DS125BR111: EEPROM and Max trace length

Antony,

The device will try to read the EEPROM.  The outcome is hard to determine, it depends on the actual data in the EEPROM.  An EEPROM with all 0's may load the bad data into the DS125BR111.  An EEPROM with all 1's would not complete the loading since the addressing and CRC would be incorrect.

The maximum trace length is not a constant.  The supported trace length will depend on the system Tx and Rx capabilities and on the level of noise on the channel.  The DS125BR111 is a linear equalizer designed to extend the maximum supported link attenuation by as much as 10 dB.  Typically, the DS125BR111 can improve the channel reach by ~ 5-8 dB.

The placement of the DS125BR111 within the channel is not super critical.  The rule of thumb is that there should be a least as much attenuation between the DS125BR111 input and the system Tx as the equalization setting on the DS125BR111 can add to the incoming signal.

Regards,

Lee 

Hi Lee,

Thanks for your comments.

Customer told me when EEPROM is not programmed, all the data content are 0xFF.  With such EEPROM on their board, it can still pass funcion test.  May you explain why?

Besides, with such all 0xFF EEPROM content on board, is there any way our chip can detect such case?  (e.g. any update on DONE pin?

Antony

Antony,

The DONE pin should not go low.  The DS125BR111 default register values are being used by the analog datapath and the system is working.

Regards,

Lee