Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Problems of using the "dft" function of caculator

Status
Not open for further replies.

stephenlucky

Newbie level 1
Joined
Nov 9, 2006
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,296
In Cadence,
I want to simulate the SFDR of my design by "dft" in caculator. But I got a very small SFDR, no more than 20dB. I think this is not correct, not because of my design, but my simulation. I check my simulation, and I find a strange thing that when I caculate the input signal (by ''dft'' in caculator), which is set as a ideal sin wave (the frequency is "fsin", and the period is "Tsin") by ADE in simulation setup, I got a big pulse in the point of "fsin" in frequency domain, and many small pulse of other frequency, just like the backgroud noise. In frequency domain, The difference of the max value of "fsin" and the max value of those small pulses is only about 66dB. That means I input a NOT clean singal to my circuits, how could I got a good SFDR? And I think I should get only a pure frequency point "fsin" when I "dft" the input ideal sin wave, but why those small pulses appear?
I setup the "dft" condition fsin=40M, start time=2u, end time=4u, numble of sample=1024, the length of my simulation time is 6u. why I could not get the ideal dft result?

who could help me? many thanks!
 

Re: Problems of using the "dft" function of cacula

refer to this paper.
 

several parameters will affect your dft simulation results. Your fin is 40M, i.e, the period is about 25ns. so my suggestion is that is start time=2u, end time is 2.1u, number of sample is 16384. If the simulation results is still no good enough, you should set a smaller transient simulation step.
 

Re: Problems of using the "dft" function of cacula

pfd001 said:
several parameters will affect your dft simulation results. Your fin is 40M, i.e, the period is about 25ns. so my suggestion is that is start time=2u, end time is 2.1u, number of sample is 16384. If the simulation results is still no good enough, you should set a smaller transient simulation step.


I have meet the same problem.
why select "start time=2u, end time is 2.1u, number of sample is 16384"? thanks
 

Re: Problems of using the "dft" function of cacula

i think you need restrict the convergence rule, setting a smaller value for tolerance. you should set a smaller transient simulation step too as pfd001 say.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top