I was recently asked by an engineering manager if running IR drop analysis was really necessary. The argument to support his question was that his engineering team always over-designs the power rails, and so the risk of getting high IR drop was so small that analysis was not required.
The easiest way to answer his question was to relate what actually happened during the creation of a DAC demo a number of years ago.
The demo was designed to initially suffer from high IR drop in the center of the design, and the plan was to short the VDD of this region to a solid VDD, and hence fix the high IR drop. Figure 1 shows the initial IR drop analysis results, with the high IR drop in the center of the design. You can also see the solid VDD net just below the red region of IR drop.
While the added jumper did help reduce the highest IR drop in the center, it also enabled current to flow through a completely different path from the lower right boundary VDD I/O pad to the center of the design. This new path caused higher currents to flow around a memory, and resulted in new electromigration (EM) failures around the sense-amps of the memory. Figure 2 shows a zoomed in view of the lower right of the design, where you can see the added VDD jumper and the new EM failures.
When we saw this result, it surprised everyone involved and we were all experts with many years of experience in both chip design and IR drop analysis. No-one could have predicted how the addition of a small jumper in the center of the design could have caused new EM problems towards the corner of the design.
That's the reason why power rail analysis is required!
While we would like to believe that we can predict how current flows through the power network, in reality the current often flows unpredictably through multiple levels of interconnect and complex gridded power networks. It is the inability to predict how the current flows that forces the need for analysis. How can you possibly over-design the power rails when you don't know where, and how much, current is flowing though them?
Anyone have other thoughts on this?
Good article, giving a practical example. I am new to Voltagestorm or EPS and have used for static analysis only. However, it helped a lot in our design. I liked the feature in EPS, where it shows how much current flows through a node, the resistance, current density, etc. It also tells us in micro level what's the capacity of a metal strip to carry current. It helped us in defining the width of the metals to design analog power network.
I would also like to explore the dynamic EMIR flow.
Kari hits a great point on missing vias, or single vias where there was meant to be an array. Unfortunately neither DRC nor LVS can help you much, because of the gridded nature of the power networks. Tools like the Encounter Power System (and previously VoltageStorm), can easily find these problems and report them to you. It is even possible to find a VDD-VSS short, if you know what you're doing :)
I have empathy with Daniel, I had a similar experience when micro-probing an older design, except that my problem looked more like a volcano, the metal completely splattered ... the good old days, when we were lucky if DRC actually completed on designs!
In the olden days I watched my DRAM die under a microscope as a portion of the 6um metal interconnect would bubble and overheat due to excessive currents. IR drop and EM has always been an issue, even in the 6um era. I would've loved to use a software tool to pinpoint these failures so that I wouldn't need to make 21 spins of silicon to get my chip correct.
I completely agree, Pete. There are MANY reasons to do IR-drop analysis. One reason is that you can over-design the grid all you want, but if you happen to be missing a few key vias, that robust grid is not going to do what you think. IR-drop analysis can point out some things you didn't know were missing.
Also, it is important to not forget about the package. Designers really need to look at the die and the package together when doing IR Drop analysis.