Jumper CE_NAND_E

[David Goadby]

I have two A10S boards, one with 4GB NAND and one without. On both boards the CE_NAND_E jumper is made. The manual says that if I open (cut) CE_NAND_E the NAND memory is disabled. The circuit diagram shows the CE_NAND_E as a chip enable so I guess that is what happens in practice.

I am debugging a booting issue with an A10S-4GB board. If I cut CE_NAND_CE I can see that the memory will not be enabled but what happens to the boot sequence? The A10S-4GB tries to boot the NAND first if present and then proceeds to check the SD card. I specifically do not want the A10S to try and boot the NAND at this moment.

Do I need to change anything else other than cutting the link to do this?

David Goadby

[martinayotte]

Hi David,

The boot order is microSD first, if it fails, then it will boot from NAND.
So, I don't think you need to cut any wire.

[David Goadby]

Hi Martin,

Thanks for that. I tried it and it does work that way. Come to think of it it's obvious that the boot should work that way. It's been a tough week and it's only Tuesday 

Presumably, if I want access to the extra I/O lines that the NAND steals, cutting the CE_NAND_E is desirable rather than necessary.

David

[Robert Leacroft]

Hi all,

I also have two A10S boards, one with nand and one without.  I wanted to use all the pins on GPIO 2 connector for various inputs and outputs and have made a board that plugs on to the 40 pin with ULN2803 buffers for outputs and optos with holdups (to 3V3) for the inputs. I am running Debian from  SD card.

The board with nand on it boots perfectly from SD card and a python progam controls everything OK.
The board without nand will not boot at all from SD card while my in/out board is plugged into the 40 pin. If I remove the plug and then boot it and then re-plug in my board after it is running  it all works OK.  At boot up some of these pins are obviously examined by the A10S and perhaps my holdups or something give it the wrong information and thus it never boots. Perhaps the additional load of the nand chip on the other board changes the pin states so they do get read correctly and then allows it to boot from SD.

As this problem was on the board without nand there seemed little point in cutting the nand jumper.

I have found a solution and that involved changing nand_used to = 0  in the script.bin   nand_para section file thus disabling the nand (even though it is not there). Everything now works fine and my board boots from SD and all IO pins are controlled OK by my python program.

For future information does anyone know which of the GPIO/nand pins are being looked at by the A10S at boot?

Robert