The beginning of the most complete MOSFET Miller effect analysisThe most complete MOSFET Miller effect analysis of the resistive load One person and one sentence cheers for the graduate school entrance examination Inductive load is the most analyzed version on the Internet, because inductive load is the most used scenario, such as switching power supplies, motors, relays, etc. The circuit analyzed on the Internet is the circuit that controls the relay, and then the coil of the relay is analyzed as an inductance, and because the inductor current cannot be abruptly changed, it is regarded as a constant current source. Here I think there is a certain problem, the first problem is that the specification of the relay is very clear, the coil is resistive and not small, tens of thousands of ohms. The second mistake is to regard him as a constant current source, of course, because his inductance is very large and the current changes slowly, but the relay coil is constant current, which means that the relay has been working, which is obviously unreasonable, and in the switching power supply, the inductor is ripple current, so I think the whole analysis is a little worse. In the resistive load, it is also mentioned that the whole process does not mention the reason for the decline of VDS, that is, the diode is not conductive and there is no clamping, so it starts to fall, and the explanation is not transparent enough.
First of all, let's take a look at the relay switching circuit, the circuit equivalent diagram is as follows.
In the above figure, it can be found that there is a difference in the resistive load in the above figure, that is, the current of ID does not almost reach the maximum at the end of the Miller platform, but is much lower than the maximum value, and it is still increasing after the Miller platform.
T0 T1 Phase:
This phase is the same as for resistive loads, because the current is 0 when the relay is not working. At this time, VIN is OV, VGS
T1 T2 Phase:
The waveform may be more resistive than the load that will almost always be greater than the supply voltage vcc. That is, the current of IGD is affected by the relay coil inductance L, because the voltage at both ends of the inductor is UL=LDi DT, and IGD=CGDDU DT, so VDS will be lifted up, when the forward voltage drop of the freewheeling diode D1 is greater than that of the relay, the diode will start to work, and VDS will be clamped to VCC+VF. VGS at this stage
T2 T3 Phase:
At T2, the VGS increases to VTH, at which point the MOS tube starts to open, the current starts to flow, and the state happens. VDS > VGS-VTH, so now the MOS tube has come from the cut-off zone to the constant flow zone.
At this time, as long as the VGS is a little larger than VTH, it will be amplified GM times and become ID, when the relay coil current is almost 0 at this time, when ID-IGS=IL>0, the inductor begins to charge, because there is a change in current, so there is UL at both ends of the inductor, and there is the existence of UR, so VDS=VCC-UL-UR begins to decline. VDS Drop According to the analysis of resistive load, the DU DT of CGD will become larger, resulting in most of the current flowing to CGD, and only a small part of the current flowing into CGS, resulting in a slow change of voltage at both ends of CGS, so a platform appears. This phase ends when VDS drops to VGS-VTH, where the current does not reach the maximum VCC R due to the influence of inductance, because it is not reached until UL=0.
T3 T4 and post-T4 phases:
At this time, VDS at this time, VCC charges L through R, so that UL gradually decreases, and IL or ID will gradually increase to VCC R.
Next, let's see that the inductor current is not 0, take the boost circuit as an example, the circuit is as follows:
As you can see from the waveform chart, the waveforms of VDS, ID, and VGS are completely different from those explained earlier, and of course they are also different from those described in other articles. Why it's different, let's take a look.
T0 T1 Phase:
At this stage, vin is ov, vgs
T1 T2 Phase:
This stage is in the VGS
T2 T3 Phase:
At this point, the diode is cut off, but since the diode has a reverse recovery current, you can see that the id will rush very high, drop back, and then slowly rise. Because of this, the ideal state is to enter the Miller platform in and out of the VA, but because the diode recovers the voltage, the VGS has been fluctuating, until the ID will fall back and then slowly rise. But at this time, the VGS still changes, so the ID is changed, so there will be UL, so the VDS will decrease. This phase ends when VDS drops to VGS-VTH, where the current does not reach the maximum IL we want due to the inductance.
T3 T4 and post-T4 phases:
At this point, VDS
In view of the Maitreya effect, make a summary. The reason for the parameter Miller effect, which is also the Miller platform, is that in the constant current region, a small change in VGS will also bring about a relatively large change in ID, while the increase in ID, the decrease in VDS, the decrease in DVGD, the increase in IGD, the decrease in IGS, and the change in VGS are very small, so it seems that there is a platform. Instead of the VGS not changing because the ID does not change, it is impossible for the ID to remain the same unless the load is a true constant current source.