Board-ID EEPROM on BeagleBone Black vs. Beagle Bone and SRM

GoogleGroups
1

Hi everyone,

while looking at the BBB Documetation i got quite confiused. The BBB-SRM (A5.6) says that teh eeporm on the BBB ist the same as on the original BeagleBone:

5.3.2 32KB EEPROM
A single 32KB EEPROM is provided on I2C0 that holds the board information. This information includes board name, serial number, and revision information. This is the same as found on the original BeagleBone. It has a test point to allow the device to be programmed and otherwise to provide write protection when not grounded.

But form what I see on the board and in the schematic, it seems that there is a 4kByte-Prom (24LC32) on the BBB while the one on the original BeagleBone is a 32kByte (CAT24C256) one.

Finally confusion was perfect when I read in the BOM fro the original BeagleBone:

CAT24C256W | IC EEPROM 32KBIT 400KHZ SOT23-6

| ON | CAT24C256WI-G |

  • | - | - | - |

while CAT24C256W is a 256KBIT EEPROM.

Is this a “bug” in the BBB-SRM and the Beagle-Bone Bom? I searched quite a while now bit did not find any note about EEPORM changes on the BBB.

Also it seems that the EEPROM Map from the original BeagleBone does not completely fit the on the one used on the BBB-EEPROM (in the first 4kBytes of course). Looking at the teble in teh Original BeagleBone SRM (RevA3_1.0, page46) there seems to be no “Configuration Option” and reserved fields that are zeroed on the BBB-Eprom. instead, everything after offset 0x1c is still erased to 0xFF.

2

I will look into this when I get back from vacation next week.

Gerald

3

OK. I went ahead and checked. They are different parts. I will fix the “bug” in the SRM at a later release and remove the statement that they are the same. They are not the same. They are different.

Gerald

4

I just noticed this too. It’s also repeated in section 8.2 (page 94).

I looked for a newer version that Rev A5.6, but couldn’t find one.

5

I will add it to my list of things to look at at some point one of these days. I may have already fixed it in the upcoming version, but I will double check.

Gerald