| View previous topic :: View next topic |
| Author |
Message |
incubus22
Joined: 15 Nov 2007
Posts: 14
|
|
 Posted: Sun Nov 25, 2007 10:30 am |
 |
|
| Why isn't anybody answering??? Hello Miniman... Are you still there? |
|
 |
andreluizeng
Joined: 04 Apr 2006
Posts: 117
Location: Brasil
|
|
| Posted: Mon Nov 26, 2007 5:05 am |
|
|
Hi, im trying to simulate this code on proteus VSM, but im having some problems.... the MMCInit function does not work (but i already fixed it)... and a INITFat function does not work too.
my question is:
where all those structs and variables are initizalized ?
in this piece of code
| Code: | memcpy(&DiskInfo,gFiles[MAXFILES-1].IOpuffer,sizeof(DiskInfo));
|
DiskInfo struct is initialized with the data in IOPuffer, but IOpuffer are not initialized either.
| Code: |
if (gFiles[MAXFILES-1].IOpuffer[0] != 0xEB)
{
gStartSectorL = gFiles[MAXFILES-1].IOpuffer[0x1C6];
gStartSectorH = gFiles[MAXFILES-1].IOpuffer[0x1C7];
gStartSectorHL = gFiles[MAXFILES-1].IOpuffer[0x1C8];
ReadSector (gFAT32Vars.gStartSector, gFiles[MAXFILES-1].IOpuffer);
}
|
the data in IOpuffer never will be 0xEB, insted it is 0xFF
when this calculus
| Code: |
actsector = gFAT32Vars.gStartSector + DiskInfo.Reserved1;
|
or
| Code: |
gFAT32Vars.gFirstDataSector = gFAT32Vars.gStartSector + DiskInfo.FATCopies * DiskInfo.hSectorsPerFat + DiskInfo.Reserved1 - 2;
|
is done:
we got a overflow in gFat32Vars.gFirstDataSector...
All those variables in that math have a 0xffff... value.
somebody have any sugestion ????
thanks in advance.
regards.
_________________
Andre |
|
|
incubus22
Joined: 15 Nov 2007
Posts: 14
|
|
| Posted: Mon Nov 26, 2007 5:54 am |
|
|
Hello andreluizeng!
Mine is working properly. The only problem is the writing speed.  |
|
|
Milhaus
Joined: 20 Oct 2007
Posts: 9
|
|
| Posted: Mon Nov 26, 2007 6:15 am |
|
|
incubus
if your HW SPI does not work, you have probably connected SPI device to different pins than use your PIC for HW SPI.
Milos |
|
|
Milhaus
Joined: 20 Oct 2007
Posts: 9
|
|
| Posted: Mon Nov 26, 2007 6:30 am |
|
|
andreluizeng: the first sector MBR( start with 0xEB ) is read in InitFAT function look below. If not then SPI communication does not work or you have any problem with MMCInit.Do you have card formatted to FAT32 ?
void InitFAT()
{
int32 actsector;
char i;
gFirstEmptyCluster = 0;
gFAT32Vars.gStartSector = 0;
ReadSector(gFAT32Vars.gStartSector,gFiles[MAXFILES-1].IOpuffer);
if (gFiles[MAXFILES-1].IOpuffer[0] != 0xEB) |
|
|
andreluizeng
Joined: 04 Apr 2006
Posts: 117
Location: Brasil
|
|
| Posted: Mon Nov 26, 2007 7:02 am |
|
|
Milhaus,
my spi is totally functional, as i said, im using proteus VSM (simulation), probably my mmc initialization isnt working very good.
im try again.
thanks.
_________________
Andre |
|
|
incubus22
Joined: 15 Nov 2007
Posts: 14
|
|
| Posted: Mon Nov 26, 2007 8:01 am |
|
|
| Milhaus wrote: | incubus
if your HW SPI does not work, you have probably connected SPI device to different pins than use your PIC for HW SPI.
Milos |
Hello Milhaus! Thank you for replying! 
I connected the MMC to the right SPI hardware pins on the PIC. Please try the hardware SPI configuration and your code (with hardware SPI settings on) on your own and let's compare notes. Please...  |
|
|
andreluizeng
Joined: 04 Apr 2006
Posts: 117
Location: Brasil
|
|
| Posted: Mon Nov 26, 2007 8:25 am |
|
|
Incubus,
try this code.. it probably will work.
| Code: |
#define SPI_ENABLE 0x20
#define SPI_DISABLE 0x00
#define CLK_L_TO_H 0x10
#define CLK_H_TO_L 0x00
#define CLK_DIV_4 0x00
#define CLK_DIV_16 0x01
#define CLK_DIV_64 0x02
void SPIConfig (void)
{
SSPCON1 = 0x00;
SSPCON1 = SPI_ENABLE | CLK_L_TO_H | CLK_DIV_4;
}
|
change configuration whatever way you want.
regards.
_________________
Andre |
|
|
Miniman
Joined: 30 Apr 2007
Posts: 44
|
|
| Posted: Mon Nov 26, 2007 2:39 pm |
|
|
Hi!
sorry for being off a while. If your HW SPI does not work maybe the SDI pin don't recognize 3.3V as a logic '1'. Check the datasheet to see what type of pin it is when using it with SPI. You may need to convert 3.3v to 5v to get it to work properly. I believe i saw a thread in the discussion part of this forum about 3.3v to 5v and 5v to 3.3v conversion not long ago. This may be the problem.. otherwise i do not know what i could be..
Also make sure that these registers have the right address:
| Code: |
#byte SSPBUF = 0xFC9
#byte SSPSTAT = 0xFC7
#byte SSPCON1 = 0xFC6
#bit BF = SSPSTAT.0
|
I believe this would also work (and better for using in different pic:s):
| Code: |
#byte SSPBUF = Getenv("SFR:SSPBUF")
#byte SSPSTAT = Getenv("SFR:SSPSTAT")
#byte SSPCON1 = Getenv("SFR:SSPCON1")
#bit BF = SSPSTAT.0
|
hope it helps..
and to you andreluizeng, I do not know anything about proteus, but i know the code works in real life..
best regards
miniman |
|
|
andreluizeng
Joined: 04 Apr 2006
Posts: 117
Location: Brasil
|
|
| Posted: Mon Nov 26, 2007 4:21 pm |
|
|
Miniman, whats up ?
I fixed de code... now im using fat32 on pic18f2525.
im having now little problems like: no removing an existing file, or.. when i try to open a file with 'r' argument.. it creates another one... but i think that is not 'too' hard to fix.
thank you for the reply..
regards.
_________________
Andre |
|
|
incubus22
Joined: 15 Nov 2007
Posts: 14
|
|
| Posted: Tue Nov 27, 2007 12:22 am |
|
|
Hello andreluizeng! 
I managed it to work by implementing what you said. Thank you!
I have another question. What delay should I use with this line of code in CCS?
#use delay(clock=20000000)
I am using a 20MHz crystal.
I also used HS for the oscillator option. Are my settings correct to get the maximum speed of writing data on the MMC? I really need to write data in a very fast manner.
Thanks in advance! |
|
|
vikram.kadian
Joined: 28 Nov 2007
Posts: 1
Location: CHENNAI,INDIA
|
| mmc -spi |
| Posted: Wed Nov 28, 2007 12:47 am |
|
|
| can anyone tell me hoe to write mmc in fat format . i have seen few programs with including files such as MMC_SPI_FAT32.C or .h . anybody have these files with them then please do send me to my mail id [email protected] . thanks in advance bro. |
|
|
BLL
Joined: 11 Nov 2006
Posts: 181
Location: Birmingham, UK
|
| FAT32 on SD/MMC |
| Posted: Mon Feb 04, 2008 6:01 am |
|
|
Hi,
I have been trying to get the modified version of Tomi's code to work on a 18F2620, using hardware SPI. The code works providing I use DIV16 on the SPI, not DIV4.
The problem I find is this:
I write a small (about 300 bytes) file to the card - I can read this fine on the PC.
I then write a second file.
When I look at the card on the PC, either:
The second file id there and OK and the first file is corrupt and has lost its filename
or:
There is only one corrupted file with no name on the card.
The only way to get rid of the corrupted file is to reformat the card!!
The corrupted file says 256 bytes when mouse hovering over it, but 0 bytes in properties!
Can anyone help please?
Thanks
Brian |
|
|
andreluizeng
Joined: 04 Apr 2006
Posts: 117
Location: Brasil
|
|
| Posted: Tue Feb 05, 2008 3:33 pm |
|
|
Incubus22,
Im using 20MHz clock and clock_div_64 (312.5 khz clock) until those commands were sents:
1 - CMD0 (Reset)
2 - CMD1 (SPI)
3 - CMD16 (BLOCK SIZE)
4 - CMD10 (CID)
5 - CMD9 (CSD)
6 - CMD6 (HI-SPEED MODE)
then you can change spi clock to clock_div_4 (at 20MHZ fullspeed is 5mhz (theorical)) and write or read at a higher speed. (im not using any delay between comands)
im using it and still working.
hope it helps.
regards.
_________________
Andre |
|
|
BLL
Joined: 11 Nov 2006
Posts: 181
Location: Birmingham, UK
|
|
| Posted: Tue Feb 05, 2008 3:43 pm |
|
|
Hi,
Thanks for that - I'll go and play with it.
Brian |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
