 |
 |
| View previous topic :: View next topic |
| Author |
Message |
Mike Walne
Joined: 19 Feb 2004
Posts: 1785
Location: Boston Spa UK
|
| Strange MPSIM for LATB & PORTB with PIC18F26K22 [SOLVED] |
 Posted: Mon Feb 27, 2017 11:55 am |
 |
|
Is this a known MPLAB SIM problem, or am I missing something?
I'm wanting to drive a stepper motor and started from the CCS example.
When I simulated in MPLAB for a PIC18F26K22 I got rubbish with CCS code.
I've played with the code and still get gibberish with the '26' device.
MPLAB is fine with PIC18F46K22 and PIC18F252 parts.
The real thing behaves as expected.
I've used the code below. On each line I've shown both '26' '46' outputs after the first time round.
I'm showing the states before execution of the instruction, i.e. where the cursor is pointing.
I'm using MPLAB IDE v8.92 and CCS v5.059
The disassembly listing for both '*K22' parts more or less agree.
| Code: | // Port B Test
#include <18F26K22.h>
//#include <18F46K22.h>
#fuses ECH,NOWDT,NOPROTECT,NOLVP,NOPLLEN
//#include <18F252.h>
//#fuses HS,NOWDT,NOPROTECT,NOLVP
#use delay(clock=4000000)
main()
{
while (TRUE) // PIC18F26K22 PIC18F46k22
{ // LATB PORTB LATB PORTB
output_low(PIN_B3); // 0b0000 0b0111 0b1000 0b1000
output_high(PIN_B0); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B0); // 0b0000 0b0111 0b0100 0b0100
output_high(PIN_B1); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B1); // 0b0000 0b0111 0b0010 0b0010
output_high(PIN_B2); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B2); // 0b0000 0b0111 0b0001 0b0001
output_high(PIN_B3); // 0b0111 0b0000 0b0000 0b0000
}
}
|
Disassembly listing looks OK to me. | Code: | 1: // Port B Test
0000 EF02 GOTO 0x4
2:
3: #include <18F26K22.h>
4: //#include <18F46K22.h>
5: #fuses ECH,NOWDT,NOPROTECT,NOLVP,NOPLLEN
6: //#include <18F252.h>
7: //#fuses HS,NOWDT,NOPROTECT,NOLVP
8: #use delay(clock=4000000)
9:
10: main()
0004 6AF8 CLRF 0xff8, ACCESS
0006 9ED0 BCF 0xfd0, 0x7, ACCESS
0008 010F MOVLB 0xf
000A 6B38 CLRF 0x38, BANKED
000C 6B39 CLRF 0x39, BANKED
000E 6B3A CLRF 0x3a, BANKED
0010 6A77 CLRF 0xf77, ACCESS
0012 6A78 CLRF 0xf78, ACCESS
0014 6A79 CLRF 0xf79, ACCESS
11: {
12: while (TRUE)
13: {
14: output_low(PIN_B3);
0016 9693 BCF 0xf93, 0x3, ACCESS
0018 968A BCF 0xf8a, 0x3, ACCESS
15: output_high(PIN_B0);
001A 9093 BCF 0xf93, 0, ACCESS
001C 808A BSF 0xf8a, 0, ACCESS
16: output_low(PIN_B0);
001E 9093 BCF 0xf93, 0, ACCESS
0020 908A BCF 0xf8a, 0, ACCESS
17: output_high(PIN_B1);
0022 9293 BCF 0xf93, 0x1, ACCESS
0024 828A BSF 0xf8a, 0x1, ACCESS
18: output_low(PIN_B1);
0026 9293 BCF 0xf93, 0x1, ACCESS
0028 928A BCF 0xf8a, 0x1, ACCESS
19: output_high(PIN_B2);
002A 9493 BCF 0xf93, 0x2, ACCESS
002C 848A BSF 0xf8a, 0x2, ACCESS
20: output_low(PIN_B2);
002E 9493 BCF 0xf93, 0x2, ACCESS
0030 948A BCF 0xf8a, 0x2, ACCESS
21: output_high(PIN_B3);
0032 9693 BCF 0xf93, 0x3, ACCESS
0034 868A BSF 0xf8a, 0x3, ACCESS
0036 D7EF BRA 0x16
22: }
23: }
24: /*
0038 0003 SLEEP
|
Does anyone else get the same as I do?
Last edited by Mike Walne on Tue Feb 28, 2017 3:47 am; edited 2 times in total |
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Mon Feb 27, 2017 1:24 pm |
|
|
| Quote: | | When I simulated in MPLAB for a PIC18F26K22 I got rubbish with CCS code. |
What exactly is rubbish ? and where are you looking in MPLAB 8.92
when you see it. Give precise instructions on how to duplicate the
problem. |
|
|
Ttelmah
Joined: 11 Mar 2010
Posts: 19496
|
|
| Posted: Mon Feb 27, 2017 1:52 pm |
|
|
The problem you have is the ECCP.
On the smaller PIC ECCP1 is multiplexed onto the low PortB pins.
The ECCP is designed so that it wakes up driving it's pins, so that a power driver using it, does not have it's inputs floating.
You need to turn off the ECCP so that the pins can be used for normal I/O.
Currently B0, is behaving as the ECCP fault control, so overrides the other pins.... |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Mon Feb 27, 2017 2:00 pm |
|
|
Thank you Ttelmah. I understand now. I never simulate i/o ports.
I only simulate code. That's why I didn't understand his question. |
|
|
Ttelmah
Joined: 11 Mar 2010
Posts: 19496
|
|
| Posted: Mon Feb 27, 2017 2:43 pm |
|
|
The great thing is how good the 'team' here is. You'll spot oddities it'd take me ages to see, while I then see other things. Add a few others (Jay etc.), and there is little we don't know about the PIC.  |
|
|
Mike Walne
Joined: 19 Feb 2004
Posts: 1785
Location: Boston Spa UK
|
|
| Posted: Mon Feb 27, 2017 6:04 pm |
|
|
| PCM programmer wrote: | | Quote: | | When I simulated in MPLAB for a PIC18F26K22 I got rubbish with CCS code. |
What exactly is rubbish ? and where are you looking in MPLAB 8.92
when you see it. Give precise instructions on how to duplicate the
problem. |
I'm looking at LATB and PORTB in a watch window as I single step through either the CCS source code or the disassembly listing.
The values I've added to the source code were copied from the watch window.
What's really throwing me is the fact that the hardware behaves perfectly with all the chips mentioned but the 18K26K22 does not tally with MPLAB SIM.
Thanks for your response.
Mike |
|
|
Mike Walne
Joined: 19 Feb 2004
Posts: 1785
Location: Boston Spa UK
|
|
| Posted: Mon Feb 27, 2017 6:18 pm |
|
|
| Ttelmah wrote: | The problem you have is the ECCP.
On the smaller PIC ECCP1 is multiplexed onto the low PortB pins.
The ECCP is designed so that it wakes up driving it's pins, so that a power driver using it, does not have it's inputs floating.
You need to turn off the ECCP so that the pins can be used for normal I/O.
Currently B0, is behaving as the ECCP fault control, so overrides the other pins.... |
Thanks Ttelmah. I sort of follow what you're telling me. My code is now | Code: | // Port B Test
#include <18F26K22.h>
#fuses ECH,NOWDT,NOPROTECT,NOLVP,NOPLLEN
#use delay(clock=4000000)
main()
{
setup_ccp1(CCP_OFF);
while (TRUE) // PIC18F26K22 PIC18F46k22
{ // LATB PORTB LATB PORTB
output_low(PIN_B3); // 0b0000 0b0111 0b1000 0b1000
output_high(PIN_B0); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B0); // 0b0000 0b0111 0b0100 0b0100
output_high(PIN_B1); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B1); // 0b0000 0b0111 0b0010 0b0010
output_high(PIN_B2); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B2); // 0b0000 0b0111 0b0001 0b0001
output_high(PIN_B3); // 0b0111 0b0000 0b0000 0b0000
}
} |
I still get the same output as before from MPLAB SIM.
Is there something else I have to do?
What am I missing when the hardware behaves as I expect?
Mike |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Mon Feb 27, 2017 8:36 pm |
|
|
There is a bug with MPSIM and the 18F26K22. Read the yellow post by
dan1138 on the Microchip forum:
http://www.microchip.com/forums/m603514.aspx
He says you have to set ADCON1 to 0x0F to fix it. I did that. It works.
Try it.
| Code: | // Port B Test
#include <18F26K22.h>
#fuses ECH,NOWDT,NOPROTECT,NOLVP,NOPLLEN
#use delay(clock=4000000)
#byte ADCON1 = getenv("SFR:ADCON1")
void main()
{
ADCON1 = 0x0F;
while (TRUE) // PIC18F26K22 PIC18F46k22
{ // LATB PORTB LATB PORTB
output_low(PIN_B3); // 0b0000 0b0111 0b1000 0b1000
output_high(PIN_B0); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B0); // 0b0000 0b0111 0b0100 0b0100
output_high(PIN_B1); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B1); // 0b0000 0b0111 0b0010 0b0010
output_high(PIN_B2); // 0b0111 0b0000 0b0000 0b0000
output_low(PIN_B2); // 0b0000 0b0111 0b0001 0b0001
output_high(PIN_B3); // 0b0111 0b0000 0b0000 0b0000
}
} |
|
|
|
Mike Walne
Joined: 19 Feb 2004
Posts: 1785
Location: Boston Spa UK
|
|
| Posted: Tue Feb 28, 2017 3:46 am |
|
|
Yes, it works OK for me now.
I've just tried stepping through my original code in the MPLAB debugger. It operates correctly.
I suppose that's because MPLAB is reading the registers in the chip rather than simulating.
I'll add solved to the title, and hope it helps someone else.
I'm relieved to know that I wasn't losing my marbles.
I would probably have never found the fix on my own.
Thank you PCM.
Mike
I'm new to the *K22 chips, and was alerted to the problem when I adapted the CCS stepper motor code.
I was uncertain, for some time, whether the problem was CCS or MPLAB.
It seemed odd that Microchip could get something so basic so badly wrong.
Really pleased to have it solved, as I was wary of continuing to trust MPLAB. |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
