为什么analoglib中的sw仿真时候高电平不导通

fantasy19853

之前用理想sw的时候都是直接用，没设置过导通电压和断开电压，之前都能用，这次好奇怪用不了了，但我设置之后仿真不收敛了
我很奇怪为什么之前不设置都能正常用，但这次就是不导通，欢迎路过大佬不吝赐教，也欢迎讨论，谢谢大家！！

fantasy19853

把SW换成TG门也不行，仿真方法换成trap还是不行，应该是电路哪里有简单问题，导致仿真一直不收敛
Notice from spectre at time = 45.5863 fs during transient analysis `tran'.
    Newton iteration fails to converge at time = 45.5863 fs step = 20e-21 s.
        Disaster recovery algorithm is enabled to search for a converged solution.
Notice from spectre at time = 3.56568 ps during transient analysis `tran'.
    Newton iteration fails to converge at time = 3.56568 ps step = 20e-21 s.
        Disaster recovery algorithm is enabled to search for a converged solution.
Notice from spectre at time = 3.56585 ps during transient analysis `tran'.
    Newton iteration fails to converge at time = 3.56585 ps step = 20e-21 s.
        Disaster recovery algorithm is enabled to search for a converged solution.
Notice from spectre at time = 3.56585 ps during transient analysis `tran'.
    Newton iteration fails to converge at time = 3.56585 ps step = 20e-21 s.
        Disaster recovery algorithm is enabled to search for a converged solution.
Notice from spectre at time = 3.56585 ps during transient analysis `tran'.
    Newton iteration fails to converge at time = 3.56585 ps step = 20e-21 s.
        Disaster recovery algorithm is enabled to search for a converged solution.
        Further occurrences of this notice will be suppressed.
Warning from spectre at time = 3.56585 ps during transient analysis `tran'.
    WARNING (SPECTRE-16191): Minimum time step used.  Solution might be in error.
Warning from spectre at time = 3.56585 ps during transient analysis `tran'.
    WARNING (SPECTRE-16191): Minimum time step used.  Solution might be in error.
Warning from spectre at time = 3.56585 ps during transient analysis `tran'.
    WARNING (SPECTRE-16191): Minimum time step used.  Solution might be in error.
Warning from spectre at time = 3.56585 ps during transient analysis `tran'.
    WARNING (SPECTRE-16191): Minimum time step used.  Solution might be in error.
Error found by spectre at time = 843.34 ns during transient analysis `tran'.
    ERROR (SPECTRE-16192): No convergence achieved with the minimum time step specified.  Last acceptable solution computed at 3.56585 ps.

The values for those nodes that did not converge on the last Newton iteration are given below.  The manner in which the convergence criteria were not satisfied is also given.
            Failed test: | Value | > RelTol*Ref + AbsTol

Top 10 Solution too large Convergence failure:
    I(E0:p) = 5.86508 A, previously 24.1731 fA.
        update too large:  | 1.28001 GA | > 122.21 uA + 1 pA
    I(V11:p) = 4.61391 kA, previously -1.64709 uA.
        update too large:  | 43.4052 MA | > 94.53 mA + 1 pA
    I(V12:p) = 60.4996 kA, previously 2.95919 nA.
        update too large:  | 150.336 MA | > 1.23949 A + 1 pA
    V(net058) = 15.582 V, previously 487.392 mV.
        update too large:  | -300 mV | > 339.726 uV + 1 uV
    I(E1:p) = 980.581 GA, previously -2.07018 mA.
        update too large:  | 7.3067 GA | > 20.0898 MA + 1 pA
    I(V3:p) = 862.129e+15 A, previously 441.531 kA.
        update too large:  | -1.11034e+15 A | > 17.663 TA + 1 pA
    I(V7:p) = 2.44722e+15 A, previously 2.00277 kA.
        update too large:  | -2.63895 TA | > 50.1376 GA + 1 pA
    V(I0.I33.net056) = 3.02378 kV, previously 383.182 mV.
        update too large:  | -600 mV | > 61.9705 mV + 1 uV
    V(I0.I33.net058) = 6.74698 kV, previously -300.661 nV.
        update too large:  | -600 mV | > 138.25 mV + 1 uV
    V(I0.I33.net055) = 3.50921 kV, previously 33.2362 mV.
        update too large:  | -300 mV | > 71.9159 mV + 1 uV
Top 10 Residue too large Convergence failure:
    V(net10) = 18.6025 GV, previously 1.25092 V.
        residue too large: | -171.049e+15 A | > 3.50438 TA + 1 pA
    V(net22) = 7.66248 GV, previously 1.2504 V.
        residue too large: | -29.0214e+15 A | > 594.579 GA + 1 pA
    V(net21) = 7.66248 GV, previously 1.2504 V.
        residue too large: | -29.0214e+15 A | > 594.579 GA + 1 pA
    V(rst) = 0 V, previously 2.49911 V.
        residue too large: | -163.17 TA | > 3.34296 GA + 1 pA
    V(net33) = 0 V, previously 1.60854 V.
        residue too large: | -319.691 GA | > 6.54971 MA + 1 pA
    V(EN_CLK) = 2.5 V, previously 0 V.
        residue too large: | -61.8204 TA | > 1.26655 GA + 1 pA
    V(rst_n) = 2.5 V, previously 891.463 uV.
        residue too large: | -332.537 TA | > 6.81289 GA + 1 pA
    V(en_flash_vref) = 2.5 V, previously 891.463 uV.
        residue too large: | -4.61391 kA | > 94.528 mA + 1 pA
    V(net13) = 18.8662 GV, previously 1.25094 V.
        residue too large: | -175.935e+15 A | > 3.60449 TA + 1 pA
    V(I0.net019) = 2.63035 MV, previously -82.5125 mV.
        residue too large: | -7.61056 GA | > 155.922 kA + 1 pA

fantasy19853

fantasy19853 发表于 2024-10-23 13:21
把SW换成TG门也不行，仿真方法换成trap还是不行，应该是电路哪里有简单问题，导致仿真一直不收敛
Notice fr ...

cmin和retol也改了还是不行，估计是电路哪里有问题，但是checkandsave也没问题呀，好烦

acrofoxAgain

'check and save只能检查一下最基本的错误，甚至有的“错误”都未必是错误。其实check的主要作用是建立网表信息。

zt_ic222

换个原理图试一下，新建，不要copy

fantasy19853

zt_ic222 发表于 2024-11-5 14:03
换个原理图试一下，新建，不要copy

好的，谢谢您的回答，我copy了一个可以收敛的仿真，然后在这个里面改可以吗，我现在这个就是这样，原本可以收敛，我在一阶的基础上加到了二阶（没从原本不收敛里面复制任何东西，都是从这个可以收敛的原理图上copy本身搭的），但是加成二阶就不收敛了。明天有空我再从零搭试试，非常感谢您的回答，祝您身体健康万事如意啦~

fantasy19853

zt_ic222 发表于 2024-11-5 14:03
换个原理图试一下，新建，不要copy

真行了wok，我从copy可以仿真的schmatic在里面改都不行，谢谢哥，
您在为什么analoglib中的sw仿真时候高电平不导通 - Analog/RF IC 资料共享 - EETOP 创芯网论坛 (原名：电子顶级开发网) -原帖发一下吧：

换个原理图试一下，新建，不要copy

，我给您设置最佳答案，谢谢您，祝您身体健康万事如意！！！

zt_ic222

fantasy19853 发表于 2024-11-6 16:49
真行了wok，我从copy可以仿真的schmatic在里面改都不行，谢谢哥，
您在为什么analoglib中的sw仿真时候高 ...

哈哈，我就知道小倒霉蛋的人设不止我一个，还有我也不是哥