UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Explorer
Explorer
5,667 Views
Registered: ‎05-22-2008

HSTL RGMII interface need for external Vtt

Jump to solution

I'm interfacing a Marvell 88e1116R RGMII phy chip to one of the HR Banks on a Zynq XC7Z030. The Marvell PHY uses HSTL at 1.8V. Looking at the 7 Series SelectIO guide (UG471), on pages 63 and 66, it appears that external resistors pulled up to Vtt = .9 V may be necessary. I also have a Xilinx ZC706 Board and schematic. On this Board, the identical PHY is connected to the PS_MIO, but there are no such external resistors to Vtt. I'm thinking this is possible because the MIO pins have DCI, but I can't find any references that indicate this.

 

Can anyone speak to if I need the external pullups to Vtt and why when interfacing to the MIO they don't seem to be necessary 

Tags (3)
0 Kudos
1 Solution

Accepted Solutions
9,554 Views
Registered: ‎01-08-2012

Re: HSTL RGMII interface need for external Vtt

Jump to solution

Hi.

 

Those aren't "pullups" - they're terminations.

 

I've used RGMII without terminations on a few designs without problems.

 

Terminations MAY be needed.  Whether they are actually needed for your board depends on your board.  The best way to tell is to simulate the response of the board using some SI simulation tool.  (Yes, you can even use a free version of Spice if you can manually extract the models from the IBIS files and your PCB.)

 

Rough rule of thumb:  If the trace impedance on the board is close to 50ohm AND there are no branches or stubs on the traces (i.e. point to point topology) AND you don't run the traces really close together AND the length is less than a few inches AND you use ground planes properly, you probably don't need to worry about terminating RGMII.

 

2nd rule of thumb: don't trust rules of thumb.  Perform a simulation.

 

Allan

View solution in original post

0 Kudos
2 Replies
9,555 Views
Registered: ‎01-08-2012

Re: HSTL RGMII interface need for external Vtt

Jump to solution

Hi.

 

Those aren't "pullups" - they're terminations.

 

I've used RGMII without terminations on a few designs without problems.

 

Terminations MAY be needed.  Whether they are actually needed for your board depends on your board.  The best way to tell is to simulate the response of the board using some SI simulation tool.  (Yes, you can even use a free version of Spice if you can manually extract the models from the IBIS files and your PCB.)

 

Rough rule of thumb:  If the trace impedance on the board is close to 50ohm AND there are no branches or stubs on the traces (i.e. point to point topology) AND you don't run the traces really close together AND the length is less than a few inches AND you use ground planes properly, you probably don't need to worry about terminating RGMII.

 

2nd rule of thumb: don't trust rules of thumb.  Perform a simulation.

 

Allan

View solution in original post

0 Kudos
Explorer
Explorer
5,633 Views
Registered: ‎05-22-2008

Re: HSTL RGMII interface need for external Vtt

Jump to solution

Thanks for your response Allan,

 

I kinda knew they weren't pullups I was jsut being lazy with my verbiage.

 

I get what you are saying about simulation, but I'm not certain that the person putting the schematic together has the time or abiltiy to do simulation. Boards on the cheap!

 

My EM knowledge hasn't been used in some years, but I did some reading, and If I understand it and you correctly the resistor is to match the impedance of the 50ohm transmission line, so as to remove the boundary condition, thus maximizing power delivered to the load and I think more importantly for this application minimizing signal reflection.

 

I'm going add the resistors, just to be safe. 

 

 

0 Kudos