How do chip design engineers make sure that there is no interference between different circuit modules due to electromagnetic fields, given that every circuit component is very close to each other?
In: 628
Every conductor thrhough which passes a current has an electromagnetic field. This is law. And, by extension, integrated circuits have electromagnetic fields and they can interfere with the circuit components.
The engineers are careful to protect the components through different methods that I find impossible to ELY5.
The EM fields can be calculated and where engineers consider there is a need for EMI protection they implement one of those measures for better protection. That, for example, can be done with capacitors that interrupt the circuit (there is no current passing through a capacitor because it’s basically two metal bars separated by an isulating gas. One metal plate is charged by the current and through energy transfer the second plate is charged, letting go of current through the rest of the circuit.)
Big power consumers are kept further appart to avoid EMI. For example you don’t want your microcontroller too close to the supply. Or if you have two chips in your integrated circuit, you want to give them space.
There is even more consideration than electromagnetic compatibility (EMC) and electromagnetic interferences (EMI).
Heat is also considered. Dissipated heat can also be calculated and of course the easiest way to protect against it is by keeping big power consumers further appart. For this there also exist special layers of the integrated circuit that are good heat conductors so they disippate energy. Microprocessors for example have a paste and a heat disipation material under them, and if the chip is not soldered perfectly it can actually deteriorate because the heat is not spread accordingly.
Even more, the connecting lines between components can also cause problems like parasitic capacitances and current reflection. Parasitic capacitance can happen if two lines are too close or even a right angle of the same line has the risk of making a parasitic capacitance. As i said, a capacitor is two metal bars separated by an insulator. Line reflections happen mostly when there is a power surge in the circuit, like when turning it on. When the supply switches on, there will be a power spike in the circuit before it stabilises. Power surges or even just normal curent flow can reflect on the line, meaning that when current goes from component A to component B, a very small portion comes back. The reasons behind this phenomenon are unknown to me (although I did learn about them in college), but what I do remember is that the longer the line, the higher the reflection can be. So it’s a challenge to get rid of EMI, heat, and reflection together.
There can also be cost reduction issues and they try to reduce the amount of material.
There are virtual instrumentations that allow integrated circuit design which automatically takes care of many problems so you can solve them at design level.
Of course, integrated circuits are EMC tested and if they are within accepted parameters they are considered good for sale.
All electronics have a mandatory EMC compliance and the proof of it is the small stylized “CE” printed on them.
This is all what I can remember from my college classes. I welcome anyone to correct me if I described something wrong.
Latest Answers