I designed a negative resistor LC oscillator, for the first time, I simulated it in spectre, it's frequency was about 2.5GHz, without changing any thing, I simulated it for the second time, it's frequency changed to about 5GHz! why! Anybody met this thing?
The freq of my osc is 2.4GHz. Initlal condition was the same ipulse(current1 200uA, current2 0uA, raise time 1ns)! For the majority of the time, the frequecy is about 5GHz, some time ,it is 2.5 GHz. Now, I can not get the 2.5GHz output
thanks!
if u use pss run the simulation it is possible.
pss sometime is not correct. i guess it is 5GHz. At some case, pss take two cycles as one , so cause 2.5GHz.
monte carlo analysis is the beginners level Yield analysis.
to put it simple with an example
threshold voltage is a parameter which affects u r ckts performance.
and it varies across the wafer, In spice u can model this variation into any of the random distribution.
So spice will vary the threshold voltage(foundry will provide the distribution model and the level of variation) according to the distribution do multiple simulation.
this will help u make sure u r ckt works in most of the cases.