Intel® Quartus® Prime Software
Intel® Quartus® Prime Design Software, Design Entry, Synthesis, Simulation, Verification, Timing Analysis, System Design (Platform Designer, formerly Qsys)
17252 Discussions

PIN - PIN Connection Singal not same

Altera_Forum
Honored Contributor II
1,297 Views

Hi all, 

 

i have quartus project just input clock pin connected to output pin. 

 

this is the result i got in scope. just see the jpg attached. 

 

any suggestions 

 

regards 

 

video
0 Kudos
4 Replies
Altera_Forum
Honored Contributor II
558 Views

 

--- Quote Start ---  

Hi all, 

 

i have quartus project just input clock pin connected to output pin. 

 

this is the result i got in scope. just see the jpg attached. 

 

any suggestions 

 

regards 

 

video 

--- Quote End ---  

 

 

Hi, 

 

I'm not a signal integrity specialist, but for me it looks like an un-terminate trace. 

The over- and undershoots are the result of reflections. I assume you 

have used a scope with a 10x1 probe ?
0 Kudos
Altera_Forum
Honored Contributor II
558 Views

Another thing to try is to reduce the drive strength of the output pin in Quartus. This will slow the edge rate down and reduce reflections.  

 

Are you sure the ground path on your scope probe is short enough?
0 Kudos
Altera_Forum
Honored Contributor II
558 Views

hi all, 

 

in assignment editor i changed the drive strength, but not helpful, 

 

i checked with jtag, signal tap II, the signal inside is very clear and logic is very good.  

 

regards, 

 

video_signal
0 Kudos
Altera_Forum
Honored Contributor II
558 Views

I see from your last post, that you apparently don't yet understand the difference between a logical signal viewn in Signal Tap and it's waveform shown by an oscilloscope. It may be helpful to read e. g. AN466 Cyclone III Design Guidelines to understand the issues of high speed digital signal interfaces. 

 

Just some additional remarks. 

 

1. A high drive strength FPGA output connected to an unterminated PCB trace will most likely show a waveform similar to your results. 

 

2. With minimum drive strength, assuming a regular PCB design with sufficient ground connection, the waveform should be expected to show no or minimal overshoot. 

 

3. When you connect an usual passive oscilloscope probe, you're effectively unable to see the real waveform, cause the probe impedance is distorting it considerably. The probes ground inductance does the rest. It's not completely impossible to capture realistic waveform with a passive probe, but requires some skill. 

 

An active probe or low impedance resistive divider would be required for an almost exact waveform display.
0 Kudos
Reply