| View previous topic :: View next topic |
| Author |
Message |
cile
Joined: 10 May 2009
Posts: 17
|
| keypad 4x4 problem |
 Posted: Wed Jul 22, 2009 6:52 am |
 |
|
Hello !
I'm using keypad 4x4 which is connected to PIC 18F4550 and LCD 4x20.
In my program I use drivers for keypad and LCD from PCM programmer. There is link for 4x4 keypad driver :
http://www.ccsinfo.com/forum/viewtopic.php?t=28022&start=14
Everything is OK ( compiling), but when I press any button on keypad this number is shown on display more times.. for example: for ONE press on ONE button of keypad (for ex: number 5), there is on display 5 or 4 or 3 or 2 or 1 same pressed number. Same problem is for all buttons on keypad.
example
Keypad (ONE press no. 2) ----> on LCD is shown: 2
Keypad (if I press again 2)--> ON LCD is shown sometimes 22 or 222 or 222222.. etc
There is link of keypad which I'm using:
http://www.mikroe.com/en/tools/keypad/
Please help
Thank you
P.S. sorry my english is not so good.... |
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Wed Jul 22, 2009 12:00 pm |
|
|
Did you put the keypad and the LCD on the same pins ? If so, you can't
do that. The flex lcd driver and the keypad driver won't work if they
share the same pins. Put them on different pins. |
|
|
cile
Joined: 10 May 2009
Posts: 17
|
|
| Posted: Wed Jul 22, 2009 12:44 pm |
|
|
No, the pins of LCD and the pins of keypad are not the same.
LCD is on PORTB & PORTE (B4-B7 & E0-E2 ) and the keypad is on the PORTD.
Can this problem fix in software ?? to use some delay_ms() or something else ?
Thank you ! |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Wed Jul 22, 2009 1:25 pm |
|
|
1. Post your test program that calls the functions in the keypad driver.
The program must be small and compile with no errors.
2. Also, the connector for the keypad should have Pin 1 marked with a
little triangle symbol. Look for that triangle. Make sure that you
understand the pin numbers on the connector. Post a list of
connections between the keypad connector and the PIC.
3. Do you have pull-up resistors on the Row pins ? |
|
|
cile
Joined: 10 May 2009
Posts: 17
|
|
| Posted: Thu Jul 23, 2009 4:35 am |
|
|
1.
| Code: | #include <18F4550.h>
#fuses HSPLL, PLL5, CPUDIV1, NOWDT, PUT, BROWNOUT, NOLVP
#use delay (clock=48000000)
#include <keypad1.c> // same as
// http://www.ccsinfo.com/forum/viewtopic.php?t=28022&start=14
// just pins changed
#include <Flex_LCD420.c>
void main() {
char k;
lcd_init();
kbd_init();
lcd_putc("\fReady...\n");
while (TRUE) {
k=kbd_getc();
if(k!=0)
if(k=='*')
lcd_putc('\f');
else
lcd_putc(k);
}
}
|
2.There are pull up resistors (10kOhm) on rows.
//Keypad connection:
#define row0 PIN_D4
#define row1 PIN_D5
#define row2 PIN_D6
#define row3 PIN_D7
#define col0 PIN_D0
#define col1 PIN_D1
#define col2 PIN_D2
#define col3 PIN_D3 |
|
|
arunb
Joined: 08 Sep 2003
Posts: 492
Location: India
|
| RE: |
| Posted: Thu Jul 23, 2009 7:08 am |
|
|
You need to do several things
1. Debounce the keys
2. Every time the user presses the key, it must be released for another operation. or in other words disable multiple presses.
hope this helps,
thanks
arunb |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu Jul 23, 2009 10:29 am |
|
|
arunb,
Look at the keypad driver code given in the link in his post.
The code already has a debounce feature in it.
| Code: |
#define KBD_DEBOUNCE_FACTOR 33 // Set this number to apx n/333 where
// n is the number of times you expect
// to call kbd_getc each second
if(++kbd_call_count>KBD_DEBOUNCE_FACTOR)
|
cile,
| Quote: | | #use delay (clock=48000000) |
This is a very high clock speed. The original CCS keypad driver was
tested at 20 MHz maximum. However, the driver has a parameter
that can be adjusted for the oscillator frequency.
The ratio of 48 MHz to 20 MHz is:
48 / 20 = 2.4
The old debounce value (for 20 MHz operation) is 33:
| Code: | | #define KBD_DEBOUNCE_FACTOR 33 |
So, multiply the old "debounce factor" by 2.4 to get the new value:
| Code: | | #define KBD_DEBOUNCE_FACTOR 79 |
Edit the keypad driver file, re-compile, and see if it now works. |
|
|
cile
Joined: 10 May 2009
Posts: 17
|
|
| Posted: Thu Jul 23, 2009 1:17 pm |
|
|
Ok, thank you, I will try tomorrow , I can not do it now.
I will post what happened |
|
|
cile
Joined: 10 May 2009
Posts: 17
|
|
| Posted: Fri Jul 24, 2009 7:13 am |
|
|
I have tested but still same problem !
any advice ??
thank you ! |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Fri Jul 24, 2009 11:05 am |
|
|
| Quote: | 2.There are pull up resistors (10kOhm) on rows.
//Keypad connection:
#define row0 PIN_D4
#define row1 PIN_D5
#define row2 PIN_D6
#define row3 PIN_D7
#define col0 PIN_D0
#define col1 PIN_D1
#define col2 PIN_D2
#define col3 PIN_D3 |
The keypad schematic does not label its pins as "rowx" and "colx".
It calls them RB0 to RB7.
http://www.mikroe.com/pdf/keypad_board_schematic.pdf
Post a list of connections, similar to the one above, but use the pin names
from the keypad schematic instead of "rowx" and "colx". |
|
|
arunb
Joined: 08 Sep 2003
Posts: 492
Location: India
|
| RE: |
| Posted: Sat Jul 25, 2009 4:15 am |
|
|
| Quote: |
Look at the keypad driver code given in the link in his post.
The code already has a debounce feature in it. |
I did not notice the debounce feature..thanks
It seems the key presses are repeating, if so the keys have to be disabled after each press, when the key is released the key can be enabled again.
thanks
arunb |
|
|
cile
Joined: 10 May 2009
Posts: 17
|
|
| Posted: Sun Jul 26, 2009 7:10 am |
|
|
Hello,
There is link, where is schematic of my final project. You will see on schematic how is made jack for kepyad.
http://rapidshare.com/files/260219348/picture.JPG
I think, maybe the buttons on keypad are bad.?!?!
Thank you ! |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Sun Jul 26, 2009 2:18 pm |
|
|
I was able to duplicate your problem. I fixed it by increasing the
debounce value to this:
| Code: | | #define KBD_DEBOUNCE_FACTOR 200 |
I didn't experiment to try to fine-tune it to the minimum value that will
work. I just dropped in 200 and it started working.
I've never liked the "debounce factor" method that CCS uses in kbd.c.
It's not clean enough. You have to estimate (or physically determine)
the amount of time it takes to execute the keypad routine, and then
manually calculate the debounce factor. It would be better to require
a fixed delay such as delay_ms(1), in the loop that calls the keypad
routines and then use a fixed value for the debounce factor (1 or 3).
Then you wouldn't have to do this calculation every time you change
the oscillator frequency or PIC series. |
|
|
cile
Joined: 10 May 2009
Posts: 17
|
|
| Posted: Fri Aug 07, 2009 5:37 am |
|
|
Hello. I could not answer before. I have put 150 debounce factor, and it seems better. I did not test with 200, but I will try and report.
Thank you for help ! |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
