[etherlab-users] Writing to a blank EEPROM

[Aaron Edsinger]

Thanks Florian. My concern is being able to write the first 40 words 
into a blank EEPROM using the Master. If it is unable to recognize a 
slave with a blank EEPROM, then how is it able to write these initial 
words?

--Aaron

Florian Pose wrote:
> Hi Aaron,
> 
> On Thu, Nov 15, 2007 at 06:00:18PM -0800, Aaron Edsinger wrote:
>> Is it possible for the Ethercat Master to write to a blank slave EEPROM. 
> 
> Yes. I never wrote data to a blank EEPROM, but it should be possible.
> As far as I know, there are no restrictions to write EEPROM data.
> 
>> Is default EEPROM data required to bootstrap the slave and allow the 
>> Master to recognize it?
> 
> Yes. At least the first 40 words should be filled with the specified
> information. Especially the ID object (vendor Id, product code, etc.) is
> necessary to recognize the slave. If the slave has PDOs (almost
> certainly), the EEPROM must contain at least the sync manager
> configurations. If the PDOs are fixed, and the slave does nut support
> CoE, then the PDO mapping information must be in the EEPROM either. If
> they are variable, the master can read the mapping out of the CoE
> dictionary.
> 
>> I ask because I am developing a custom ET1200 slave board and am not 
>> sure if I need to pre-burn the EEPROM before soldering it to the board. 
>> Any general advice for this application would be very much appreciated.
> 
> No, the master should be able to write it to the device.
> 

-- 
Aaron Edsinger, Ph.D.
San Francisco, CA
617.869.9691
www.mekabot.com
www.hee3.com
www.csail.mit.edu/~edsinger