SD memory card what is the dirty little secret.

[Guest]

[Guest]

[Guest]

[ckielstra]

Hans,

After a good night of sleep I suspect your problem with two FAT's must have something to do with your new formatting routine. Is your MBR correct? Are you sure you are writing the MBR to the first sector in the Flash card? How many partitions are you configuring?

Like I wrote before, the only difference between the V1.7 en V1.9 manuals has to do with updated VDD values according to the Change Notes (and the manual is 200kb smaller...). I couldn't find any reference to differences in the CMD0-CMD1 or CMD0-ACMD41 sequences, but as I don't have all documentation there is a possibility for differences between the SPI and SD-card mode.

Some more documentation for your growing collection:
MMC-System-Summary-v3.31.pdf
In Chapter 8 this document gives a description of two file formats for the MMC card, these are the same as described in the FILE_FORMAT field of the SD-card CSD-register. I mentioned this field in an earlier post and it seems that the value '0, Hard disk-like file system with partition table' is the only practical value used in DOS-compatible file systems. Very likely this value is correct in your system and can be dismissed as a possible error cause.

[Guest]

If you have Data sheet V1.7 Page 4-6 Figure 4-7 you will see a diagram of how the MMC is initialized . Eveyrthing satrts with CMDO Then CMD1 for MMC SPI mode.

Also Page 4-7 4.3.2
The MultiMediaCard will not respond to ACMD41 (actually it will not respond to APP_CMD—CMD55, that
precedes it). The MultiMediaCard will be initialized as per the MultiMediaCard spec, using SEND_OP_COND
command (CMD1 of MultiMediaCard). The host should ignore an ILLEGAL_COMMAND status in the
MultiMediaCard response to CMD3, since it is a residue of ACMD41 which is invalid in the MultiMediaCard
(CMD0, 1, 2 do not clear the status register). Actually, ACMD41 and CMD1 will be used by the host to distinguish between MultiMediaCard and SD Cards in a system.

As you see it is in the data sheet, yet you claim it is not...

Boot sector is of course at address ZERO like I would writ e it to someother location...! These cards do NOT have DOS partitions...

No. the two boot sectors are NOT because of my format routine.!

My format routine was writen to PROVE the existance of the "other boot sector" !
In the example I described, I had an original 16 MEG card factory formated by Toshiba as FAT12 readable on the USB card reader as FAT12.
I then put it in my hardware where upon reading Sector 0 I got all 00's . This led me to try and write the boot sector to what the SPI sees as adress zero.

That new FAT16 boot sector WORKS, but cannot be seen by the USB card reader which is using SD protocol, the USB card reader still sees FAT12...

I was then able to boot and read/write files using the bootsector/FAT16 that I put on the card.
However on the PC USB card reader it is still showing as FAT12 ! and my files are not showing up.

Futher in the doc. you will find a reference to the secure DOS area. It's an area I cannot "get at" because it requires the secure access protocol, which SD Org. are charging $11K for...... Arrrrrrrrgh ! bastards, I hope they fry in hell..


So if you READ the data sheet that you say you have you can find all of this...

There are TWO bootsectors and two FAT areas in the secure DOS image area. See page 1-8 Figure 1-2 where you will see a diagram showing TWO sector 1 each with Block 0,1,2,n
And on that same page you can read this
"Protected Area—The part of the Card that relates to the secured copyright management and has separate DOS
partitioning including sectors and blocks. The card write protection mechanism does not effect this area."

See the words
SEPARATE DOS PARTITIONING inclduing sectors and blocks...

it does not say in that many words, but the default protocol is the SD protocol and I suspect that is what the USB card reader is using and seeing the SEPARATE DOS partitioning...

In the MMC mode which I am using, it sees the un-secured area which ALSO has a boot sector...
Just look at the after thought of the connectors…. The MMC card had equal spacing and lots of it, they never thought further than that, and when SD came along the squeezed in the 1/3 wide connector. It is a typical “designed by committee product” both in the hardware and in the firmware, and it shows…


Thanks but unless you are offering some new information I refuse to lead you through the data sheet any more...

[ckielstra]

No new information.

Sorry for having upset you. I'm only trying to help, no personal gain for me other than the attraction of solving a good puzzle.

Good luck.
bye.

[Guest]

[ckielstra]

[Guest]

[kevin]

[ckielstra]

Hi Hans,

You really like pushing it, don't you?

I've been angry with you all day because I'm only trying to help you and then you give me this kind of [censored], but after your last post I could see the humor of it and had to laugh about the two us. I don't know if I offended you in some way by my style of writing, English is not my native language so I sometimes do miss the small nuances.

[Guest]

ckielstra,
Please check your PM box.

[Hi]

Hi,

I know that CMD1 is 0x41. What is ACMD41 ?
I have been all around the SD card datasheet and unable to figure this out.

Also, If I understand the documents right, a SD card will not initialize to SPI mode with code written for a MMC card. Is this correct ?

Given that I have enough IO pins, how hard is it to use an SD card in 4Bit SD card mode ? I expect that the ansver will explain why there's no sample code for that mode around the web.

Any clues ?

Thanks.

[ckielstra]

Hi Guest, who are you?

[direnc]

Hi ckielstra,

Thanks for your aswers.