Forum Discussion
ContributorEnable on-chip termination for Differential Transceiver clocks
Hi,
as far as I understand you are still searching for valid .qsf entry for transceiver clock input termination.
Valid settings are listed in this document: https://www.intel.com/content/www/us/en/docs/programmable/683296/24-3.html
As far as I see, the settings are the same for previous Quartus Pro versions.
The option for transceiver reference clock termination is XCVR_A10_REFCLK_TERM_TRISTATEallowed values are TRISTATE_OFF and TRISTATE_ON
TRISTATE_OFF corresponds to 100 ohm differential termination.
Some confusion is caused by the fact, that Quartus Std. uses a different option name
XCVR_REFCLK_PIN_TERMINATIONvalues are AC_COUPLING, DC_COUPLING_EXTERNAL_RESISTOR and DC_COUPLING_INTERNAL_100_OHMS
See: https://www.intel.com/content/www/us/en/docs/programmable/683084/current/settings-file-reference-manual.html
To check which termination has been actually implemented, you can review QPP fitter report/Plan Stage/Input Pins.
Regards
Frank
Super ContributorHi,
as far as I understand you are still searching for valid .qsf entry for transceiver clock input termination.
Valid settings are listed in this document: https://www.intel.com/content/www/us/en/docs/programmable/683296/24-3.html
As far as I see, the settings are the same for previous Quartus Pro versions.
The option for transceiver reference clock termination is XCVR_A10_REFCLK_TERM_TRISTATE
allowed values are TRISTATE_OFF and TRISTATE_ON
TRISTATE_OFF corresponds to 100 ohm differential termination.
Some confusion is caused by the fact, that Quartus Std. uses a different option name
XCVR_REFCLK_PIN_TERMINATION
values are AC_COUPLING, DC_COUPLING_EXTERNAL_RESISTOR and DC_COUPLING_INTERNAL_100_OHMS
See: https://www.intel.com/content/www/us/en/docs/programmable/683084/current/settings-file-reference-manual.html
To check which termination has been actually implemented, you can review QPP fitter report/Plan Stage/Input Pins.
Regards
Frank
ContributorThanks for the reply and the explanation Frank!
When I check the fitter report, I see that the positive differential pairs that I added to my qsf (please refer to my entry on Nov 13th) have a termination called "refclk_divider_enable_termination=enable_term", whereas the negative pairs have the termination "tristate_off" (I'm assuming that's the default bcs I didn't specify the termination for the negative differential pair). But as I said earlier in this thread, my clocks and the design work as expected. I was thinking that Quartus was converting the invalid syntax to the default settings under the hood, but according to the reports that's not the case either..
Anyways, I only added the positive clocks with the termination you suggested to my qsf. Should I add the negative pairs of the differential clocks as well (apparently the negative end doesn't get the assignment that the positive end gets from the .qsf) ??
Kind regards
