.. | ||
.github | ||
examples | ||
test | ||
.arduino-ci.yml | ||
CHANGELOG.md | ||
CMakeLists.txt | ||
keywords.txt | ||
library.json | ||
library.properties | ||
LICENSE | ||
MCP23x17_registers.h | ||
MCP23017.cpp | ||
MCP23017.h | ||
README.md |
MCP23017_RT
Arduino library for MCP23017 16 channel I2C port expander.
Description
This library gives easy control over the 16 pins of a (I2C) MCP23017 chip.
This IC is strongly related to the MCP23S17 SPI port expander - https://github.com/RobTillaart/MCP23S17 Programming Interface is kept the same as much as possible.
The write1(pin, value) is optimized. If a pin is not changed it will not be written again to save time.
REV D - June 2022
The I2C IO expander MCP23017 has changed according to the new data sheet. It is now a 14/16-bit IO expander. The pins GPA7 (7) and GPB7 (15) have lost their input mode, output mode still works. The chips look the same and did not change names. This implies it not possible to read 8 bits in parallel at the exact same moment any more. The REV D version now need reading both A and B register to get 8 bits parallel (with a minor delay).
Some details see:
- https://hackaday.com/2023/02/03/mcp23017-went-through-shortage-hell-lost-two-inputs/
- https://ww1.microchip.com/downloads/aemDocuments/documents/APID/ProductDocuments/DataSheets/MCP23017-Data-Sheet-DS20001952.pdf
- https://microchip.my.site.com/s/article/GPA7---GPB7-Cannot-Be-Used-as-Inputs-In-MCP23017
- https://www.elektormagazine.com/articles/breakout-board-mcp23017 (comment)
- https://forums.raspberrypi.com/viewtopic.php?t=91209&sid=f8c6df7c8ede76937a66503edfe25394
Note: the library has no provisions (yet) for detecting DEV D chips or handle them in a special way. There is an idea to implement a derived class MCP23017_REVD that provides automatic support. However low priority.
Note that the MCP23S017 (SPI version) does not have this "feature" for GPA7 and GPB7.
0.6.0 Breaking change
The version 0.6.0 has breaking changes in the interface. The rationale is that the programming environment of the Arduino ESP32 S3 board uses a remapping by means of the include file io_pin_remap.h. This file remaps the pins of several core Arduino functions. The remapping is implemented by #define macros and these implement "hard" text replacements without considering context. The effect is that methods from this class (and several others) which have the same name as those Arduino core functions will be remapped into something not working.
The following library functions have been renamed:
old name | new name | notes |
---|---|---|
analogRead() | read() | |
analogWrite() | write() | |
pinMode() | pinMode1() | |
digitalRead() | read1() | |
digitalWrite() | write1() |
0.5.0 Breaking change
Version 0.5.0 introduced a breaking change. You cannot set the pins in begin() any more. This reduces the dependency of processor dependent Wire implementations. The user has to call Wire.begin() and can optionally set the Wire pins before calling begin().
I2C
Supports 100kHz, 400kHz and 1.7MHz
TODO - add performance data
I2C multiplexing
Sometimes you need to control more devices than possible with the default address range the device provides. This is possible with an I2C multiplexer e.g. TCA9548 which creates up to eight channels (think of it as I2C subnets) which can use the complete address range of the device.
Drawback of using a multiplexer is that it takes more administration in your code e.g. which device is on which channel. This will slow down the access, which must be taken into account when deciding which devices are on which channel. Also note that switching between channels will slow down other devices too if they are behind the multiplexer.
Related
16 bit port expanders
- https://github.com/RobTillaart/MCP23017_RT
- https://github.com/RobTillaart/MCP23S17
- https://github.com/RobTillaart/PCF8575
- https://github.com/RobTillaart/TCA9555
8 bit port expanders
- https://github.com/RobTillaart/MCP23008
- https://github.com/RobTillaart/MCP23S08
- https://github.com/RobTillaart/PCF8574
Interface
#include "MCP23017.h"
Constructor
- MCP23017(uint8_t address, TwoWire *wire = &Wire) constructor, with default Wire interface.
Can be overruled with Wire0..WireN. - bool begin(bool pullup = true) initializes library, returns true if successful. Default sets the pins to INPUT PULLUP. Returns false if not connected or a register could not be set.
- bool isConnected() returns true if connected, false otherwise.
- uint8_t getAddress() returns the address set in the constructor.
Single pin interface
Please note REVD remarks at top.
- bool pinMode1(uint8_t pin, uint8_t mode) pin = 0..15, mode = INPUT, OUTPUT. Returns true if successful.
- bool write1(uint8_t pin, uint8_t value) pin = 0..15, value = LOW(0) HIGH (!0). Returns true if successful.
- uint8_t read1(uint8_t pin) pin = 0..15, returns LOW or HIGH, might set the lastError();
- bool setPolarity(uint8_t pin, bool reversed) pin = 0..15, set reversed flag. Returns true if successful.
- bool getPolarity(uint8_t pin, bool &reversed) pin = 0..15, reads reversed flag. Returns true if successful.
- bool setPullup(uint8_t pin, bool pullup) pin = 0..15, set pull-up flag. Returns true if successful.
- bool getPullup(uint8_t pin, bool &pullup) pin = 0..15, reads pull-up flag. Returns true if successful.
8 pins interface
Please note REVD remarks at top.
- bool pinMode8(uint8_t port, uint8_t value) port = 0..1, value = 0..255. Returns true if successful.
- bool write8(uint8_t port, uint8_t value) port = 0..1, value = 0..255. Returns true if successful.
- uint8_t read8(uint8_t port) port = 0..1, reads 8 pins into one byte.
- bool setPolarity8(uint8_t port, uint8_t mask) port = 0..1, sets polarity for 8 channels at once. Returns true if successful.
- bool getPolarity8(uint8_t port, uint8_t &mask) port = 0..1, reads polarity of 8 channels at once. Returns true if successful.
- bool setPullup8(uint8_t port, uint8_t mask) port = 0..1, sets pull-up for 8 channels at once. Returns true if successful.
- bool getPullup8(uint8_t port, uint8_t &mask) port = 0..1, reads pull-up for 8 channels at once. Returns true if successful.
16 pins interface
Please note REVD remarks at top.
- bool pinMode16(uint16_t value) value = 0..0xFFFF, returns true if successful.
- bool write16(uint16_t value) value = 0..0xFFFF, returns true if successful.
- uint16_t read16() reads 16 pins into an uint16_t.
- bool setPolarity16(uint16_t mask) sets polarity for 16 channels. Returns true if successful.
- bool getPolarity16(uint16_t &mask) reads polarity of 16 channels. Returns true if successful.
- bool setPullup16(uint16_t mask) sets pull-up for 16 channels. Returns true if successful.
- bool getPullup16(uint16_t &mask) reads pull-up for 16 channels. Returns true if successful.
The reading and writing to registers have been performance optimized for the 16 bit interface. If there are problems please open an issue.
Interrupts (experimental 0.6.3)
Read the datasheet for the details, page 24,25.
Note: Error handling is limited.
pin = 0..15
mode = { RISING, FALLING, CHANGE }
- bool enableInterrupt(uint8_t pin, uint8_t mode) Returns true if successful. Returns MCP23017_PIN_ERROR if pin > 15.
- bool disableInterrupt(uint8_t pin) Returns true if successful. Returns MCP23017_PIN_ERROR if pin > 15.
16 pin interface, overrides all earlier settings. Sets all pins to the same interrupt mode { RISING, FALLING, CHANGE }.
- bool enableInterrupt16(uint16_t mask, uint8_t mode) mask = 0x0000..0xFFFF.
- bool disableInterrupt16(uint16_t mask)
Determine which pins caused the Interrupt. (datasheet).
-
uint16_t getInterruptFlagRegister() Reads all 16 pins.
-
uint16_t getInterruptCaptureRegister() Reads all 16 pins. Is used to detect if multiple pins triggered an interrupt.
-
bool setInterruptPolarity(uint8_t ipol) polarity: 0 = LOW, 1 = HIGH, 2 = NONE/ODR
-
uint8_t getInterruptPolarity() return set value.
Merge INTA and INTB into one signal so only one line handles all interrupts. This reduces the number of interrupt lines to handle, however one has to read more registers to find the changed ones.
- bool mirrorInterrupts(bool on) enables / disables mirror mode.
- bool isMirroredInterrupts() returns set option (0,1 or 2).
IO Control Register
The library supports setting bit fields in the IO control register. Read the datasheet carefully!
- bool enableControlRegister(uint8_t mask) set IOCR bit fields
- bool disableControlRegister(uint8_t mask) clear IOCR bit fields
constant | mask | description |
---|---|---|
MCP23x17_IOCR_BANK | 0x80 | Controls how the registers are addressed. |
MCP23x17_IOCR_MIRROR | 0x40 | INT Pins Mirror bit. |
MCP23x17_IOCR_SEQOP | 0x20 | Sequential Operation mode bit. |
MCP23x17_IOCR_DISSLW | 0x10 | Slew Rate control bit for SDA output. |
MCP23x17_IOCR_HAEN | 0x08 | Hardware Address Enable bit (MCP23S17 only). |
MCP23x17_IOCR_ODR | 0x04 | Configures the INT pin as an open-drain output. |
MCP23x17_IOCR_INTPOL | 0x02 | This bit sets the polarity of the INT output pin. |
MCP23x17_IOCR_NI | 0x01 | Not implemented. |
Error codes
If one of the above functions return false, there might be an error.
- int lastError() Above functions set an error flag that can be read with this function.
Reading it will reset the flag to MCP23017_OK.
name | value | description |
---|---|---|
MCP23017_OK | 0x00 | No error |
MCP23017_PIN_ERROR | 0x81 | |
MCP23017_I2C_ERROR | 0x82 | (compatibility) |
MCP23017_VALUE_ERROR | 0x83 | |
MCP23017_PORT_ERROR | 0x84 | |
MCP23017_REGISTER_ERROR | 0xFF | low level. |
MCP23017_INVALID_READ | 0xFF | low level. |
Future
Must
- Improve and extend documentation
- add examples
Should
- keep functional in sync
- sync error codes to MCP23x17
- buy additional hardware
- test with multiple devices.
- multi SELECT lines
- add example with interrupts
- test
- extend error codes
- optimize code - squeeze footprint
- fix TODO's in code
- investigate if REV D chips can be detected.
Could
- check need for writing in all functions (Polarity / Pull-up)
- check if bit mask changes.
- what is performance gain vs footprint?
- investigate and reimplement the INPUT_PULLUP for pinMode() ?
- initial value (16 bit?) as begin parameter (breaking change)
- depends on input output pull-up etc.
- create a derived class MCP23017_REVD
Wont
Support
If you appreciate my libraries, you can support the development and maintenance. Improve the quality of the libraries by providing issues and Pull Requests, or donate through PayPal or GitHub sponsors.
Thank you,