Tech Forums
Coupling Capacitance and aggressor nets
/ 0
- Details
- Category: ASIC - Backend
- Written by Harsh
- Hits: 159
Hi ,
I have two nets net1 & net2. They have some coupling cap inbetween them. StarRc (extraction tool) generates coupling report file.
In that i could see for the same two nets coupling cap is different in two different lines. Let me give an example …
1.62uf net1 net2
3.42uf net2 net1
In above two lines first column represents coupling cap between those two nets, second coulmn is for victim nets & last column is for agressor nets.
1) Can we have different coupling caps for the same two nets, like explained above?
2) In above example net1 is victim in first case & agressor in second case. It is reverse in net2 case. I am not understanding how the same net is behaving as victim in one case & agressor in other? I mean to say how net1 is victim as well as agressor when analyged with net2?
It can be either victim or agressor but not both when analyged with same net.
Please let me know if my understanding is wrong.
vinod
Hi Vinod,
See the spacing between these two nets. I suppose the two nets are long ets with different spacings.hence there may be chance of difference in the caps.
moreover the same net can be aggressor and victim also,
———
Kiran
Hi Vinod,
> >
> >
> >
> > As Kiran pointed is the distance between net1 & net2
> > is constant?
> >
> >
> >
> > Can the width of the same net be different at
> > different points? Is this
> > also the reason in your case…..
> >
> > Net1 with small width at one point and going to a
> > via attached to the
> > next metal layer with more width and so
> > on………….this also could
> > have that effect….. I am guessing….
> >
> >
> >
> > Regards,
> >
> > Praveen
Hi all,
As per my understanding 2 nets must have same coupling cap between them. Hey Vinod, tell me are those the only columns you saw. No matter what the distance of seperation be, if Net 1 has xfarads with net Net2 then net Net2 MUST have same x farads with Net 1. Thats why it is called coupling cap.
Its like this.. If you measure the distance from net 1 to net 2.. you will get the same distance when you measure from net 2 to net 1. and similarly the relative area is also same…
Kishore, I think ones we saw that in starRC.. 1 net, n a, was having some coupling cap with another net, nb. But net nb did not give any coupling cap with net na. Then we realized that if coupling cap is significant enough, if it exceeds certain limit then only that cap is reported. But what i am seeing here(in vinod’s exmple)is a bit different. coming to 2nd, aggressors and victims are just for terminology sake. When analysis is made on net 1 then its always a victim.. and can have 0, 1 or more aggressors. Though the driver of aggressor is far weaker than the victim, it still is an aggressor net.
Prashant C