模拟电路后仿仿真Step越跑越慢,不动了,怎么解决
时间:10-02
整理:3721RD
点击:
跑一个模拟电路的后仿仿真,tran仿真。
可是仿真时 仿真的 step越来越小,都到几 fs了,仿真时间根本就不动了,不知道是什么问题,怎么解决?
下面是仿真log文件的一部分:
....
cmin = 0 F
gmin = 1 pS
maxrsd = 0 Ohm
mos_method = s
mos_vres = 50 mV
tran: time = 108.7 ns(1.09 %), step = 3.142 fs(31.4 n%)
Warning from spectre at time = 108.733 ns during transient analysis `tran'.
Convergence difficulties resulted in error requirements being unsatisfied.
tran: time = 108.7 ns(1.09 %), step = 1.221 ps(12.2 u%)
Warning from spectre at time = 108.837 ns during transient analysis `tran'.
Convergence difficulties resulted in error requirements being unsatisfied.
tran: time = 108.8 ns(1.09 %), step = 104.2 ps(1.04 m%)
tran: time = 108.8 ns(1.09 %), step = 794.9 as(7.95 n%)
....
可是仿真时 仿真的 step越来越小,都到几 fs了,仿真时间根本就不动了,不知道是什么问题,怎么解决?
下面是仿真log文件的一部分:
....
cmin = 0 F
gmin = 1 pS
maxrsd = 0 Ohm
mos_method = s
mos_vres = 50 mV
tran: time = 108.7 ns(1.09 %), step = 3.142 fs(31.4 n%)
Warning from spectre at time = 108.733 ns during transient analysis `tran'.
Convergence difficulties resulted in error requirements being unsatisfied.
tran: time = 108.7 ns(1.09 %), step = 1.221 ps(12.2 u%)
Warning from spectre at time = 108.837 ns during transient analysis `tran'.
Convergence difficulties resulted in error requirements being unsatisfied.
tran: time = 108.8 ns(1.09 %), step = 104.2 ps(1.04 m%)
tran: time = 108.8 ns(1.09 %), step = 794.9 as(7.95 n%)
....
有人说是内存的问题,内存不够。但是没有说解决方法。不知道有谁碰到过类似的问题没有,怎么解决呢
是不收敛的问题。
在那个时间点上有大信号阶跃响应。
解决办法的话,我一般是修改absi 和 reli 这两个参数。
非常感谢。先去试一下
是不是震荡了
估计是不收敛,如果电路当中有太多floating点的话可能会引起这种现象,或者就按3楼的方法,调一下spectre的精度,retol 等等,这样可能就会解决了。