer

Table 1 shows the interface definition.

8.6.1 Device Address

Figure 27 shows the address byte of the TCA9539.

Figure 27. TCA9539 Address

Table 2 shows the address reference of the TCA9539.

The last bit of the slave address defines the operation (read or write) to be performed. A high (1) selects a read operation, while a low (0) selects a write operation.

Note that the I²C addresses shown above are the 7-bit, right-justified hexadecimal values.

8.6.2 Control Register and Command Byte

Following the successful acknowledgment of the address byte, the bus master sends a command byte shown in Table 3 that is stored in the control register in the TCA9539. Three bits of this data byte state the operation (read or write) and the internal register (input, output, Polarity Inversion or Configuration) that is affected. This register can be written or read through the I²C bus. The command byte is sent only during a write transmission.

When a command byte has been sent, the register pair that was addressed continues to be accessed by reads until a new command byte has been sent. Figure 28 shows the control register bits.

Figure 28. Control Register Bits

8.6.3 Register Descriptions

The Input Port registers (registers 0 and 1) shown in Table 4 reflect the incoming logic levels of the pins, regardless of whether the pin is defined as an input or an output by the Configuration register. It only acts on read operation. Writes to these registers have no effect. The default value, X, is determined by the externally applied logic level.

Before a read operation, a write transmission is sent with the command byte to indicate to the I²C device that the Input Port register is accessed next. See the Writes section for more information and examples.

The Output Port registers (registers 2 and 3) shown in Table 5 show the outgoing logic levels of the pins defined as outputs by the Configuration register. Bit values in this register have no effect on pins defined as inputs. In turn, reads from this register reflect the value that is in the flip-flop controlling the output selection, not the actual pin value.

The Polarity Inversion registers (registers 4 and 5) shown in Table 6 allow Polarity Inversion of pins defined as inputs by the Configuration register. If a bit in this register is set (written with 1), the corresponding port pin's polarity is inverted. If a bit in this register is cleared (written with a 0), the corresponding port pin's original polarity is retained.

The Configuration registers (registers 6 and 7) shown in Table 7 configure the directions of the I/O pins. If a bit in this register is set to 1, the corresponding port pin is enabled as an input with a high-impedance output driver. If a bit in this register is cleared to 0, the corresponding port pin is enabled as an output.

8.6.3.1 Bus Transactions

Data is exchanged between the master and the TCA9539 through write and read commands, and this is accomplished by reading from or writing to registers in the slave device.

Registers are locations in the memory of the slave which contain information, whether it be the configuration information or some sampled data to send back to the master. The master must write information to these registers in order to instruct the slave device to perform a task.

8.6.3.1.1 Writes

To write on the I²C bus, the master sends a START condition on the bus with the address of the slave, as well as the last bit (the R/W bit) set to 0, which signifies a write. After the slave sends the acknowledge bit, the master then sends the register address of the register to which it wishes to write. The slave acknowledges again, letting the master know it is ready. After this, the master starts sending the register data to the slave until the master has sent all the data necessary (which is sometimes only a single byte), and the master terminates the transmission with a STOP condition.

See the Register Descriptions section to see list of the TCA9539s internal registers and a description of each one.

Figure 29 shows an example of writing a single byte to a slave register.

Figure 29. Write to Register

<br/>

Figure 30. Write to the Polarity Inversion Register

Figure 31. Write to Output Port Registers

8.6.3.1.2 Reads

Reading from a slave is very similar to writing, but requires some additional steps. In order to read from a slave, the master must first instruct the slave which register it wishes to read from. This is done by the master starting off the transmission in a similar fashion as the write, by sending the address with the R/W bit equal to 0 (signifying a write), followed by the register address it wishes to read from. When the slave acknowledges this register address, the master sends a START condition again, followed by the slave address with the R/W bit set to 1 (signifying a read). This time, the slave acknowledges the read request, and the master releases the SDA bus but continues supplying the clock to the slave. During this part of the transaction, the master becomes the master-receiver, and the slave becomes the slave-transmitter.

The master continues to send out the clock pulses, but releases the SDA line so that the slave can transmit data. At the end of every byte of data, the master sends an ACK to the slave, letting the slave know that it is ready for more data. When the master has received the number of bytes it is expecting, it sends a NACK, signaling to the slave to halt communications and release the bus. The master follows this up with a STOP condition.

See the Register Descriptions section for the list of the TCA9539s internal registers and a description of each one.

Figure 32 shows an example of reading a single byte from a slave register.

Figure 32. Read from Register

After a restart, the value of the register defined by the command byte matches the register being accessed when the restart occurred. For example, if the command byte references Input Port 1 before the restart, the restart occurs when Input Port 0 is being read. The original command byte is forgotten. If a subsequent restart occurs, Input Port 0 is read first. Data is clocked into the register on the rising edge of the ACK clock pulse. After the first byte is read, additional bytes may be read, but the data now reflect the information in the other register in the pair. For example, if Input Port 1 is read, the next byte read is Input Port 0.

Data is clocked into the register on the rising edge of the ACK clock pulse. There is no limitation on the number of data bytes received in one read transmission, but when the final byte is received, the bus master must not acknowledge the data.

A. Transfer of data can be stopped at any time by a Stop condition. When this occurs, data present at the latest acknowledge phase is valid (output mode). It is assumed that the command byte previously has been set to 00 (Read Input Port register).

B. This figure eliminates the command byte transfer, a restart, and slave address call between the initial slave address call and actual data transfer from the P port (see Figure 32 for these details).

Figure 33. Read Input Port Register, Scenario 1

<br/>

A. Transfer of data can be stopped at any time by a Stop condition. When this occurs, data present at the latest acknowledge phase is valid (output mode). It is assumed that the command byte previously has been set to 00 (Read Input Port register).

B. This figure eliminates the command byte transfer, a restart, and slave address call between the initial slave address call and actual data transfer from the P port (see Figure 32 for these details).

Figure 34. Read Input Port Register, Scenario 2

9.2.1 Design Requirements

9.2.1.1 Calculating Junction Temperature and Power Dissipation

When designing with this device, it is important that the Recommended Operating Conditions not be violated. Many of the parameters of this device are rated based on junction temperature. So junction temperature must be calculated in order to verify that safe operation of the device is met. The basic equation for junction temperature is shown in Equation 1.

Equation 1.

θ_JA is the standard junction to ambient thermal resistance measurement of the package, as seen in Thermal Information table. P_d is the total power dissipation of the device, and the approximation is shown in Equation 2.

Equation 2.

Equation 2 is the approximation of power dissipation in the device. The equation is the static power plus the summation of power dissipated by each port (with a different equation based on if the port is outputting high, or outputting low. If the port is set as an input, then power dissipation is the input leakage of the pin multiplied by the voltage on the pin). Note that this ignores power dissipation in the INT and SDA pins, assuming these transients to be small. They can easily be included in the power dissipation calculation by using Equation 3 to calculate the power dissipation in INT or SDA while they are pulling low, and this gives maximum power dissipation.

Equation 3.

Equation 3 shows the power dissipation for a single port which is set to output low. The power dissipated by the port is the V_OL of the port multiplied by the current it is sinking.

Equation 4.

Equation 4 shows the power dissipation for a single port which is set to output high. The power dissipated by the port is the current sourced by the port multiplied by the voltage drop across the device (difference between V_CC and the output voltage).

9.2.1.2 Minimizing I_CC When I/Os Control LEDs

When an I/O is used to control an LED, normally it is connected to V_CC through a resistor see Figure 35. Because the LED acts as a diode, when the LED is off, the I/O V_IN is about 1.2 V less than V_CC. The ΔI_CC parameter in the Electrical Characteristics table shows how I_CC increases as V_IN becomes lower than V_CC. For battery-powered applications, it is essential that the voltage of I/O pins is greater than or equal to V_CC, when the LED is off, to minimize current consumption.

Figure 36 shows a high-value resistor in parallel with the LED. Figure 37 shows V_CC less than the LED supply voltage by at least 1.2 V. Both of these methods maintain the I/O V_CC at or above V_CC and prevent additional supply-current consumption when the LED is off.

Figure 36. High-Value Resistor in Parallel with LED

Figure 37. Device Supplied by Lower Voltage

9.2.2 Detailed Design Procedure

The pull-up resistors, R_P, for the SCL and SDA lines need to be selected appropriately and take into consideration the total capacitance of all slaves on the I²C bus. The minimum pull-up resistance is a function of V_CC, V_OL,(max), and I_OL as shown in Equation 5.

Equation 5.

The maximum pull-up resistance is a function of the maximum rise time, t_r (300 ns for fast-mode operation,
f_SCL = 400 kHz) and bus capacitance, C_b, see Equation 6.

Equation 6.

The maximum bus capacitance for an I²C bus must not exceed 400 pF for standard-mode or fast-mode operation. The bus capacitance can be approximated by adding the capacitance of the TCA9554A, C_i for SCL or C_io for SDA, the capacitance of wires, connections and traces, and the capacitance of additional slaves on the bus.

9.2.3 Application Curves

Standard-mode	Fast-mode
(fSCL= 100 kHz, tr = 1 µs)	(fSCL= 400 kHz, tr= 300 ns)

Figure 38. Maximum Pull-Up Resistance (R_p(max)) vs Bus Capacitance (C_b)

VOL = 0.2 × VCC, IOL = 2 mA when VCC ≤ 2 V

VOL = 0.4 V, IOL = 3 mA when VCC > 2 V

Figure 39. Minimum Pull-Up Resistance (R_p(min)) vs Pull-Up Reference Voltage (V_CC)