View previous topic :: View next topic |
Author |
Message |
siripot12
Joined: 03 Dec 2016 Posts: 3
|
Input Capture Problem with pic24FJ128FB110 |
Posted: Sat Dec 03, 2016 3:06 am |
|
|
Hi Everyone, I have a problem with pic24FJ128GB110 on Input capture interrupt(INT_ICx). I want to use this function to measure the period of PWM.
So, I have set following EX_PinSelect2.c by using Input Capture channel 5 (IC5) but when I inject a pulse to IC5 Pin the controller not jump to #INT_IC5 loop.
Is it correct? If it not correct how can I do?
This is my code.
Code: | #include <24FJ128GB110.h>
#fuses HS,NOWDT,NOPROTECT,PR
#use delay(clock=8000000)
#pin_select U1TX=PIN_B2
#pin_select U1RX=PIN_E9
#pin_select INT1 = PIN_B0 // Setup External Interrupt INT_EXT1
#pin_select IC5 = PIN_C1 // setup Input Capture INT_IC4
#use rs232(baud=9600, UART1)
#INT_EXT1
void external_interrupt1(void)
{
printf("\n\r EXT1 Trigger");
}
#INT_IC5
void Input_Capture1(void)
{
printf("\n\r IC5 Trigger");
}
void main()
{
printf("\n\rRESET");
// Following setup required for Input Capture
setup_timer2(TMR_INTERNAL|TMR_DIV_BY_1); // Start timer 2
setup_capture(5,CAPTURE_EE|CAPTURE_TIMER2);
enable_interrupts(INT_EXT1);
enable_interrupts(INT_IC5);
enable_interrupts(INTR_GLOBAL);
while(1)
{
delay_ms(1000);
}
} |
Thank you. |
|
|
Ttelmah
Joined: 11 Mar 2010 Posts: 19553
|
|
Posted: Sat Dec 03, 2016 4:21 am |
|
|
There are a few little 'bewares' in what you have posted:
1) When using the hardware UART, you should always have 'ERRORS' in the #USE RS232, unless you are adding your own error handler code. CCS (unfortunately) don't do this in a lot of their examples....
2) On a PIC24, 8MHz, requires the XT oscillator, not HS. You'll probably be over-driving the oscillator a bit. Minimum for HS is 10Mhz. Same comment applies here about the CCS example!....
However none of these should prevent the test working.
What is your compiler version?. (You should _always_ tell us this, for any problem like this, since we can then compile the code and see if the compiler is setting the registers correctly).
Does INT_EXT1 work correctly?.
How old is your chip?. Devrev?.
There was an issue in the first release chips with input capture. 2008 erratum sheet.
How are you testing this?.
MPLAB SIM has an issue with these chips on some versions and some capture channels.
Tell us the compiler version, and we can then have a look at the code being generated and see if the compiler is doing something wrong. |
|
|
temtronic
Joined: 01 Jul 2010 Posts: 9246 Location: Greensville,Ontario
|
|
Posted: Sat Dec 03, 2016 6:37 am |
|
|
query...
is this...
enable_interrupts(INTR_GLOBAL);
correct ?
I don't use that PIC, so curious as to the syntax...
enable_interrupts(GLOBAL);
is what I've always used .
Jay |
|
|
siripot12
Joined: 03 Dec 2016 Posts: 3
|
|
Posted: Sat Dec 03, 2016 7:11 am |
|
|
Ttelmah wrote: | There are a few little 'bewares' in what you have posted:
1) When using the hardware UART, you should always have 'ERRORS' in the #USE RS232, unless you are adding your own error handler code. CCS (unfortunately) don't do this in a lot of their examples....
2) On a PIC24, 8MHz, requires the XT oscillator, not HS. You'll probably be over-driving the oscillator a bit. Minimum for HS is 10Mhz. Same comment applies here about the CCS example!....
However none of these should prevent the test working.
What is your compiler version?. (You should _always_ tell us this, for any problem like this, since we can then compile the code and see if the compiler is setting the registers correctly).
Does INT_EXT1 work correctly?.
How old is your chip?. Devrev?.
There was an issue in the first release chips with input capture. 2008 erratum sheet.
How are you testing this?.
MPLAB SIM has an issue with these chips on some versions and some capture channels.
Tell us the compiler version, and we can then have a look at the code being generated and see if the compiler is doing something wrong. |
Thank you for your reply.
-I have tried to change from HS to XT but it still not working.
-And the code that I posted for INT_EXT1 working properly but only input compare not working.
-And for the old of the chip I don't know for exact old because I bought from the shop 2 week ago.
my code is complied on CCS version 4.104. |
|
|
siripot12
Joined: 03 Dec 2016 Posts: 3
|
|
Posted: Sat Dec 03, 2016 7:16 am |
|
|
temtronic wrote: | query...
is this...
enable_interrupts(INTR_GLOBAL);
correct ?
I don't use that PIC, so curious as to the syntax...
enable_interrupts(GLOBAL);
is what I've always used .
Jay |
I think enable_interrupts(INTR_GLOBAL) already correct because in my header file of "24FJ128GB110.h" don't have "GLOBAL". Thank for your reply ^_^. I'm waiting for help T_T |
|
|
Ttelmah
Joined: 11 Mar 2010 Posts: 19553
|
|
Posted: Sat Dec 03, 2016 7:41 am |
|
|
temtronic wrote: | query...
is this...
enable_interrupts(INTR_GLOBAL);
correct ?
I don't use that PIC, so curious as to the syntax...
enable_interrupts(GLOBAL);
is what I've always used .
Jay |
INTR_GLOBAL, is really 'more correct'. The comment again 'GLOBAL' on PIC24's, says 'for reverse compatibility only'....
Ouch. Do you really mean 4.104!....
I'd suspect the #pin select might well not work right for anything but the most basic pins on a compiler this old.
Will try and see. |
|
|
temtronic
Joined: 01 Jul 2010 Posts: 9246 Location: Greensville,Ontario
|
|
Posted: Sat Dec 03, 2016 9:33 am |
|
|
It was the 'R' that caught my eye...
Seems all the other INTerrupt syntax is INT_xxx, so from a 'syntax' or 'visual' perspective, I'd have thought INT_GLOBAL would have been better ?
But it boils down to what's in the header that the compiler wants....
Jay |
|
|
Ttelmah
Joined: 11 Mar 2010 Posts: 19553
|
|
Posted: Sat Dec 03, 2016 12:56 pm |
|
|
The standard 'level enables' for a PIC 24, are:
Code: |
#define INTR_GLOBAL 0x400
#define INTR_NORMAL 0x100
#define INTR_ALTERNATE 0x200
#define INTR_LEVEL0 0x500
#define INTR_LEVEL1 0x501
#define INTR_LEVEL2 0x502
#define INTR_LEVEL3 0x503
#define INTR_LEVEL4 0x504
#define INTR_LEVEL5 0x505
#define INTR_LEVEL6 0x506
#define INTR_LEVEL7 0x507
|
The V5 compilers then add:
#define INT_GLOBAL INTR_GLOBAL //for reverse compatibility
On the V4 compilers, you always had to use INTR_GLOBAL for the global enable.
Anyway that is not the problem. I'd suspect the compiler version is.
I don't think that the #pin_select was fully working till about V5.020. Even more recently than that, there were problems on some chips with some interrupts. |
|
|
|