cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Highlighted
904 Views
Registered: ‎03-03-2017

XVphy_IBufDsEnable(InstancePtr, QuadId, XVPHY_DIR_TX, (FALSE)) causing GTREF input DC level to go to 0V

Hi,

   I am working on debugging a new piece of hardware using Kintex-7 325T using Vivado/SDK 2018.1 and I seem to be having issues with the HDMI and/or the GTX setup on the hardware side.   If I program the FPGA in "bootloop" then the MGTREFCLK0P_115 / MGTREFCLK0N_115 seem to power up to around 0.8V (which is what is expected according to ds182_Kintex_7_Data_Sheet.pdf (page 4 table 54 (Vcmin) 2/3*Vmgtavtt of 1.2V).

   But when I step through the SDK code and run the XVphy_IBufDsEnable(InstancePtr, QuadId, XVPHY_DIR_TX, (FALSE)) command (line 470 of bsp/proc_mblz/libsrc/vphy_v1_7/src/xvphy_hdmi.c) the GTX reference input common mode voltage goes down to 0V and nothing in the HDMI SDK seems to work.   I know the SDK code is good since I copied it over from a similar working design/hardware.

   Can somebody please help in determining what might be going on with the hardware to cause this?

   I have verified that MGTAVTT=1.20V being powered by a MAX8556ETE LDO which is supplied by a 1.7V coming from a IAH12040A007V-007 DC/DC being powered by a 12V 5A wall-wart power adapter, so I am pretty certain that the 1.2V is solid.

 

Thanks.

Tim

Tags (3)
0 Kudos
Reply
1 Reply
Highlighted
893 Views
Registered: ‎03-03-2017

Never mind, this turned out to be an issue where since I forgot to program my HDCP keys, the HDMI IP never brought the IBUFDS inputs back to their 0.8V levels.   Once I programmed the keys properly now the HDMI seems to work.

The IBUFDS inputs still do go to 0V, but eventually come back.

Is this expected behavior?

 

Thanks.

Tim

0 Kudos
Reply