DP83822IF: capacity of equalization

Hi Aniruddha san

Thank you for your reply!

Could you provide more details about the application and why changing equalization is needed?
->Our customer uses twisted pairs 30mm cable instead of CAT5E cable for 1000Base-T. So, we would like to break down the issue into parts whether this problem causes of cable or not.
   So, according to increasing equalization level, if the problem solve, we understand that it depends on cable.
   Of course, we recognize that our customer should use CAT5E cable.

Kind regards,

Hirotaka Matsumoto

Aniruddha san

Thank you for your reply always!
Can you also elaborate on the problem seen by customer? Is it link instability?
->Yes, the link is instability. As the condition, it occurs link up and link down.
   Our customer's configuration is follows;
   DP83867 <-> Pulse trans <-> Connector <-> Cable(30mm) <-> Connector <-> Pulse trans <-> Micrel's Giga phy
   About cable, please refer to the attachment file.
20170823_DP83867's_cable.pdf

And then, pulse trans is satisfied with requirements which you mentioned insersion loss, return loss, turn ration, Cross talk, Differential to Common Rejection ratio and Isolation. 

We assume that it is not relation to RGMII I/F like as shifted mode and aligned mode.
We guess that it depends on follows;
-Pulse trans <-> Connector <-> Cable(30mm) <-> Connector <-> Pulse trans
-Auto negotiation setting between DP83867 and Micrel's phy

And then, we suggest that our customer uses CAT5E cable and RJ45 connector.

If you have some additional comments, could you give us?

Kind regards,

Hirotaka Matsumoto

Pulse trans

Aniruddha san

We really apologize that we incorrectly wrote device number.
The correct device No is "DP83867IR" Giga Phy.

TI's phy : DP83867IR(Giga Phy)
Micrel phy : KSZ9031(Giga Phy)

Kind regards,

Hirotaka Matsumoto