How MOSFET leaks

[ayeaye]

Yes, as you see on the currents diagram, the opposite current caused by the inductance is much less than the charging and discharging current of Cgs. Also, the inductance doesn't work very well when it doesn't have a closed circuit with a low resistance, and your circuit through drain-source has an almost infinite resistance. So your test mostly shows passing voltage through the capacitance.

Yet, in your channel one, do you notice that the upper edges of the pulses go slightly down, and the lower edges after pulses slightly rise? This is likely the effect of the inductance, because inductance works against the change of voltage, it tries to bring it down after it rises, and to bring it up, after it falls.

If you are interested, below is the diagram from the same simulation i did before, of the voltages at the output of the pulse voltage source (blue) and at the gate (red), between them is a 10k resistor. The voltage at the gate mostly shows the charging and discharging of Cgs with exponential curves. Yet, we see the inductance at the end of the discharge, where the voltage very slightly rises. We should also see it at the end of the charge, but this weird horizontal line makes it impossible to see. I don't know what it is. Because of inductance there should be a decrease of the rise of the voltage slightly more than by the exponential curve.

But, about the mosfet leaking, the mosfet did not anyhow considerably leak without the inductance, this seems to be the point.

[TinselKoala]

When you say "weird horizontal line" are you referring to the Miller plateau?

Maybe this Vishay application note will help.

http://www.vishay.com/docs/73217/73217.pdf

[TinselKoala]

Yet, in your channel one, do you notice that the upper edges of the pulses go slightly down, and the lower edges after pulses slightly rise? This is likely the effect of the inductance, because inductance works against the change of voltage, it tries to bring it down after it rises, and to bring it up, after it falls.

There is substantially more inductance in the probe connections, particularly the Ground clip wire, than there is in the mosfet itself.

[TinselKoala]

Here's another example from real life. You can clearly see the Miller plateaus, as well as the inductive switching transient ringing at the end of the pulse.

[ayeaye]

Thanks for your info about the Miller plateau. As much as i understand, this has to do with the increase of capacitance during the channel formation. As i understand, the capacitance is not only between the electrodes, but also the n regions or something, so when they grow, the capacitance grows.

Well, my primitive understanding of the mosfet, n-mosfet, it consists of a p semiconductor, or as i prefer to say, p-conductor, which though is only slightly p, or something, so it almost doesn't conduct. Then, the positive charge at the gate attracts electrons into it, making it an n-conductor, which forms a conducting channel between the source and the drain.

I prefer to say n-conductor, because the prefix "semi" is very misleading in the word semiconductor. They are conductors, but different from metals where electrons form an electron gas, the electrons there are tied to atoms, and hop from one atom to the other, something. Now this can happen when there is either excess of elecrons, or lack of electrons. In the latter case an electron jumps to the next atom to fill the lack of electrons there, called a hole, and leaves a hole behind in the atom where it jumped from, this is a p-conductor.

The inductances in the mosfets have to do with connecting the electrodes to the pins, something. By the irf530n datasheet http://www.irf.com/product-info/datasheets/data/irf530n.pdf the source inductance Ls in that mosfet is 7.5 nH. A very low inductance, i wonder that it causes so considerable effect. The drain inductance Ld is 4.5 nH.