I want to generate this kind of signal with my spartan 3 XCS200 :
formatting link
I am using the DCM to multiply my oscillator ( 50Mhz ) by 2 ; but, I don' succeed to have this "3.8 ns plate" for each level. I have it only for hig logic level. How can I do ???
Are you sure that you do not succeed? You need more than 1GHz of Scope and probe bandwidth to reliably measure signal slopes in the order of 1ns. (As the frequency is fixed, the plate width is defined by the slope of the signal and the duty cycle)
For example with a 150MHz scope you will see a sine wave, no matter what waveform the FPGA generates.
With a perfectly attached 500MHz setup you will see only the one overtone (300MHz) undamped. The 500MHz part will allready be damped. So even if the FPGA would ouput a perfect square wave your scope would see something that looks worse then the K=5 case in that URL:
formatting link
If you clamp the gnd connection of a regular probe somewhere on your board and just touch the probe to your signal you are going to see something that is a lot worse. Follow the link that Falk suggested and solder a coax cable with a resistor to your signal.
ALSO: If your signal is unterminated it will look really bad. If it is series terminated it will only look good at the receiver.
ALSO: Select fast slew rate and 24mA drive strength for your output.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.