It will flash the 2 LEDs @ c.5 & c.4
THEN HALTS without ANY i2c activity.
But will not do any i2c work, originally had the LED @ c.5 & c.4 where the LEDs DID flash so we know the pins do work.
I know that the fix is something really simple but it alludes me.
The MCP23017 is at 0x26 +R/W so that's 0x4c
' 2 Leds @ c.5 & c.4
' i2c bus @
' I2C_CLOCK PORTc.6
' I2C_DATA PORTc.7
'Chip Settings
#chip 16F1829, 4
'Defines (Constants)
'I2C settings
#define I2C_MODE Master
#define I2C_DATA PORTc.7
#define I2C_CLOCK PORTc.6
#define BlinkA portC.5
#define BlinkB portC.4
' original i2c bus connector
'#define I2C_DATA PORTb.5
'#define I2C_CLOCK PORTb.7
#define I2C_Read_BIT = 1
#define I2C_Write_BIT = 0
Dir i2c_data Out
Dir i2c_clock Out
'Variables
Dim xB As byte
ADCON0 = 0
ADCON1 = 0
clrf ANSELa
clrf ANSELb
clrf ANSELc
clrf LATa
clrf LATb
clrf LATc
'Turn off Comparators
CM1CON0 = 0
CM1CON1 = 0
CM2CON0 = 0
CM2CON1 = 0
trisC = 0
Start:
do
'Blink some lights
for xb = 1 to 30
BlinkA = 1
BlinkB = 0
wait 100 ms
BlinkA = 0
BlinkB = 1
wait 100 ms
next xb
BlinkA = 0
BlinkB = 0
' wait 500 ms
' loop
I2CStart
I2CSend 0x4c 'i2cDevice + I2C_Write_BIT ' Who + how
I2CSend 0x00 'dira 'i2cAddress ' Where to put it
I2CSend 0x00 'all out 'i2cData ' What to put there
I2CStop
wait 100 ms
I2CStart
I2CSend 0x4c 'i2cDevice + I2C_Write_BIT ' Who + how
I2CSend 0x12 'dira 'i2cAddress ' Where to put it
I2CSend 0xFF 'all out 'i2cData ' What to put there
I2CStop
wait 250 ms
I2CStart
I2CSend 0x4c 'i2cDevice + I2C_Write_BIT ' Who + how
I2CSend 0x12 'dira 'i2cAddress ' Where to put it
I2CSend 0x00 'all out 'i2cData ' What to put there
I2CStop
wait 200 ms
Loop
end
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think the problem with the I2C communication starts with how the MCP23017 is addressed. I have to say the address scheme looks different from what I'm used to. This chip has to write to a config register (IOCON,HAEN) in order to activate the address pins A1-A3, and is reset to zero every POR and RST!
So the default address is 0x40 plus the R/W bit, just stick with that unless there is a reason. Otherwise, a write to the IOCON,HAEN would be required to initialize the MCP23017 A1-A3 pins prior to main, that's just weird.
What value of pullup resistors are on the SDA, SCL pins?
EDIT: Oops, was looking at the SPI (wrong) addressing diagram.
Last edit: kent_twt4 2013-09-28
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
KENT:
You were correct about the pull ups when I moved the ports away from my original placement of Sck = b.6 & Sda =b.5, the pullups were on the motherboard, not the i2c_IO_board. Need to move them to the last i2c_IO_board in the chain, RTfM, dooh!
RTfM= read the FINE manual :)
Issue # 2 is using HSERsend or hserprint causes Sck to be permanently allocated, so that caused some tail chasing, fixed that with:
saving rcsta
shutdown the RS232 serial port
i2C stuff
then turning it back on.
I'm sure that will cause havoc when I change the i2c port to somewhere else :).
Again thanks for the whack upside the head. :)
This Now Works!
#define I2C_Read_BIT = 1
#define I2C_Write_BIT = 0
...Main code ...
xb = 0
i2cWrite (IO_Chip6, IODIRA, xb) ' make portA outputs
i2cWrite (IO_Chip6, IODIRB, xb) ' make portB outputs
xb = 0xFF
i2cWrite (IO_Chip6, GPIOA, xb) ' turn all pins on
i2cWrite (IO_Chip6, GPIOB, xb) ' turn all pins on
Wait 250 ms
xb = 0xAA '170
i2cWrite (IO_Chip6, GPIOA, xb) ' swap pins on/off
i2cWrite (IO_Chip6, GPIOB, xb) '
zb= i2cread (IO_Chip6, GPIOA) ' make portA outputs
HSerPrint " GPIOA= ": HSerprint xB : HSerPrint " " : HSerprint zB ' make portA outputs
zb= i2cread (IO_Chip6, GPIOB) ' make portB outputs
HSerPrint " GPIOB= ": HSerprint xB : HSerPrint " " : HSerprint zB ' make portA outputs
Wait 250 ms
xb = 0x55 '85 '
i2cWrite (IO_Chip6, GPIOA, xb) ' swap pins on/off
i2cWrite (IO_Chip6, GPIOB, xb) '
zb= i2cread (IO_Chip6, GPIOA) ' make portA outputs
HSerPrint " GPIOA= ": HSerprint xB : HSerPrint " " : HSerprint zB ' make portA outputs
zb= i2cread (IO_Chip6, GPIOB) ' make portB outputs
HSerPrint " GPIOB= ": HSerprint xB : HSerPrint " " : HSerprint zB ' make portA outputs
Wait 250 ms
xb = 0x00 '0 '
i2cWrite (IO_Chip6, GPIOA, xb) ' turn all pins oFF
i2cWrite (IO_Chip6, GPIOB, xb) ' turn all pins oFF
zb= i2cread (IO_Chip6, GPIOA) ' make portA outputs
HSerPrint " GPIOA= ": HSerprint xB : HSerPrint " " : HSerprint zB ' make portA outputs
zb= i2cread (IO_Chip6, GPIOB) ' make portB outputs
HSerPrint " GPIOB= ": HSerprint xB : HSerPrint " " : HSerprint zB ' make portA outputs
Wait 250 ms
.....
sub i2cWrite (i2cDevice, i2cAddress, i2cData) #NR
reg_status = RCSTA
rcsta = 0 ' shutdown the serial port to share with the i2c bus
Rotate i2cDevice left Simple' *2 = << 1 ' is shifted left 1 because of The RW bit: $26 << 1 = $4C
i2cstart
I2CSend i2cDevice + I2C_Write_BIT ' Who + how
I2CSend i2cAddress ' Where to put it
I2CSend i2cData ' What to put there
I2CStop
RCSTA = reg_status
end sub
Function i2cRead (i2cDevice, i2cAddress) 'fixed'
Rotate i2cDevice left Simple ' *2 = << 1 ' is shifted left 1 because of The RW bit: $26 << 1 = $4C
reg_status = RCSTA
rcsta = 0 ' shutdown the serial port to share with the i2c bus
I2CStart
I2CSend i2cDevice + I2C_Write_BIT ' Who + how, Tell it where to read
I2CSend i2cAddress ' Where to GET it, Give it the location
I2CStop
I2CStart
I2CSend i2cDevice + I2C_Read_BIT ' Who + how, Tell it you want to read
I2CReceive i2cData ' What to GET from there
I2CStop
i2cRead = i2cData
RCSTA = reg_status
end function
Last edit: ofuzzy1 2013-09-28
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The MCP23017 is really easy to use, and has a whole lot of features, unlike the PCF8575.
Just set up ONE sets of registers just like a PIC.
1. Direction port IODIR portA=0x00 or portB=0x01
2. Then move the Data through GPIO portA=0x12 or portB=0x13
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Kent asks: You have to use HSER because the free Tiny Bootloader version doesn't support software serial?
Ayup.
And that is fine, I just need to move the i2c away from the HSER port.
The ALT location c.4=[EN - CS] & c.5=[RS] is being used by the LCD.
So both locations are important.
I guess I could use the LCD d4-7 pins c.0-.3 while not trying to do a LCD write :)
Hold the LCD EN (CS) low and I can torment all the other lines.
-- This will allow 3 sets of i2c Busses :) c.0=Sck and c.1.2.3 as the Sda for i2c.1 .2 .3 -- that would be a whole lot of IO! 8163 =384
Oh but wait I could also use the RS line to that make it 512, now that is just plain ridiculous. :0)
Getting loopy here time for bed.
Again, thanks for the help.
G'nite.
Last edit: ofuzzy1 2013-09-28
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Also, the 23017 comes in DIP28 or soic28 for easy and almost easy prototyping work @$1.20. The 8575 is only soic @$1.50.
Prices from my order in 2012. 23017 is cheaper and more fun.
newark.com
Last edit: ofuzzy1 2013-09-28
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It will flash the 2 LEDs @ c.5 & c.4
THEN HALTS without ANY i2c activity.
But will not do any i2c work, originally had the LED @ c.5 & c.4 where the LEDs DID flash so we know the pins do work.
I know that the fix is something really simple but it alludes me.
The MCP23017 is at 0x26 +R/W so that's 0x4c
' 2 Leds @ c.5 & c.4 ' i2c bus @ ' I2C_CLOCK PORTc.6 ' I2C_DATA PORTc.7 'Chip Settings #chip 16F1829, 4 'Defines (Constants) 'I2C settings #define I2C_MODE Master #define I2C_DATA PORTc.7 #define I2C_CLOCK PORTc.6 #define BlinkA portC.5 #define BlinkB portC.4 ' original i2c bus connector '#define I2C_DATA PORTb.5 '#define I2C_CLOCK PORTb.7 #define I2C_Read_BIT = 1 #define I2C_Write_BIT = 0 Dir i2c_data Out Dir i2c_clock Out 'Variables Dim xB As byte ADCON0 = 0 ADCON1 = 0 clrf ANSELa clrf ANSELb clrf ANSELc clrf LATa clrf LATb clrf LATc 'Turn off Comparators CM1CON0 = 0 CM1CON1 = 0 CM2CON0 = 0 CM2CON1 = 0 trisC = 0 Start: do 'Blink some lights for xb = 1 to 30 BlinkA = 1 BlinkB = 0 wait 100 ms BlinkA = 0 BlinkB = 1 wait 100 ms next xb BlinkA = 0 BlinkB = 0 ' wait 500 ms ' loop I2CStart I2CSend 0x4c 'i2cDevice + I2C_Write_BIT ' Who + how I2CSend 0x00 'dira 'i2cAddress ' Where to put it I2CSend 0x00 'all out 'i2cData ' What to put there I2CStop wait 100 ms I2CStart I2CSend 0x4c 'i2cDevice + I2C_Write_BIT ' Who + how I2CSend 0x12 'dira 'i2cAddress ' Where to put it I2CSend 0xFF 'all out 'i2cData ' What to put there I2CStop wait 250 ms I2CStart I2CSend 0x4c 'i2cDevice + I2C_Write_BIT ' Who + how I2CSend 0x12 'dira 'i2cAddress ' Where to put it I2CSend 0x00 'all out 'i2cData ' What to put there I2CStop wait 200 ms Loop endI think the problem with the I2C communication starts with how the MCP23017 is addressed. I have to say the address scheme looks different from what I'm used to. This chip has to write to a config register (IOCON,HAEN) in order to activate the address pins A1-A3, and is reset to zero every POR and RST!
So the default address is 0x40 plus the R/W bit, just stick with that unless there is a reason. Otherwise, a write to the IOCON,HAEN would be required to initialize the MCP23017 A1-A3 pins prior to main, that's just weird.
What value of pullup resistors are on the SDA, SCL pins?
EDIT: Oops, was looking at the SPI (wrong) addressing diagram.
Last edit: kent_twt4 2013-09-28
KENT:
You were correct about the pull ups when I moved the ports away from my original placement of Sck = b.6 & Sda =b.5, the pullups were on the motherboard, not the i2c_IO_board. Need to move them to the last i2c_IO_board in the chain, RTfM, dooh!
RTfM= read the FINE manual :)
Issue # 2 is using HSERsend or hserprint causes Sck to be permanently allocated, so that caused some tail chasing, fixed that with:
saving rcsta
shutdown the RS232 serial port
i2C stuff
then turning it back on.
I'm sure that will cause havoc when I change the i2c port to somewhere else :).
Again thanks for the whack upside the head. :)
This Now Works!
Last edit: ofuzzy1 2013-09-28
The MCP23017 is really easy to use, and has a whole lot of features, unlike the PCF8575.
Just set up ONE sets of registers just like a PIC.
1. Direction port IODIR portA=0x00 or portB=0x01
2. Then move the Data through GPIO portA=0x12 or portB=0x13
That MCP23017 is certainly an interesting part.
My BAD, looking again the ICON,HAEN only applies the the "SPI" addressing scheme, head slap, now that makes more sense.
You have to use HSER because the free Tiny Bootloader version doesn't support software serial?
Kent asks: You have to use HSER because the free Tiny Bootloader version doesn't support software serial?
Ayup.
And that is fine, I just need to move the i2c away from the HSER port.
The ALT location c.4=[EN - CS] & c.5=[RS] is being used by the LCD.
So both locations are important.
I guess I could use the LCD d4-7 pins c.0-.3 while not trying to do a LCD write :)
Hold the LCD EN (CS) low and I can torment all the other lines.
-- This will allow 3 sets of i2c Busses :) c.0=Sck and c.1.2.3 as the Sda for i2c.1 .2 .3 -- that would be a whole lot of IO! 8163 =384
Oh but wait I could also use the RS line to that make it 512, now that is just plain ridiculous. :0)
Getting loopy here time for bed.
Again, thanks for the help.
G'nite.
Last edit: ofuzzy1 2013-09-28
Also, the 23017 comes in DIP28 or soic28 for easy and almost easy prototyping work @$1.20. The 8575 is only soic @$1.50.
Prices from my order in 2012. 23017 is cheaper and more fun.
newark.com
Last edit: ofuzzy1 2013-09-28