- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic to the Top
- Bookmark
- Subscribe
- Printer Friendly Page
RXUSRCLK-s ourced problem in V5 GTX
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content
07-15-2012 07:57 PM
When RXUSRCLK is sourced by TXOUtCLK+BUFG,the GTX works normally. However, when When RXUSRCLK is sourced by RXRECCLK+BUFG,the GTX does not work well. In the configuration, he clock correction is used, and TX/RX buffer is also used. My question is why RXUSRCLK can not be sourced by RXRECCLK+BUFG?
Thanks
Re: RXUSRCLK-s ourced problem in V5 GTX
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content
07-24-2012 02:11 AM
Re: RXUSRCLK-s ourced problem in V5 GTX
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content
08-03-2012 04:55 AM
When RXUSRCLK is sourced by RXRECCLK+BUFG, the RX side of GTX can not properly receive data ( part of FC core 3.3). However, it works during simulation.
Is it necessary to present the parameter configuration of GTX?
Re: RXUSRCLK-s ourced problem in V5 GTX
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content
08-03-2012 05:02 AM
CDR takes reference of the serial clock generated by PLL and recovers the clock based on the transitions in the incoming data. So, the stability of the RXRECCLK depends on the amount of toggling in the receive data and the quality of the REFCLK given to the PLL. By quality of the REFCLK, I mean the amount of jitter in the REFCLK input should be less.
If there is any jitter on the REFCLK it can translate to RXRECCLK.
Re: RXUSRCLK-s ourced problem in V5 GTX
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content
08-03-2012 06:32 AM
I try the case. REFCLKOUT of GTX drives the DCM, whose ouput clock (+BUFG) drives RXRECCLK. It works. Does it mean the amount of jitter in the REFCLK input is less?
Re: RXUSRCLK-s ourced problem in V5 GTX
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content
08-09-2012 06:06 AM
Hi, venkata.
May the jitter of REFCLK be the key problem?











