The Tech Blog

You would have heard this a countless number of times, “Schottky diodes are fast”. Ever stopped to wonder what is “fast” about them and why? Let’s explore that today

When a Silicon PN jn diode is forward-biased, current flows through it once the voltage is above the depletion layer potential. Now assume the case, where you suddenly remove the forward voltage and apply a negative voltage across it. We expect the diode to become reverse-biased and have no current flow. This occurs, but not as immediately as you think. There is a finite delay post the switching, where the current flowing across the junction goes negative(Or flows opposite). This occurs because when the diode is conducting, minority carriers (electrons in the P-type material and holes in the N-type material) accumulate near the junction. When the diode switches off, these minority carriers must return to their original regions or recombine. The accumulated charge in the junction decreases to zero. Now the current starts flowing in the opposite direction and this is known as the reverse recovery current. It goes up to a negative value and then slowly recombines back to a zero state when the diode becomes non-conducting. The delay time needed for it is called Reverse Recovery Time.

This delay causes power losses whenever a diode voltage switches so you always try to use diodes with low reverse recovery time. It becomes critical in switching power supplies with high switching rates. Schottky diodes don’t have this problem because it’s not a PN junction. It’s formed with a N-type material and a metal junction. Hence they have low switching loss and have a “fast” switching time with usually an order of magnitude faster times. So always check the switching times of diodes in the datasheet and use it to find the fastest edge switching rate in your application for the diode.

I was doing some layouts for high current measurements and thought we should address it here. Current Shunt resistors are normally connected in series with a load to measure the current flowing across it. So you tap the voltage across the two terminals across a shunt and feed to an amplifier(if needed) and then to an ADC to measure the voltage across it.

Since these resistors are usually for accurate measurements you really need to keep the following layout considerations in mind.

1. The length of the connections from the shunt to the amp should be symmetrical or be length-matched so that the PCB trace lengths are the same on both arms. It needs to be as short as possible to prevent any trace losses.

2. Connect them preferably with thick traces to reduce impedance issues(But not a big factor as the input impedance of the amplifier is very large anyway).

3. Make sure to tap it at two distinct points such that it’s away from the normal current path. This is known as a 4-wire Kelvin connection. This is a method used to measure very low resistances with high accuracy. It involves the use of two pairs of wires: one pair delivers the supply current to the component being measured, and the other pair is used to measure the voltage drop across the component (We can discuss this in a future post on why it’s needed). This method eliminates the impact of the resistance of the lead wires and contact resistance in the circuit, which can introduce significant errors in low resistance measurement. For this reason, you will find current sense resistors with unique 4 terminals. Check images.

4. If you are having multiple current sense resistors in parallel(for power dissipation reasons), make sure to connect it in a way that current flows through each of the resistors, Tapping at let’s say a wrong pad can cause a current mismatch in the arms.

Hoping these help next you do a physical layout for current sensing. Happy Designing!