I tried the simplest thing ever, to route the 200 MHz differential clocks
NET "clk_200M_p" LOC = AD12; # 200 MHz NET "clk_200M_n" LOC = AD11; #
directly to the two GPIO SMA connectors at
NET "test_outA" LOC = Y23; # GPIO-SMA, J13.1 NET "test_outB" LOC = Y24; # GPIO-SMA, J14.1
For whatever reason, my 500MHz scope just does not show up anything of these two 200MHz at the two SMA connectors.
I don't believe my KC705 board is defective, since I tested the 8 GPIO LEDs, and they are working at least.
NET "GPIO_LED(0)" LOC = AB8; NET "GPIO_LED(1)" LOC = AA8; NET "GPIO_LED(2)" LOC = AC9; NET "GPIO_LED(3)" LOC = AB9; NET "GPIO_LED(4)" LOC = AE26; NET "GPIO_LED(5)" LOC = G19; NET "GPIO_LED(6)" LOC = E18; NET "GPIO_LED(7)" LOC = F16;
Is there any comment? I did quite a lot of programming with ML506/ML507/ML605/SP605, and had never ever got this strange issue.
What IO Standard is the output? If it is LVDS, you need a termination resistor across the _p and _n signals. Do you have one? Most, if not all differential signal standards require some termination scheme. What is the bandwidth of your scope probe?