急问_为什么Specture方针时总报下列一个warning！

benbenbear

Warning from spectre during IC analysis, during transient analysis `tran'.
    I0.I20.M0: The bulk-source junction current exceeds `imelt'.  The results
        computed by Spectre are now incorrect because the junction current
        model has been linearized.
    I0.I20.M0: The bulk-source junction current exceeds `imax'.
    I0.I20.M0: The bulk-drain junction current exceeds `imelt'.  The results
        computed by Spectre are now incorrect because the junction current
        model has been linearized.
非常不理解为什么，就是把那个MOS管的宽长比加大好几倍也会出现这种情况
并且好多个单元都会这样，流过传输门的电路也不大，大概也就有3~4mA吧，可是管子的宽长比已经接近100：1
了。
鄙人感觉非常迷惑！！！
请求高人解答。
谢谢！

benbenbear

Warning from spectre during IC analysis, during transient analysis `tran'.
    I0.I20.M0: The bulk-source junction current exceeds `imelt'.  The results
        computed by Spectre are now incorrect because the junction current
        model has been linearized.
    I0.I20.M0: The bulk-source junction current exceeds `imax'.
    I0.I20.M0: The bulk-drain junction current exceeds `imelt'.  The results
        computed by Spectre are now incorrect because the junction current
        model has been linearized.
非常不理解为什么，就是把那个MOS管的宽长比加大好几倍也会出现这种情况
并且好多个单元都会这样，流过传输门的电流也不大，大概也就有3~4mA吧，可是管子的宽长比已经接近100：1
了。
鄙人感觉非常迷惑！！！
请求高人解答。
谢谢！

lwjbh

能把电路图贴出来么，看一下

moorehuang

平时好像仿真不收敛的时候经常会有这种warning
我觉得这是仿真器的问题，通过很复杂的矩阵运算求解，有可能解越来越大，因此出现什么超过最大电流这类的warning
我个人觉得没有必要修改管子的尺寸，如果能收敛，关注结果就可以了

xzyan

有的时候是激励源设得太陡了（例如上升时间设得很短1ps）会出现这个问题。

langtaosha

imet是指能够融化器件的超大电流！电路仿真有时流过MOS器件的源漏到衬底的寄生二极管发生正偏导致极大电流，从而迫使Spectre将这个非线性器件进行线性化来得到一个收敛解，但此时的结果已经不太准确了！
通常有两个原因导致会出现这个问题：1、模型中的寄生（特别是与二极管串联的电阻）没有。2、电路工作条件设置得不太合理！

santaxy

我也碰到过类似问题，学习一下

linuxwinner

I think any warning message cannot be neglected unless it is truly recognized. As specified, if the device is used in transmission gate, I think its bulk is connected to wrong place. You can try to connect it to its source if it is pmos.

amay

学习了

arong00

请问你最后怎么解决这个问题的，我仿真时也遇到这个问题了，非常麻烦，请多多指教，万分感谢！