WEBVTT 00:00:00.000 --> 00:00:01.479 align:middle line:90% 00:00:01.479 --> 00:00:02.465 align:middle line:90% [SQUEAKING] 00:00:02.465 --> 00:00:04.437 align:middle line:90% [RUSTLING] 00:00:04.437 --> 00:00:05.916 align:middle line:90% [CLICKING] 00:00:05.916 --> 00:00:09.670 align:middle line:90% 00:00:09.670 --> 00:00:11.130 align:middle line:84% XIN ZAN: So I think we can start. 00:00:11.130 --> 00:00:12.920 align:middle line:90% So welcome to my lecture. 00:00:12.920 --> 00:00:14.920 align:middle line:84% Today, we are going to learn about the switching 00:00:14.920 --> 00:00:16.219 align:middle line:90% loss and snubbers. 00:00:16.219 --> 00:00:18.760 align:middle line:84% So before that, I would like to have a quick review 00:00:18.760 --> 00:00:20.540 align:middle line:90% about our devices. 00:00:20.540 --> 00:00:22.210 align:middle line:84% So one of the thing power electronics 00:00:22.210 --> 00:00:24.250 align:middle line:90% deal with is about the circuits. 00:00:24.250 --> 00:00:27.670 align:middle line:84% And nowadays, these circuits play with the transistors. 00:00:27.670 --> 00:00:30.940 align:middle line:84% And the transistors can be a bipolar transistor or the field 00:00:30.940 --> 00:00:32.299 align:middle line:90% effect transistors. 00:00:32.299 --> 00:00:35.890 align:middle line:84% So we use transistors as switches, which is on and off. 00:00:35.890 --> 00:00:39.400 align:middle line:84% And other analog circuits, or advanced analog circuits, they 00:00:39.400 --> 00:00:41.920 align:middle line:84% use transistors as a current source. 00:00:41.920 --> 00:00:43.840 align:middle line:84% And for the switches, we are focusing 00:00:43.840 --> 00:00:47.630 align:middle line:84% on there could be ideal switches or practical switches. 00:00:47.630 --> 00:00:51.430 align:middle line:84% For ideal switches, we assume the switches 00:00:51.430 --> 00:00:53.980 align:middle line:84% will instantaneously turn on and turn off. 00:00:53.980 --> 00:00:57.970 align:middle line:84% And also, during the switch-on state, there is no voltage drop, 00:00:57.970 --> 00:00:59.770 align:middle line:90% or the resistor is 0. 00:00:59.770 --> 00:01:03.400 align:middle line:84% And it can carry both positive and negative current. 00:01:03.400 --> 00:01:06.160 align:middle line:84% And when the switch is off, we said 00:01:06.160 --> 00:01:08.230 align:middle line:84% there will be an infinite resistor, 00:01:08.230 --> 00:01:10.280 align:middle line:90% and there is no leakage current. 00:01:10.280 --> 00:01:14.630 align:middle line:84% And it can also block positive voltage and negative voltage. 00:01:14.630 --> 00:01:17.750 align:middle line:84% And also, for parasitics, we assume the ideal switch 00:01:17.750 --> 00:01:19.190 align:middle line:90% has no parasitics. 00:01:19.190 --> 00:01:21.350 align:middle line:84% And because of these characteristics, 00:01:21.350 --> 00:01:25.130 align:middle line:84% we said that there is no loss if we use the ideal switches. 00:01:25.130 --> 00:01:29.150 align:middle line:84% However, for reality, mostly, we use practical switches. 00:01:29.150 --> 00:01:30.860 align:middle line:84% So I list some practical switches 00:01:30.860 --> 00:01:36.140 align:middle line:84% in the lecture notes, that is like BJT, IGBT, thyristor, which 00:01:36.140 --> 00:01:39.260 align:middle line:90% is also called CSR or a diode. 00:01:39.260 --> 00:01:42.590 align:middle line:84% These days, we use mostly about the MOSFET or the wide bandgap 00:01:42.590 --> 00:01:44.060 align:middle line:90% gallium nitride FET. 00:01:44.060 --> 00:01:45.980 align:middle line:84% And for these practical switches, 00:01:45.980 --> 00:01:49.290 align:middle line:84% they need some finite time to turn on and turn off. 00:01:49.290 --> 00:01:51.890 align:middle line:84% And also, when the switch is during its on state, 00:01:51.890 --> 00:01:56.210 align:middle line:84% there is a finite voltage drop, or a finite small resistor. 00:01:56.210 --> 00:01:59.580 align:middle line:84% And the current it can carry has some polarity. 00:01:59.580 --> 00:02:01.730 align:middle line:84% So let's say, for the BJT, it can only 00:02:01.730 --> 00:02:03.230 align:middle line:90% carry the positive current. 00:02:03.230 --> 00:02:06.200 align:middle line:84% But for most FET, it can carry both positive and negative 00:02:06.200 --> 00:02:07.080 align:middle line:90% current. 00:02:07.080 --> 00:02:09.259 align:middle line:84% And also, when the switch is off, 00:02:09.259 --> 00:02:12.020 align:middle line:84% the off resistor will be very big. 00:02:12.020 --> 00:02:14.180 align:middle line:84% So there could be some leakage current. 00:02:14.180 --> 00:02:17.770 align:middle line:84% And also, the blocking voltage also has a polarity. 00:02:17.770 --> 00:02:20.740 align:middle line:84% And because of this, and also for parasitics, 00:02:20.740 --> 00:02:22.540 align:middle line:84% we usually have a parallel capacitance 00:02:22.540 --> 00:02:23.890 align:middle line:90% in parallel with the switch. 00:02:23.890 --> 00:02:27.040 align:middle line:84% And also, some switches will have a antiparallel diode 00:02:27.040 --> 00:02:28.150 align:middle line:90% inside of them. 00:02:28.150 --> 00:02:30.700 align:middle line:84% And because of these non-ideality factors, 00:02:30.700 --> 00:02:33.250 align:middle line:84% there will be some non-zero losses. 00:02:33.250 --> 00:02:35.200 align:middle line:84% And the losses could be conduction loss 00:02:35.200 --> 00:02:38.710 align:middle line:84% during the on state or the switching loss when the switch 00:02:38.710 --> 00:02:40.100 align:middle line:90% turn on and turn off. 00:02:40.100 --> 00:02:43.010 align:middle line:84% And also, we could have some gating loss and some other loss. 00:02:43.010 --> 00:02:45.730 align:middle line:84% So this is a quick review about our transistor. 00:02:45.730 --> 00:02:48.400 align:middle line:84% And today, we will focusing on the switching 00:02:48.400 --> 00:02:51.100 align:middle line:84% loss and the corresponding method 00:02:51.100 --> 00:02:55.960 align:middle line:84% to improve the switching loss that we call snubbers. 00:02:55.960 --> 00:02:58.630 align:middle line:84% So let's take the most basic circuit block 00:02:58.630 --> 00:03:00.850 align:middle line:90% converter as an example. 00:03:00.850 --> 00:03:03.700 align:middle line:84% So we have a input voltage source. 00:03:03.700 --> 00:03:07.630 align:middle line:84% And we can have a practical switch that is a MOSFET. 00:03:07.630 --> 00:03:11.100 align:middle line:90% And we can have a diode here. 00:03:11.100 --> 00:03:15.480 align:middle line:84% And usually, we will have a low-pass filter, L and a C, 00:03:15.480 --> 00:03:18.630 align:middle line:90% and also a resistor here. 00:03:18.630 --> 00:03:22.230 align:middle line:84% So if we assume that the current flow through this inductor 00:03:22.230 --> 00:03:25.260 align:middle line:84% is constant, which is there is no ripple here, 00:03:25.260 --> 00:03:28.170 align:middle line:84% and this inductor is pretty big, then we 00:03:28.170 --> 00:03:32.370 align:middle line:84% can model our circuit, as we still have an input voltage 00:03:32.370 --> 00:03:38.580 align:middle line:84% source, and we have a MOSFET, and we have a diode here. 00:03:38.580 --> 00:03:43.260 align:middle line:84% And now our load will be a constant current source. 00:03:43.260 --> 00:03:47.130 align:middle line:84% So we already know the switching pattern of the buck converter. 00:03:47.130 --> 00:03:48.150 align:middle line:90% So let's see. 00:03:48.150 --> 00:03:54.060 align:middle line:84% For a given switching period that is from 0 to Ts, from 0 00:03:54.060 --> 00:03:57.900 align:middle line:84% to DTs, which is a duty cycle, the active switch, 00:03:57.900 --> 00:04:00.990 align:middle line:84% or the MOSFET, will turn on this load current. 00:04:00.990 --> 00:04:05.020 align:middle line:84% So the switch current will be a square wave like this. 00:04:05.020 --> 00:04:08.510 align:middle line:90% So this will be isw. 00:04:08.510 --> 00:04:12.475 align:middle line:84% And for the diode, when the switch is off during the DTs 00:04:12.475 --> 00:04:15.770 align:middle line:84% to Ts, this diode will flow through this load current. 00:04:15.770 --> 00:04:19.250 align:middle line:84% So that will be a current waveform like this. 00:04:19.250 --> 00:04:21.829 align:middle line:90% And this is the diode current. 00:04:21.829 --> 00:04:25.160 align:middle line:84% So when we considering the conduction loss, 00:04:25.160 --> 00:04:26.400 align:middle line:90% what is the power? 00:04:26.400 --> 00:04:27.810 align:middle line:84% Conduction loss is still a power. 00:04:27.810 --> 00:04:28.680 align:middle line:90% So what is power? 00:04:28.680 --> 00:04:30.650 align:middle line:90% Power is v times i. 00:04:30.650 --> 00:04:34.040 align:middle line:84% So for MOSFET, the conduction loss 00:04:34.040 --> 00:04:40.760 align:middle line:84% will be something like i rms squared 00:04:40.760 --> 00:04:44.840 align:middle line:84% times the on-state resistance, which I said before, 00:04:44.840 --> 00:04:47.150 align:middle line:84% for a switch, for a practical switch, when it's 00:04:47.150 --> 00:04:49.100 align:middle line:84% during that on state, either it can be 00:04:49.100 --> 00:04:50.820 align:middle line:90% a resistor or a voltage drop. 00:04:50.820 --> 00:04:55.920 align:middle line:84% And for a MOSFET, it will be a Rds on resistor. 00:04:55.920 --> 00:04:59.490 align:middle line:84% And we can calculate the rms current for the MOSFET. 00:04:59.490 --> 00:05:03.420 align:middle line:84% So that will be the duty cycle times the IL squared 00:05:03.420 --> 00:05:04.840 align:middle line:90% times the Rds on. 00:05:04.840 --> 00:05:07.930 align:middle line:90% 00:05:07.930 --> 00:05:12.000 align:middle line:84% So this will be the IL for its amplitude. 00:05:12.000 --> 00:05:16.280 align:middle line:84% And a photodiode, its conduction loss 00:05:16.280 --> 00:05:19.580 align:middle line:84% will be the average current of the diode 00:05:19.580 --> 00:05:21.500 align:middle line:90% times the voltage drop. 00:05:21.500 --> 00:05:24.210 align:middle line:90% So still, the loss is a power. 00:05:24.210 --> 00:05:27.200 align:middle line:84% Power is v times i, just like, for the MOSFET, 00:05:27.200 --> 00:05:29.270 align:middle line:90% the v will be i times the r. 00:05:29.270 --> 00:05:33.410 align:middle line:84% So it will be the irms squared times the Rds on. 00:05:33.410 --> 00:05:35.870 align:middle line:84% But for the diode, it will be the average current 00:05:35.870 --> 00:05:38.440 align:middle line:90% times the voltage drop. 00:05:38.440 --> 00:05:44.860 align:middle line:84% So that will be the 1 minus D times IL times the voltage drop 00:05:44.860 --> 00:05:46.670 align:middle line:90% of the diode. 00:05:46.670 --> 00:05:49.430 align:middle line:84% So I also list, during its on state, 00:05:49.430 --> 00:05:51.470 align:middle line:84% the MOSFET looks like the resistor. 00:05:51.470 --> 00:05:55.040 align:middle line:84% And during its on state, a diode looks like a voltage drop. 00:05:55.040 --> 00:05:59.960 align:middle line:84% And for the BJT or IGBT or the SCR during its on state, 00:05:59.960 --> 00:06:01.920 align:middle line:84% it still looks like a voltage drop. 00:06:01.920 --> 00:06:05.270 align:middle line:84% So when you calculate the conduction loss for a BJT, IGBT, 00:06:05.270 --> 00:06:08.630 align:middle line:84% or a diode or a thyristor, you use the average current 00:06:08.630 --> 00:06:09.890 align:middle line:90% times the voltage drop. 00:06:09.890 --> 00:06:11.780 align:middle line:84% And when you calculate the conduction loss 00:06:11.780 --> 00:06:13.910 align:middle line:84% for MOSFET or gallium nitride FET, 00:06:13.910 --> 00:06:16.880 align:middle line:84% you use the rms squared times the Rds on. 00:06:16.880 --> 00:06:20.980 align:middle line:84% So this is about the conduction loss. 00:06:20.980 --> 00:06:23.770 align:middle line:84% And also, today, the semiconductor loss 00:06:23.770 --> 00:06:25.280 align:middle line:90% is back-of-envelope. 00:06:25.280 --> 00:06:28.540 align:middle line:84% So if you do more research, we can model this conduction loss 00:06:28.540 --> 00:06:29.660 align:middle line:90% more precisely. 00:06:29.660 --> 00:06:31.960 align:middle line:84% But generally speaking, this can be 00:06:31.960 --> 00:06:34.360 align:middle line:84% a rule of thumb for a rough calculation, 00:06:34.360 --> 00:06:38.560 align:middle line:84% and it is enough good for our design. 00:06:38.560 --> 00:06:42.610 align:middle line:84% And for the conduction loss and for the switching loss, 00:06:42.610 --> 00:06:44.960 align:middle line:84% it occurs during the turn-on and the turn-off. 00:06:44.960 --> 00:06:48.100 align:middle line:84% And we need to imagine what will be the waveform for the switch 00:06:48.100 --> 00:06:51.020 align:middle line:84% current and the switch voltage during the turn-on and turn-off. 00:06:51.020 --> 00:06:52.930 align:middle line:90% Here, we have assumption. 00:06:52.930 --> 00:06:59.830 align:middle line:84% Our assumption is we assume the vsw and the isw 00:06:59.830 --> 00:07:03.430 align:middle line:84% will be linear during the transient. 00:07:03.430 --> 00:07:08.200 align:middle line:84% And here, we also assume our diode is ideal. 00:07:08.200 --> 00:07:10.400 align:middle line:84% So what do I mean by an ideal diode? 00:07:10.400 --> 00:07:13.990 align:middle line:84% It is from the characteristic l-V curve. 00:07:13.990 --> 00:07:15.400 align:middle line:90% This is vd. 00:07:15.400 --> 00:07:16.780 align:middle line:90% This is id. 00:07:16.780 --> 00:07:19.930 align:middle line:84% So a diode can conduct any amount of current 00:07:19.930 --> 00:07:22.840 align:middle line:84% during its on state with a zero-voltage drop. 00:07:22.840 --> 00:07:25.930 align:middle line:84% And the diode can block any amount of the voltage 00:07:25.930 --> 00:07:27.240 align:middle line:90% during its off state. 00:07:27.240 --> 00:07:30.050 align:middle line:90% So this is the ideal diode. 00:07:30.050 --> 00:07:34.940 align:middle line:84% And if we notice, this term reference as this is vsw. 00:07:34.940 --> 00:07:37.450 align:middle line:90% This is isw. 00:07:37.450 --> 00:07:40.050 align:middle line:90% And this is id. 00:07:40.050 --> 00:07:42.150 align:middle line:90% And this is vd. 00:07:42.150 --> 00:07:46.770 align:middle line:84% We can simply have the KCL and KVL equation for this. 00:07:46.770 --> 00:07:49.710 align:middle line:84% The KCL equation at this node will be 00:07:49.710 --> 00:07:55.280 align:middle line:90% IL is equal to isw plus the id. 00:07:55.280 --> 00:07:57.770 align:middle line:84% And for the KVL equation, that will be the v 00:07:57.770 --> 00:08:04.640 align:middle line:84% in is equal to the vsw minus the vd. 00:08:04.640 --> 00:08:07.505 align:middle line:84% So let's imagine what will be the current and voltage 00:08:07.505 --> 00:08:08.005 align:middle line:90% waveform. 00:08:08.005 --> 00:08:12.160 align:middle line:90% 00:08:12.160 --> 00:08:14.830 align:middle line:84% So for turning off, I would first 00:08:14.830 --> 00:08:18.450 align:middle line:84% say that it will be the voltage increase first. 00:08:18.450 --> 00:08:21.660 align:middle line:84% And when the voltage increase to its maximum amplitude 00:08:21.660 --> 00:08:24.650 align:middle line:84% for the buck converter, it's v in. 00:08:24.650 --> 00:08:28.570 align:middle line:90% And then the current will drop. 00:08:28.570 --> 00:08:29.890 align:middle line:90% Why it's like this? 00:08:29.890 --> 00:08:33.429 align:middle line:84% Because from this KCL and KVL equation, 00:08:33.429 --> 00:08:37.630 align:middle line:84% the diode cannot turn on or carry current until it has 00:08:37.630 --> 00:08:39.159 align:middle line:90% the zero-voltage drop. 00:08:39.159 --> 00:08:41.320 align:middle line:84% So to have the zero-voltage drop, 00:08:41.320 --> 00:08:44.470 align:middle line:84% you need to have the switch voltage 00:08:44.470 --> 00:08:47.890 align:middle line:90% vsw to increase to its maximum. 00:08:47.890 --> 00:08:50.920 align:middle line:84% And when it's increased to its maximum, which is v in, 00:08:50.920 --> 00:08:52.930 align:middle line:90% then the vd can become 0. 00:08:52.930 --> 00:08:55.510 align:middle line:84% And later, the diode can conduct the current. 00:08:55.510 --> 00:08:58.180 align:middle line:84% And then, if the diode conducts the current, 00:08:58.180 --> 00:09:00.710 align:middle line:90% then the isw will decrease. 00:09:00.710 --> 00:09:02.830 align:middle line:90% So that is during the turn-off. 00:09:02.830 --> 00:09:06.310 align:middle line:84% The switch voltage increase first, 00:09:06.310 --> 00:09:10.730 align:middle line:84% and then the switch current decrease later. 00:09:10.730 --> 00:09:12.950 align:middle line:84% And similarly, during the turn-on, 00:09:12.950 --> 00:09:17.540 align:middle line:84% we can also see that it will be the switch current increase 00:09:17.540 --> 00:09:20.330 align:middle line:84% first and when the switch current go 00:09:20.330 --> 00:09:24.250 align:middle line:90% to its maximum, which is IL. 00:09:24.250 --> 00:09:28.020 align:middle line:84% And then the switch voltage will decrease. 00:09:28.020 --> 00:09:29.250 align:middle line:90% And why is that? 00:09:29.250 --> 00:09:32.200 align:middle line:84% It is also because of the characteristics of the style 00:09:32.200 --> 00:09:35.520 align:middle line:84% because the diode cannot be turned off or block voltage 00:09:35.520 --> 00:09:37.740 align:middle line:90% until it has 0 current. 00:09:37.740 --> 00:09:40.380 align:middle line:84% And to have the diode to have the 0 current, 00:09:40.380 --> 00:09:44.280 align:middle line:84% it means the isw should flow through all the load 00:09:44.280 --> 00:09:48.280 align:middle line:84% current, which means the load current should increase first. 00:09:48.280 --> 00:09:52.350 align:middle line:84% And at this time, the current in the diode will become 0, 00:09:52.350 --> 00:09:55.180 align:middle line:84% and the diode can be turned off and block the voltage. 00:09:55.180 --> 00:09:59.030 align:middle line:84% And once the diode blocks the voltage, the vsw can decrease. 00:09:59.030 --> 00:10:06.360 align:middle line:90% 00:10:06.360 --> 00:10:11.970 align:middle line:84% So this is switching waveforms for the switch 00:10:11.970 --> 00:10:13.740 align:middle line:84% during the turn-on and the turn-off. 00:10:13.740 --> 00:10:18.570 align:middle line:84% And usually, we notate this time interval as the t fall 00:10:18.570 --> 00:10:20.370 align:middle line:90% during this turn-off. 00:10:20.370 --> 00:10:24.480 align:middle line:84% And we have this time interval as t rise during its turn-on. 00:10:24.480 --> 00:10:28.220 align:middle line:90% 00:10:28.220 --> 00:10:32.550 align:middle line:84% For ideal switch, because this will be 0 and this will be 0, 00:10:32.550 --> 00:10:35.250 align:middle line:84% the current and the voltage will have no overlap. 00:10:35.250 --> 00:10:38.340 align:middle line:84% So the ideal switch, the switching loss will be 0. 00:10:38.340 --> 00:10:40.040 align:middle line:84% But for a practical switch, because we 00:10:40.040 --> 00:10:44.090 align:middle line:84% have the tf and the tr, the switch current and the switch 00:10:44.090 --> 00:10:46.220 align:middle line:90% voltage will have some overlap. 00:10:46.220 --> 00:10:48.650 align:middle line:84% So we will have some switching loss. 00:10:48.650 --> 00:10:49.890 align:middle line:90% And what is loss? 00:10:49.890 --> 00:10:51.420 align:middle line:90% Loss is a kind of power. 00:10:51.420 --> 00:10:55.640 align:middle line:84% So that will still be the voltage times the current. 00:10:55.640 --> 00:11:01.100 align:middle line:84% So if we draw the power loss curve, 00:11:01.100 --> 00:11:07.950 align:middle line:84% that will be the P dissipate equals the vsw times the isw. 00:11:07.950 --> 00:11:11.340 align:middle line:84% So during the time interval beside the tr and the tf, 00:11:11.340 --> 00:11:15.180 align:middle line:84% it will become 0, or it will become a pretty small number. 00:11:15.180 --> 00:11:18.346 align:middle line:84% And during this, the loss will be triangular. 00:11:18.346 --> 00:11:21.780 align:middle line:90% 00:11:21.780 --> 00:11:25.720 align:middle line:84% And a similar follows tr, follows the turn-on. 00:11:25.720 --> 00:11:27.670 align:middle line:90% It will also be triangular. 00:11:27.670 --> 00:11:30.450 align:middle line:84% It is just simply the switch voltage 00:11:30.450 --> 00:11:32.160 align:middle line:90% times the switch current. 00:11:32.160 --> 00:11:35.610 align:middle line:84% And we will find this will be the switching loss. 00:11:35.610 --> 00:11:40.020 align:middle line:84% And this part of area, which is the power integrated 00:11:40.020 --> 00:11:42.130 align:middle line:90% by the time, will be the energy. 00:11:42.130 --> 00:11:45.480 align:middle line:84% So this area will be the turn-off energy-- 00:11:45.480 --> 00:11:48.280 align:middle line:90% let's say E off. 00:11:48.280 --> 00:11:51.370 align:middle line:84% And this area will be the turn-on energy-- 00:11:51.370 --> 00:11:54.360 align:middle line:90% let's say the e on. 00:11:54.360 --> 00:11:58.110 align:middle line:84% So that is if we switch this device in one period-- we 00:11:58.110 --> 00:12:02.460 align:middle line:84% switch off and we switch on-- we will lose part of this energy. 00:12:02.460 --> 00:12:05.460 align:middle line:84% And as we said, the switching our power converters, 00:12:05.460 --> 00:12:06.840 align:middle line:90% they switch frequently. 00:12:06.840 --> 00:12:09.190 align:middle line:84% There is a switching frequency definition. 00:12:09.190 --> 00:12:13.580 align:middle line:84% So then when we think about the switching loss, 00:12:13.580 --> 00:12:15.920 align:middle line:84% that will be this triangular area, 00:12:15.920 --> 00:12:23.270 align:middle line:84% which will be 1/2 tf times the V in times the IL, 00:12:23.270 --> 00:12:27.620 align:middle line:84% which is the turn-off energy plus the turn-on energy, 00:12:27.620 --> 00:12:33.500 align:middle line:84% which is the time times the peak value V in times the IL 00:12:33.500 --> 00:12:36.770 align:middle line:84% and times by the switching frequency. 00:12:36.770 --> 00:12:39.490 align:middle line:90% That will be our switching loss. 00:12:39.490 --> 00:12:41.710 align:middle line:84% And you can also consider with times 00:12:41.710 --> 00:12:44.020 align:middle line:84% the switching frequency is equivalent to divided 00:12:44.020 --> 00:12:45.490 align:middle line:90% by the switching period. 00:12:45.490 --> 00:12:47.560 align:middle line:84% And if you divide it by the switching period, 00:12:47.560 --> 00:12:49.730 align:middle line:84% that will be an average power of it. 00:12:49.730 --> 00:12:51.490 align:middle line:84% So it's still the same definition, 00:12:51.490 --> 00:12:55.040 align:middle line:84% but you can look it in two aspect. 00:12:55.040 --> 00:12:57.000 align:middle line:84% And also, for the switching loss, 00:12:57.000 --> 00:13:00.230 align:middle line:84% this is killing us in our current converters. 00:13:00.230 --> 00:13:02.810 align:middle line:84% So to decrease the switching loss, 00:13:02.810 --> 00:13:05.390 align:middle line:84% we can decrease the switching frequency. 00:13:05.390 --> 00:13:07.700 align:middle line:84% However, if we're decreasing the switching frequency, 00:13:07.700 --> 00:13:10.820 align:middle line:84% our passive components, like the inductor or the capacitor, 00:13:10.820 --> 00:13:14.330 align:middle line:84% will be very big, which limits our power density and our valve 00:13:14.330 --> 00:13:15.870 align:middle line:90% volume of our converter. 00:13:15.870 --> 00:13:18.740 align:middle line:84% So we usually want to increase our switching frequency. 00:13:18.740 --> 00:13:22.430 align:middle line:84% But the switching loss will limit us, how large 00:13:22.430 --> 00:13:24.350 align:middle line:90% or how fast can we switch. 00:13:24.350 --> 00:13:28.010 align:middle line:84% So another thing to reduce the switching frequency 00:13:28.010 --> 00:13:31.940 align:middle line:84% is, Can we decrease the tf or the tr, 00:13:31.940 --> 00:13:37.260 align:middle line:84% or we decrease the turn-off energy or the turn-on energy? 00:13:37.260 --> 00:13:40.980 align:middle line:90% So that comes with our snubber. 00:13:40.980 --> 00:13:45.060 align:middle line:84% So before that, let's have a demo of the switching loss. 00:13:45.060 --> 00:13:47.100 align:middle line:90% So Rafa will do a demo. 00:13:47.100 --> 00:13:50.050 align:middle line:84% And before we start the demo, do you have any questions? 00:13:50.050 --> 00:13:52.660 align:middle line:90% 00:13:52.660 --> 00:13:54.580 align:middle line:84% RAFA ISLAM: So I have a circuit here 00:13:54.580 --> 00:13:58.660 align:middle line:84% that's designed to help us look at the voltage and the current 00:13:58.660 --> 00:14:01.900 align:middle line:84% of a MOSFET when we're turning it on or off. 00:14:01.900 --> 00:14:07.580 align:middle line:84% So here, you can see two transitions, the purple or pink. 00:14:07.580 --> 00:14:10.180 align:middle line:84% The pink waveform is the voltage waveform. 00:14:10.180 --> 00:14:12.820 align:middle line:84% And the yellow waveform is the current waveform. 00:14:12.820 --> 00:14:17.350 align:middle line:84% So on the left, the yellow waveform rises, 00:14:17.350 --> 00:14:19.990 align:middle line:84% and the voltage is falling, which 00:14:19.990 --> 00:14:22.240 align:middle line:84% means there's no voltage across the switch, 00:14:22.240 --> 00:14:24.170 align:middle line:90% and the switch turns on. 00:14:24.170 --> 00:14:29.050 align:middle line:84% On the right, the current, the yellow waveform, is falling, 00:14:29.050 --> 00:14:32.740 align:middle line:84% and the pink waveform is rising again, 00:14:32.740 --> 00:14:35.740 align:middle line:84% which means the switch is turning off. 00:14:35.740 --> 00:14:40.790 align:middle line:84% Now let me zoom on in one transition. 00:14:40.790 --> 00:14:46.340 align:middle line:90% 00:14:46.340 --> 00:14:51.130 align:middle line:84% Let me zoom in on the turn-on transition. 00:14:51.130 --> 00:14:53.390 align:middle line:84% So as you can notice, the yellow waveform 00:14:53.390 --> 00:14:55.020 align:middle line:90% is rising pretty quickly. 00:14:55.020 --> 00:14:58.040 align:middle line:84% But the pink one, which is the voltage waveform, 00:14:58.040 --> 00:15:00.260 align:middle line:84% takes some time to completely hit 00:15:00.260 --> 00:15:04.460 align:middle line:84% 0, which is the rise time and fall time Xin was talking about. 00:15:04.460 --> 00:15:06.460 align:middle line:90% And what will this do? 00:15:06.460 --> 00:15:08.270 align:middle line:84% This means that there is a overlap time 00:15:08.270 --> 00:15:11.690 align:middle line:84% between the pink and the yellow waveform crossing. 00:15:11.690 --> 00:15:15.630 align:middle line:84% And that's the overlap time creating the overlap loss. 00:15:15.630 --> 00:15:18.410 align:middle line:90% Now there's another loss here. 00:15:18.410 --> 00:15:21.890 align:middle line:84% Notice when that switch is turning on, 00:15:21.890 --> 00:15:24.800 align:middle line:84% the voltage is non-zero, which means 00:15:24.800 --> 00:15:28.970 align:middle line:84% if the switch's capacitance is storing energy in it. 00:15:28.970 --> 00:15:34.740 align:middle line:84% Now, let me also show you the power loss. 00:15:34.740 --> 00:15:36.690 align:middle line:84% XIN ZAN: So you can see, the yellow curve 00:15:36.690 --> 00:15:39.060 align:middle line:84% will be the current, and the purple curve 00:15:39.060 --> 00:15:40.240 align:middle line:90% will be the voltage. 00:15:40.240 --> 00:15:42.240 align:middle line:84% And you can see that, for hard switching, 00:15:42.240 --> 00:15:45.000 align:middle line:84% it is a voltage increase first, and then the current 00:15:45.000 --> 00:15:47.910 align:middle line:84% will decrease, which match with our theory. 00:15:47.910 --> 00:15:51.120 align:middle line:84% And you can also observe that, for the yellow curve, which 00:15:51.120 --> 00:15:52.470 align:middle line:90% is the voltage-- 00:15:52.470 --> 00:15:55.150 align:middle line:84% sorry, the purple curve, which is the voltage. 00:15:55.150 --> 00:15:57.700 align:middle line:84% And during its on state, it has a very small value. 00:15:57.700 --> 00:15:59.830 align:middle line:84% So that will be on-state voltage drop. 00:15:59.830 --> 00:16:03.402 align:middle line:84% It could be the rms current times the Rds up. 00:16:03.402 --> 00:16:04.110 align:middle line:90% RAFA ISLAM: Yeah. 00:16:04.110 --> 00:16:08.160 align:middle line:84% It's really hard to see, but the purplish waveform, the zigzag 00:16:08.160 --> 00:16:10.030 align:middle line:90% waveform, is the power loss. 00:16:10.030 --> 00:16:14.490 align:middle line:84% You can also see that, in here, in the off state, 00:16:14.490 --> 00:16:21.990 align:middle line:84% the line is almost to 0 because there is no voltage-- 00:16:21.990 --> 00:16:24.340 align:middle line:84% because there is no current in the switch. 00:16:24.340 --> 00:16:28.230 align:middle line:84% In the other hand, whenever you are seeing all those ripples, 00:16:28.230 --> 00:16:33.220 align:middle line:84% that's because when the switch is on, it acts like a resistor. 00:16:33.220 --> 00:16:37.080 align:middle line:84% So it has some Rds on and non-zero current, 00:16:37.080 --> 00:16:40.740 align:middle line:84% i square r, which makes the loss. 00:16:40.740 --> 00:16:41.730 align:middle line:90% XIN ZAN: Yeah. 00:16:41.730 --> 00:16:44.550 align:middle line:84% The math shows the time of the multiple 00:16:44.550 --> 00:16:46.170 align:middle line:90% of the voltage and the current. 00:16:46.170 --> 00:16:48.732 align:middle line:84% And you can observe there is a triangular of that. 00:16:48.732 --> 00:16:49.440 align:middle line:90% RAFA ISLAM: Yeah. 00:16:49.440 --> 00:16:52.920 align:middle line:84% So we turn on and off our switch how many times-- 00:16:52.920 --> 00:16:56.400 align:middle line:84% 10 million times or 100,000 times per second. 00:16:56.400 --> 00:16:59.130 align:middle line:84% It can seem trivial for one time. 00:16:59.130 --> 00:17:01.350 align:middle line:84% But if we keep doing it this many times, 00:17:01.350 --> 00:17:05.180 align:middle line:90% we cannot neglect this anymore. 00:17:05.180 --> 00:17:06.569 align:middle line:90% XIN ZAN: Thank you. 00:17:06.569 --> 00:17:08.369 align:middle line:90% So another thing I want to-- 00:17:08.369 --> 00:17:09.930 align:middle line:90% I'm not sure if you observed. 00:17:09.930 --> 00:17:14.140 align:middle line:84% So during the switch turn-off, we do observe the voltage, 00:17:14.140 --> 00:17:17.089 align:middle line:90% it have some resonance here. 00:17:17.089 --> 00:17:19.880 align:middle line:84% So this is because, as I said, we want 00:17:19.880 --> 00:17:22.369 align:middle line:90% to decrease the tr and the tf. 00:17:22.369 --> 00:17:24.440 align:middle line:90% But we have some limitations. 00:17:24.440 --> 00:17:28.820 align:middle line:84% So the first limitations will be the parasitics. 00:17:28.820 --> 00:17:31.730 align:middle line:84% So the parasitics is some unwanted parameters inside 00:17:31.730 --> 00:17:32.830 align:middle line:90% of the circuit. 00:17:32.830 --> 00:17:36.440 align:middle line:84% So for example, the parasitics could come from-- 00:17:36.440 --> 00:17:40.510 align:middle line:84% there may be a parallel capacitance 00:17:40.510 --> 00:17:42.820 align:middle line:90% in parallel with the switch. 00:17:42.820 --> 00:17:46.060 align:middle line:84% And also, there could be some stray inductance. 00:17:46.060 --> 00:17:47.830 align:middle line:84% And when the switch is turned off, 00:17:47.830 --> 00:17:50.200 align:middle line:84% this parallel capacitance can resonate 00:17:50.200 --> 00:17:52.210 align:middle line:90% with this stray inductance. 00:17:52.210 --> 00:17:55.690 align:middle line:84% And that will cause an overshoot during the switch turn-off. 00:17:55.690 --> 00:17:58.420 align:middle line:84% And this overshoot can kill the switch. 00:17:58.420 --> 00:18:02.170 align:middle line:84% So this is one thing like we have the limitation for the tr 00:18:02.170 --> 00:18:03.490 align:middle line:90% and tf. 00:18:03.490 --> 00:18:09.520 align:middle line:84% And the second is about during the transient, 00:18:09.520 --> 00:18:12.460 align:middle line:84% if we have pretty low, we have pretty small, tf 00:18:12.460 --> 00:18:16.430 align:middle line:84% and the tr, our di/dt or the dv/dt could be very large. 00:18:16.430 --> 00:18:19.000 align:middle line:84% So what do I mean by the di/dt and dv/dt? 00:18:19.000 --> 00:18:22.120 align:middle line:84% So that will be the slope for the rising 00:18:22.120 --> 00:18:25.930 align:middle line:84% voltage or the falling current or the slope for the rising 00:18:25.930 --> 00:18:28.000 align:middle line:90% current and the falling voltage. 00:18:28.000 --> 00:18:30.580 align:middle line:84% And this di/dt and the dv/dt, they 00:18:30.580 --> 00:18:33.830 align:middle line:84% can have some coupling inductor or coupling capacitor. 00:18:33.830 --> 00:18:37.300 align:middle line:84% So imagine if we have a coupling inductor, which is L, 00:18:37.300 --> 00:18:40.250 align:middle line:84% this L di/dt will cause some voltage. 00:18:40.250 --> 00:18:42.950 align:middle line:84% And if we have a coupling capacitance, which is C 00:18:42.950 --> 00:18:45.870 align:middle line:84% this C dv/dt will cause some current. 00:18:45.870 --> 00:18:49.670 align:middle line:84% So previously, if we have our di/dt or dv/dt pretty small 00:18:49.670 --> 00:18:51.410 align:middle line:84% and our parasitic is pretty small, 00:18:51.410 --> 00:18:54.630 align:middle line:84% these things will not match our control circuit. 00:18:54.630 --> 00:18:56.660 align:middle line:84% But either if we have a large inductance 00:18:56.660 --> 00:18:59.450 align:middle line:84% or a large capacitance together with a large di/dt 00:18:59.450 --> 00:19:04.110 align:middle line:84% or dv/dt, this voltage or current, they are unexpected. 00:19:04.110 --> 00:19:07.110 align:middle line:84% They can cause some errors for our control circuit. 00:19:07.110 --> 00:19:10.550 align:middle line:84% So this is the second limitation. 00:19:10.550 --> 00:19:14.100 align:middle line:84% And for the third limitation, for the device itself, 00:19:14.100 --> 00:19:19.630 align:middle line:84% we sometimes have the di/dt or the dv/dt requirement. 00:19:19.630 --> 00:19:23.570 align:middle line:84% So the second one is for the EMI issues. 00:19:23.570 --> 00:19:26.390 align:middle line:84% But for this, this is bad for the device itself. 00:19:26.390 --> 00:19:28.520 align:middle line:84% It has nothing to do with the control circuits. 00:19:28.520 --> 00:19:31.330 align:middle line:84% So for example, if we have a large current rating device-- 00:19:31.330 --> 00:19:38.080 align:middle line:84% let's say the IGBT or the SCR, if the di/dt is pretty large, 00:19:38.080 --> 00:19:41.560 align:middle line:84% in a short time, the current cannot spread for the parallel 00:19:41.560 --> 00:19:43.820 align:middle line:84% diode or the multiple wave bound. 00:19:43.820 --> 00:19:47.420 align:middle line:84% So maybe one diode or maybe one wave bound will blow up. 00:19:47.420 --> 00:19:49.610 align:middle line:90% So that will destroy our device. 00:19:49.610 --> 00:19:52.420 align:middle line:84% And for the dv/dt, if we have a high dv/dt 00:19:52.420 --> 00:19:55.750 align:middle line:84% inside of the thyristor, it can trigger the thyristor 00:19:55.750 --> 00:19:56.600 align:middle line:90% back to on. 00:19:56.600 --> 00:19:58.700 align:middle line:84% So usually, what we want to turn off the device. 00:19:58.700 --> 00:20:01.960 align:middle line:84% But because of this high dv/dt inside of the thyristor, 00:20:01.960 --> 00:20:04.640 align:middle line:84% it can trigger the thyristor back on. 00:20:04.640 --> 00:20:08.200 align:middle line:84% So this is the third limitations for the tr and tf. 00:20:08.200 --> 00:20:11.170 align:middle line:84% And usually, the fourth one, we said every device 00:20:11.170 --> 00:20:13.750 align:middle line:90% has a safe operating area. 00:20:13.750 --> 00:20:18.930 align:middle line:84% So let's draw the vi curve of a device if this is v, this is i. 00:20:18.930 --> 00:20:22.970 align:middle line:84% So for a normal device, we have a maximum current rating. 00:20:22.970 --> 00:20:26.660 align:middle line:84% And we also have a maximum off-voltage rating. 00:20:26.660 --> 00:20:29.270 align:middle line:84% But this is not like a square because we also 00:20:29.270 --> 00:20:34.170 align:middle line:84% have some power limitation or some thermal limitation. 00:20:34.170 --> 00:20:38.270 align:middle line:84% So this is a safe operating area for a device. 00:20:38.270 --> 00:20:43.550 align:middle line:84% And we cannot have our switch voltage or switch current exceed 00:20:43.550 --> 00:20:48.050 align:middle line:84% this safe operating area for steady-state operation 00:20:48.050 --> 00:20:50.480 align:middle line:84% or for an instantaneous operation. 00:20:50.480 --> 00:20:52.730 align:middle line:84% And here, for the snubbers, we care 00:20:52.730 --> 00:20:54.650 align:middle line:84% about the instantaneous operation. 00:20:54.650 --> 00:20:57.170 align:middle line:84% And the next lecture, when we learn about the thermal, 00:20:57.170 --> 00:20:59.150 align:middle line:84% we care about the static operation. 00:20:59.150 --> 00:21:03.870 align:middle line:84% That is, the rms voltage or the rms current cannot exceed. 00:21:03.870 --> 00:21:05.940 align:middle line:84% But here, we care the instantaneous one. 00:21:05.940 --> 00:21:08.450 align:middle line:84% So if we think about this hard switching 00:21:08.450 --> 00:21:11.360 align:middle line:84% during the turn-on and turn-off, let's say 00:21:11.360 --> 00:21:17.430 align:middle line:84% for during the turn-off, the current will hold constant 00:21:17.430 --> 00:21:19.150 align:middle line:90% until the voltage increase. 00:21:19.150 --> 00:21:20.950 align:middle line:90% And then the current will drop. 00:21:20.950 --> 00:21:22.830 align:middle line:84% And if it's hard switching, we can 00:21:22.830 --> 00:21:26.370 align:middle line:84% find that our instantaneous current or voltage 00:21:26.370 --> 00:21:28.750 align:middle line:84% will exceed the safe operating area. 00:21:28.750 --> 00:21:32.050 align:middle line:84% And this is not good, and this can destroy our device. 00:21:32.050 --> 00:21:33.910 align:middle line:84% So that's why we want the snubber. 00:21:33.910 --> 00:21:37.140 align:middle line:84% We want the snubber to manipulate with this switching 00:21:37.140 --> 00:21:39.310 align:middle line:90% locus of the iv curves. 00:21:39.310 --> 00:21:43.620 align:middle line:84% And also, for snubbers, we want to decrease this turn-off energy 00:21:43.620 --> 00:21:46.350 align:middle line:84% or the turn-on energy of the switches, inside 00:21:46.350 --> 00:21:47.290 align:middle line:90% of the switches. 00:21:47.290 --> 00:21:51.440 align:middle line:84% So that's the motivation of our snubbers. 00:21:51.440 --> 00:21:55.985 align:middle line:84% So before we start snubbers, do you have any questions? 00:21:55.985 --> 00:21:56.485 align:middle line:90% Cool. 00:21:56.485 --> 00:22:01.720 align:middle line:90% 00:22:01.720 --> 00:22:05.290 align:middle line:84% So for snubbers, we have two kind of motivation. 00:22:05.290 --> 00:22:08.230 align:middle line:84% One is we want to maintain the instantaneous current 00:22:08.230 --> 00:22:11.490 align:middle line:84% and the voltage of the switch in the safe operating area. 00:22:11.490 --> 00:22:15.540 align:middle line:84% That is, also, we want to have low di/dt or dv/dt. 00:22:15.540 --> 00:22:18.660 align:middle line:90% 00:22:18.660 --> 00:22:20.580 align:middle line:84% And the second motivation will be 00:22:20.580 --> 00:22:23.370 align:middle line:84% we want to decrease the switching loss 00:22:23.370 --> 00:22:27.130 align:middle line:90% inside of the switches. 00:22:27.130 --> 00:22:29.910 align:middle line:84% So please have attention, What do I mean by the inside? 00:22:29.910 --> 00:22:35.510 align:middle line:90% 00:22:35.510 --> 00:22:39.140 align:middle line:84% And now let's think about the turn-off snubber first. 00:22:39.140 --> 00:22:42.680 align:middle line:84% So still, we use a spark converter as our example. 00:22:42.680 --> 00:22:45.350 align:middle line:84% We have the input voltage source. 00:22:45.350 --> 00:22:48.170 align:middle line:90% And we have a MOSFET. 00:22:48.170 --> 00:22:51.380 align:middle line:90% And we have a diode. 00:22:51.380 --> 00:22:55.730 align:middle line:84% And this will be our constant load current source. 00:22:55.730 --> 00:23:00.380 align:middle line:84% So previously, at this terminal, we said the diode cannot turn 00:23:00.380 --> 00:23:05.000 align:middle line:84% on or carry current until its voltage becomes 0. 00:23:05.000 --> 00:23:08.420 align:middle line:84% But now we can have a capacitor that 00:23:08.420 --> 00:23:10.430 align:middle line:90% is in parallel with a switch. 00:23:10.430 --> 00:23:14.270 align:middle line:84% And previously at this node, we will have the KCL equation 00:23:14.270 --> 00:23:22.250 align:middle line:84% as IL is equal to id plus the isw. 00:23:22.250 --> 00:23:25.430 align:middle line:84% But now, we have an extra term, which is the ic. 00:23:25.430 --> 00:23:28.190 align:middle line:84% So previously, the diode cannot conduct the current. 00:23:28.190 --> 00:23:31.010 align:middle line:84% But now the capacitance can flow through the current. 00:23:31.010 --> 00:23:32.810 align:middle line:84% And now we can have instantaneously 00:23:32.810 --> 00:23:35.480 align:middle line:84% the small voltage on a current during the turn-off 00:23:35.480 --> 00:23:36.370 align:middle line:90% at the same time. 00:23:36.370 --> 00:23:39.320 align:middle line:90% 00:23:39.320 --> 00:23:47.360 align:middle line:84% So if we draw the I-v curve for the switch current. 00:23:47.360 --> 00:23:49.370 align:middle line:84% We still see like it will we still 00:23:49.370 --> 00:23:53.510 align:middle line:84% assume that will be a linear transitions 00:23:53.510 --> 00:24:00.260 align:middle line:84% and when the switch current decrease from the ill the 00:24:00.260 --> 00:24:02.470 align:middle line:90% capacitor current can conduct. 00:24:02.470 --> 00:24:07.240 align:middle line:90% 00:24:07.240 --> 00:24:10.220 align:middle line:90% And pay attention at this time. 00:24:10.220 --> 00:24:14.860 align:middle line:84% The id of the diode is still in the off state. 00:24:14.860 --> 00:24:18.970 align:middle line:84% So when the switch has a linear ramp of the current, 00:24:18.970 --> 00:24:21.490 align:middle line:84% the capacitor will, similarly, have 00:24:21.490 --> 00:24:23.010 align:middle line:90% the linear ramp of the current. 00:24:23.010 --> 00:24:25.930 align:middle line:90% 00:24:25.930 --> 00:24:27.605 align:middle line:84% And at this, this will be the IL. 00:24:27.605 --> 00:24:32.810 align:middle line:90% 00:24:32.810 --> 00:24:35.240 align:middle line:84% And the full capacitor voltage, we 00:24:35.240 --> 00:24:42.250 align:middle line:84% know that the capacitance current 00:24:42.250 --> 00:24:45.700 align:middle line:90% will be c times the dv/dt. 00:24:45.700 --> 00:24:48.400 align:middle line:84% And correspondingly, the voltage of a capacitor 00:24:48.400 --> 00:24:51.495 align:middle line:84% will be the integral of the capacitance current. 00:24:51.495 --> 00:24:54.160 align:middle line:90% 00:24:54.160 --> 00:24:56.680 align:middle line:84% And here, if we have a linear of the switch 00:24:56.680 --> 00:25:01.180 align:middle line:84% of the capacitance current, we will have the quadratic 00:25:01.180 --> 00:25:03.040 align:middle line:90% of the capacitor voltage. 00:25:03.040 --> 00:25:06.370 align:middle line:84% And that capacitor voltage will be the switch voltage 00:25:06.370 --> 00:25:07.260 align:middle line:90% at the same time. 00:25:07.260 --> 00:25:10.710 align:middle line:90% 00:25:10.710 --> 00:25:12.230 align:middle line:84% Maybe I should change another color. 00:25:12.230 --> 00:25:20.790 align:middle line:90% 00:25:20.790 --> 00:25:25.660 align:middle line:84% So if we draw it, the waveform, this 00:25:25.660 --> 00:25:27.730 align:middle line:90% will be the capacitance current. 00:25:27.730 --> 00:25:31.060 align:middle line:84% If we draw the capacitance voltage, because this is linear, 00:25:31.060 --> 00:25:34.850 align:middle line:84% the integral of the linear will be quadratic. 00:25:34.850 --> 00:25:38.150 align:middle line:84% So at this time, the switch current will be off. 00:25:38.150 --> 00:25:40.430 align:middle line:84% And after that, the capacitance current 00:25:40.430 --> 00:25:44.310 align:middle line:84% will fully conduct the load current. 00:25:44.310 --> 00:25:47.910 align:middle line:84% And when the capacitance current is a constant, 00:25:47.910 --> 00:25:51.090 align:middle line:84% the capacitance voltage will become linear. 00:25:51.090 --> 00:25:52.680 align:middle line:84% But we don't care about this part 00:25:52.680 --> 00:25:55.350 align:middle line:84% anymore because we care the switching loss. 00:25:55.350 --> 00:25:57.870 align:middle line:84% I mean, for this time, we don't care. 00:25:57.870 --> 00:26:00.300 align:middle line:84% We only care the switching current and voltage 00:26:00.300 --> 00:26:05.310 align:middle line:84% during this time interval, from 0 to tf. 00:26:05.310 --> 00:26:09.980 align:middle line:84% So we only care the energy loss during this. 00:26:09.980 --> 00:26:16.170 align:middle line:84% So actually, we can have a calculation about this energy. 00:26:16.170 --> 00:26:17.230 align:middle line:90% So let's see. 00:26:17.230 --> 00:26:25.380 align:middle line:84% Our switch current is IL 1 minus t over tf. 00:26:25.380 --> 00:26:28.140 align:middle line:84% And the corresponding layer, our capacitance current, 00:26:28.140 --> 00:26:32.850 align:middle line:90% will be IL times t times tf. 00:26:32.850 --> 00:26:35.760 align:middle line:84% And correspondingly, our capacitance voltage, 00:26:35.760 --> 00:26:37.800 align:middle line:84% which is also the switch voltage, 00:26:37.800 --> 00:26:39.755 align:middle line:90% will be the integral of this. 00:26:39.755 --> 00:26:42.300 align:middle line:90% 00:26:42.300 --> 00:26:49.630 align:middle line:84% That will be the IL times the t over tf dt. 00:26:49.630 --> 00:26:56.410 align:middle line:84% So that will be IL tf times C t squared. 00:26:56.410 --> 00:27:01.176 align:middle line:84% So let me check if I do my algebra correct. 00:27:01.176 --> 00:27:02.710 align:middle line:90% Oh, sorry. 00:27:02.710 --> 00:27:04.920 align:middle line:84% So I should plug into the switch current. 00:27:04.920 --> 00:27:10.220 align:middle line:90% 00:27:10.220 --> 00:27:11.240 align:middle line:90% Oh, sorry. 00:27:11.240 --> 00:27:11.753 align:middle line:90% My bad. 00:27:11.753 --> 00:27:12.420 align:middle line:90% This is correct. 00:27:12.420 --> 00:27:16.210 align:middle line:90% 00:27:16.210 --> 00:27:21.250 align:middle line:84% So that will be IL t squared C times the tf. 00:27:21.250 --> 00:27:23.780 align:middle line:84% So that will be the switch voltage expression. 00:27:23.780 --> 00:27:27.610 align:middle line:84% And similarly, if we calculate the switching loss, 00:27:27.610 --> 00:27:28.670 align:middle line:90% what is loss? 00:27:28.670 --> 00:27:30.640 align:middle line:90% Loss is a, still, kind of power. 00:27:30.640 --> 00:27:36.970 align:middle line:84% That will be the v switch times the i switch. 00:27:36.970 --> 00:27:39.880 align:middle line:90% So we have v switch as this. 00:27:39.880 --> 00:27:44.740 align:middle line:84% That will be IC times t squared over the capacitance value 00:27:44.740 --> 00:27:48.070 align:middle line:84% times the tf times the switch current, which 00:27:48.070 --> 00:27:49.360 align:middle line:90% is this expression. 00:27:49.360 --> 00:27:54.880 align:middle line:84% That will be the IL 1 minus t over tf. 00:27:54.880 --> 00:27:58.030 align:middle line:84% So that will be the instantaneous power expression. 00:27:58.030 --> 00:28:01.120 align:middle line:84% And if we integrate this instantaneous power, 00:28:01.120 --> 00:28:03.790 align:middle line:84% we will have the turn-off energy. 00:28:03.790 --> 00:28:09.920 align:middle line:84% That is like we integrate this PSW during the time 00:28:09.920 --> 00:28:14.660 align:middle line:90% interval from 0 to the tf. 00:28:14.660 --> 00:28:17.510 align:middle line:84% And if you do the algebra by yourself, 00:28:17.510 --> 00:28:24.740 align:middle line:84% we will find the result, which is 00:28:24.740 --> 00:28:28.490 align:middle line:84% IL squared times the tf squared over the 24 00:28:28.490 --> 00:28:31.510 align:middle line:90% times the capacitance. 00:28:31.510 --> 00:28:33.990 align:middle line:84% So do you have any questions about this part? 00:28:33.990 --> 00:28:34.755 align:middle line:90% Yeah, go ahead. 00:28:34.755 --> 00:28:38.550 align:middle line:84% AUDIENCE: ic is just equal to IL 1 minus t over tf, 00:28:38.550 --> 00:28:43.350 align:middle line:84% is that coming from that IL is equal to id plus x plus xt? 00:28:43.350 --> 00:28:43.990 align:middle line:90% XIN ZAN: Yes. 00:28:43.990 --> 00:28:46.390 align:middle line:90% So here, we have assumption. 00:28:46.390 --> 00:28:48.570 align:middle line:84% So even if we have a snubber, we still 00:28:48.570 --> 00:28:51.270 align:middle line:84% assume the transient for the switch current is linear. 00:28:51.270 --> 00:28:53.040 align:middle line:84% And we know that the current will 00:28:53.040 --> 00:28:54.520 align:middle line:90% decrease during the turn-off. 00:28:54.520 --> 00:28:57.790 align:middle line:84% So we have assumption like this will be a linear current. 00:28:57.790 --> 00:29:02.550 align:middle line:84% So that's why we have it start as a part. 00:29:02.550 --> 00:29:03.690 align:middle line:90% And sure. 00:29:03.690 --> 00:29:07.470 align:middle line:84% And previously, because we don't have another alternate path 00:29:07.470 --> 00:29:12.360 align:middle line:84% for current to flow, so we need to wait until this id. 00:29:12.360 --> 00:29:13.702 align:middle line:90% But id is still 0 at this time. 00:29:13.702 --> 00:29:14.910 align:middle line:90% AUDIENCE: OK, it makes sense. 00:29:14.910 --> 00:29:17.370 align:middle line:90% XIN ZAN: Yeah, cool. 00:29:17.370 --> 00:29:20.260 align:middle line:84% So then we can calculate this turn-off energy. 00:29:20.260 --> 00:29:21.990 align:middle line:84% And what we can observe from this 00:29:21.990 --> 00:29:26.370 align:middle line:84% is we can have a control handle, which is our capacitance. 00:29:26.370 --> 00:29:29.580 align:middle line:84% So if we want to decrease our turn-off energy, 00:29:29.580 --> 00:29:32.750 align:middle line:84% we can increase our capacitance value. 00:29:32.750 --> 00:29:34.200 align:middle line:90% So we can do this. 00:29:34.200 --> 00:29:35.910 align:middle line:90% However, nothing is free. 00:29:35.910 --> 00:29:38.270 align:middle line:90% We have to pay for something. 00:29:38.270 --> 00:29:41.120 align:middle line:84% So imagine if we have a capacitance in parallel 00:29:41.120 --> 00:29:45.290 align:middle line:84% with a switch, every time when the switch is off, 00:29:45.290 --> 00:29:48.200 align:middle line:84% we will store the energy inside of the capacitance. 00:29:48.200 --> 00:29:49.885 align:middle line:90% That will be 1/2 cv squared. 00:29:49.885 --> 00:29:52.820 align:middle line:90% 00:29:52.820 --> 00:29:58.140 align:middle line:84% So we reduce the energy loss in the inside of the switch. 00:29:58.140 --> 00:30:00.950 align:middle line:84% But outside the switch, we need to dissipate 00:30:00.950 --> 00:30:02.360 align:middle line:90% this part of energy. 00:30:02.360 --> 00:30:05.600 align:middle line:84% So that's what do I mean by we decrease the switching loss 00:30:05.600 --> 00:30:09.260 align:middle line:84% inside of the switches, but we also have to pay for something. 00:30:09.260 --> 00:30:12.680 align:middle line:84% That is the energy stored in this snubber capacitance. 00:30:12.680 --> 00:30:15.170 align:middle line:84% So during the turn-on, if we do nothing, 00:30:15.170 --> 00:30:17.910 align:middle line:84% this energy will be dumped into these switches. 00:30:17.910 --> 00:30:19.190 align:middle line:90% So it will still loss. 00:30:19.190 --> 00:30:21.980 align:middle line:84% And this part of energy could be very large. 00:30:21.980 --> 00:30:24.630 align:middle line:84% It can still destroy our devices. 00:30:24.630 --> 00:30:29.540 align:middle line:84% So this is not a practical snubber as the turn-off snubber. 00:30:29.540 --> 00:30:33.100 align:middle line:84% So a practical one, we need to do something else. 00:30:33.100 --> 00:30:35.620 align:middle line:84% So let's still take this circuit. 00:30:35.620 --> 00:30:37.110 align:middle line:90% We have the input voltage. 00:30:37.110 --> 00:30:40.140 align:middle line:90% 00:30:40.140 --> 00:30:41.970 align:middle line:90% And we have the MOSFET. 00:30:41.970 --> 00:30:44.310 align:middle line:90% We have the diode. 00:30:44.310 --> 00:30:47.960 align:middle line:84% And we have the ideal current source. 00:30:47.960 --> 00:30:51.650 align:middle line:90% So we use two other components. 00:30:51.650 --> 00:30:56.470 align:middle line:84% One is the diode in parallel with a resistor 00:30:56.470 --> 00:31:00.860 align:middle line:84% and then the [INAUDIBLE] with this capacitance. 00:31:00.860 --> 00:31:05.410 align:middle line:84% So during its-- well, we want to turn off the switch. 00:31:05.410 --> 00:31:08.770 align:middle line:84% The current will still flow through this, 00:31:08.770 --> 00:31:11.230 align:middle line:84% through this diode and the capacitance. 00:31:11.230 --> 00:31:13.590 align:middle line:84% And then, when we want to turn on the switch, 00:31:13.590 --> 00:31:15.280 align:middle line:84% so then, because of we are charging 00:31:15.280 --> 00:31:18.140 align:middle line:84% this capacitance to the v in, this will be positive, 00:31:18.140 --> 00:31:19.420 align:middle line:90% this will be negative. 00:31:19.420 --> 00:31:22.090 align:middle line:84% And when we turn on the switch, this capacitance 00:31:22.090 --> 00:31:24.940 align:middle line:84% will have a current path through this resistor 00:31:24.940 --> 00:31:27.910 align:middle line:90% and through this MOSFET. 00:31:27.910 --> 00:31:29.980 align:middle line:84% And most of the energy we can control. 00:31:29.980 --> 00:31:33.650 align:middle line:84% We can dump it into this resistor instead of this MOSFET. 00:31:33.650 --> 00:31:35.560 align:middle line:84% And also, we can use this resistor 00:31:35.560 --> 00:31:39.080 align:middle line:84% to limit the current that flow through this MOSFET. 00:31:39.080 --> 00:31:44.200 align:middle line:84% And in this way, we kind solve this part of energy 00:31:44.200 --> 00:31:47.770 align:middle line:84% that could destroy our device during the turn-on. 00:31:47.770 --> 00:31:51.230 align:middle line:84% But again, we are not recover this part of energy. 00:31:51.230 --> 00:31:53.560 align:middle line:84% This part of energy is still dissipated, 00:31:53.560 --> 00:31:56.530 align:middle line:84% mostly on this resistor and the small part 00:31:56.530 --> 00:31:58.640 align:middle line:90% of amount in this MOSFET. 00:31:58.640 --> 00:32:01.320 align:middle line:90% But we can improve this. 00:32:01.320 --> 00:32:05.060 align:middle line:84% So do you have any questions for a practical turn-off snubber? 00:32:05.060 --> 00:32:08.190 align:middle line:90% 00:32:08.190 --> 00:32:08.850 align:middle line:90% OK. 00:32:08.850 --> 00:32:13.350 align:middle line:84% So then let's go to the turn-on snubber. 00:32:13.350 --> 00:32:16.080 align:middle line:84% So the beauty of the circuit is that sometimes we 00:32:16.080 --> 00:32:18.790 align:middle line:84% can have a dual circuit of the original circuit. 00:32:18.790 --> 00:32:20.940 align:middle line:84% So a turn-on snubber is the real-- the dual, 00:32:20.940 --> 00:32:22.730 align:middle line:90% circuit of the turn-off snubber. 00:32:22.730 --> 00:32:25.620 align:middle line:84% So let's draw the buck converter again. 00:32:25.620 --> 00:32:28.080 align:middle line:90% So we have a voltage source. 00:32:28.080 --> 00:32:31.080 align:middle line:90% We have a transistor. 00:32:31.080 --> 00:32:34.150 align:middle line:90% And we have a diode. 00:32:34.150 --> 00:32:36.700 align:middle line:84% And we have the constant current load source, 00:32:36.700 --> 00:32:38.890 align:middle line:90% a load, constant current load. 00:32:38.890 --> 00:32:44.410 align:middle line:84% And now I insert a inductor in series with this transistor, 00:32:44.410 --> 00:32:46.220 align:middle line:90% with this MOSFET. 00:32:46.220 --> 00:32:49.040 align:middle line:84% So previously, we have the KVL equation, 00:32:49.040 --> 00:32:55.930 align:middle line:84% which is v in is equal to the vsw minus the vd. 00:32:55.930 --> 00:32:59.200 align:middle line:84% And now we have an extra voltage, extra components, 00:32:59.200 --> 00:33:03.370 align:middle line:84% that can sustain the voltage, which is vL. 00:33:03.370 --> 00:33:07.570 align:middle line:84% So previously, we are limited by the vi curve of this diode. 00:33:07.570 --> 00:33:10.150 align:middle line:84% And now we have a second component 00:33:10.150 --> 00:33:12.370 align:middle line:84% to sustain this input voltage, which 00:33:12.370 --> 00:33:14.770 align:middle line:84% means when we turn on the switch, 00:33:14.770 --> 00:33:17.750 align:middle line:84% we can have simultaneously small current and voltage 00:33:17.750 --> 00:33:18.730 align:middle line:90% at the same time. 00:33:18.730 --> 00:33:22.270 align:middle line:84% We don't need to sustain the linear one during the overlap 00:33:22.270 --> 00:33:24.040 align:middle line:90% as the triangular. 00:33:24.040 --> 00:33:28.420 align:middle line:84% So if we draw the waveform of the turn-on, 00:33:28.420 --> 00:33:31.150 align:middle line:84% that is still like we have the input. 00:33:31.150 --> 00:33:35.320 align:middle line:84% We have the switch voltage, which is vsw. 00:33:35.320 --> 00:33:37.940 align:middle line:84% They will have a linear decrease. 00:33:37.940 --> 00:33:40.150 align:middle line:90% This is our assumption. 00:33:40.150 --> 00:33:43.780 align:middle line:84% And also, because we have a linear decrease of our switch 00:33:43.780 --> 00:33:45.940 align:middle line:84% voltage, we will have a linear increase 00:33:45.940 --> 00:33:48.450 align:middle line:90% of our inductor voltage. 00:33:48.450 --> 00:33:52.950 align:middle line:84% So the inductor voltage will increase linearly. 00:33:52.950 --> 00:33:58.500 align:middle line:84% And the most basic equation for an inductor, that will be vL 00:33:58.500 --> 00:33:59.540 align:middle line:90% is L di/dt. 00:33:59.540 --> 00:34:02.100 align:middle line:90% 00:34:02.100 --> 00:34:05.550 align:middle line:84% So correspondingly, if we want to find the inductor current, 00:34:05.550 --> 00:34:09.449 align:middle line:84% we can just do the integral, which is 1 over L 00:34:09.449 --> 00:34:12.210 align:middle line:90% times the vL times the dt. 00:34:12.210 --> 00:34:15.270 align:middle line:84% So if we know that the inductor voltage will 00:34:15.270 --> 00:34:18.750 align:middle line:84% be a linear increase, then the inductor current-- 00:34:18.750 --> 00:34:21.420 align:middle line:84% and if you pay attention, this inductor current 00:34:21.420 --> 00:34:23.280 align:middle line:84% will be same as the switch current 00:34:23.280 --> 00:34:25.770 align:middle line:90% because they are in series. 00:34:25.770 --> 00:34:31.239 align:middle line:84% And now, this switch current will increase as quadratical. 00:34:31.239 --> 00:34:33.699 align:middle line:84% And during this, this is the quadratical 00:34:33.699 --> 00:34:36.770 align:middle line:90% of the switch current. 00:34:36.770 --> 00:34:40.790 align:middle line:84% And after this tr, the switch is fully on. 00:34:40.790 --> 00:34:45.400 align:middle line:90% So the inductor current-- 00:34:45.400 --> 00:34:46.070 align:middle line:90% sorry. 00:34:46.070 --> 00:34:48.139 align:middle line:90% The switch is fully on. 00:34:48.139 --> 00:34:49.360 align:middle line:90% This will be 0. 00:34:49.360 --> 00:34:50.830 align:middle line:90% And this is still 0. 00:34:50.830 --> 00:34:53.659 align:middle line:84% And this vL will fully take the input voltage. 00:34:53.659 --> 00:34:56.230 align:middle line:84% So this will be a constant value. 00:34:56.230 --> 00:34:59.170 align:middle line:84% And if it's a constant value for inductor voltage, 00:34:59.170 --> 00:35:00.855 align:middle line:90% then the current will be linear. 00:35:00.855 --> 00:35:04.550 align:middle line:90% 00:35:04.550 --> 00:35:07.070 align:middle line:84% But again, we don't care this part for the switches. 00:35:07.070 --> 00:35:08.870 align:middle line:90% Because it is after the tr. 00:35:08.870 --> 00:35:11.330 align:middle line:84% The switch has already passed the transient. 00:35:11.330 --> 00:35:13.490 align:middle line:84% And we can still calculate what will 00:35:13.490 --> 00:35:16.070 align:middle line:84% be the turn-on energy during this part 00:35:16.070 --> 00:35:18.600 align:middle line:84% and compare it to the original hard switching. 00:35:18.600 --> 00:35:21.770 align:middle line:84% We can see that we have small overlap 00:35:21.770 --> 00:35:24.380 align:middle line:84% at the same time for the high voltage and high current. 00:35:24.380 --> 00:35:27.810 align:middle line:84% So we can improve the turn-on energy during this. 00:35:27.810 --> 00:35:31.820 align:middle line:84% And also, the same problem will be we still store the energy. 00:35:31.820 --> 00:35:36.510 align:middle line:84% That is 1/2 Li squared in this inductor. 00:35:36.510 --> 00:35:39.620 align:middle line:84% We need to dissipate this part of energy somewhere. 00:35:39.620 --> 00:35:45.140 align:middle line:84% And for practical use, we still have a resistor 00:35:45.140 --> 00:35:46.825 align:middle line:90% in series with a diode. 00:35:46.825 --> 00:35:49.350 align:middle line:90% 00:35:49.350 --> 00:35:53.220 align:middle line:84% And if we want to dissipate the energy of this inductor, 00:35:53.220 --> 00:35:55.510 align:middle line:84% we will have a current flow like this 00:35:55.510 --> 00:35:57.420 align:middle line:90% when the switch is turned off. 00:35:57.420 --> 00:35:59.560 align:middle line:84% So that's the beauty of the circuit. 00:35:59.560 --> 00:36:05.280 align:middle line:84% So for a concept, we need a turn-off parallel capacitance. 00:36:05.280 --> 00:36:09.280 align:middle line:84% And for the concept, we need a series turn-on inductor. 00:36:09.280 --> 00:36:14.040 align:middle line:84% And for a more practical use, we need a diode in parallel 00:36:14.040 --> 00:36:17.100 align:middle line:84% with a resistor, then in series with this capacitor. 00:36:17.100 --> 00:36:19.410 align:middle line:84% And for the practical turn-off snubber, 00:36:19.410 --> 00:36:23.170 align:middle line:84% we need a diode in series with a resistor, 00:36:23.170 --> 00:36:25.000 align:middle line:84% then in parallel with this inductor. 00:36:25.000 --> 00:36:27.460 align:middle line:84% So this is the fully dual of the circuit. 00:36:27.460 --> 00:36:31.120 align:middle line:84% So this is a practical, one of the practical, turn-on snubber. 00:36:31.120 --> 00:36:34.710 align:middle line:84% And this is the practical turn-off snubber. 00:36:34.710 --> 00:36:37.130 align:middle line:90% So do you have any questions? 00:36:37.130 --> 00:36:38.370 align:middle line:90% Yeah, go ahead. 00:36:38.370 --> 00:36:39.980 align:middle line:84% AUDIENCE: Would you use both at once? 00:36:39.980 --> 00:36:40.563 align:middle line:90% XIN ZAN: Yeah. 00:36:40.563 --> 00:36:42.260 align:middle line:84% That's, actually, a very good question. 00:36:42.260 --> 00:36:45.110 align:middle line:84% So this is only we care about the turn-off. 00:36:45.110 --> 00:36:47.090 align:middle line:84% This is what we care about the turn-on. 00:36:47.090 --> 00:36:52.400 align:middle line:84% And also, again, we dissipate this amount of energy fully. 00:36:52.400 --> 00:36:54.560 align:middle line:84% So in the practical, we can design 00:36:54.560 --> 00:36:57.840 align:middle line:84% some snubber that can both deal with the turn-on and turn-off. 00:36:57.840 --> 00:37:00.890 align:middle line:84% And we can imagine, some semiconductor devices, we 00:37:00.890 --> 00:37:02.960 align:middle line:84% may care more about the turn-on, or we 00:37:02.960 --> 00:37:04.220 align:middle line:90% may care about the turn-off. 00:37:04.220 --> 00:37:05.900 align:middle line:84% So we can select a trade-off, like we 00:37:05.900 --> 00:37:08.360 align:middle line:84% want to mitigate the turn-on loss more, 00:37:08.360 --> 00:37:10.650 align:middle line:84% or we want to mitigate the turn-off loss more. 00:37:10.650 --> 00:37:12.620 align:middle line:84% And also, here, we are dissipating 00:37:12.620 --> 00:37:16.320 align:middle line:84% the energy of the half cv squared and the half Li squared. 00:37:16.320 --> 00:37:19.650 align:middle line:84% In the practical, we can still design some recovery circuit. 00:37:19.650 --> 00:37:22.760 align:middle line:84% We can recycle this part of energy. 00:37:22.760 --> 00:37:24.710 align:middle line:90% So what's your question? 00:37:24.710 --> 00:37:28.460 align:middle line:84% AUDIENCE: In the [INAUDIBLE], we have a [INAUDIBLE], 00:37:28.460 --> 00:37:31.792 align:middle line:84% can you just mention what vd is doing during all of that? 00:37:31.792 --> 00:37:33.000 align:middle line:90% XIN ZAN: You mean this diode? 00:37:33.000 --> 00:37:34.320 align:middle line:90% AUDIENCE: Yeah. 00:37:34.320 --> 00:37:38.610 align:middle line:84% XIN ZAN: So after this, when the-- 00:37:38.610 --> 00:37:43.680 align:middle line:84% let's say this is the inductor current, which 00:37:43.680 --> 00:37:45.150 align:middle line:90% also the switch current. 00:37:45.150 --> 00:37:49.320 align:middle line:84% So when the switch current is going to the IL, 00:37:49.320 --> 00:37:53.850 align:middle line:84% and, also, similar here, when the capacitor 00:37:53.850 --> 00:38:00.660 align:middle line:84% voltage, or the switch voltage, is fully go to the v in, 00:38:00.660 --> 00:38:03.660 align:middle line:84% then the diode will make a row in here. 00:38:03.660 --> 00:38:05.315 align:middle line:84% That will be similar in this case. 00:38:05.315 --> 00:38:08.110 align:middle line:90% 00:38:08.110 --> 00:38:13.753 align:middle line:84% So the diode will play a role after this time interval. 00:38:13.753 --> 00:38:16.420 align:middle line:84% AUDIENCE: Why is dv not changing at all to the left of that time 00:38:16.420 --> 00:38:16.850 align:middle line:90% interval? 00:38:16.850 --> 00:38:17.350 align:middle line:90% XIN ZAN: Oh. 00:38:17.350 --> 00:38:19.180 align:middle line:84% That will be the same reason of this. 00:38:19.180 --> 00:38:22.080 align:middle line:84% Because we still assume this is an ideal diode. 00:38:22.080 --> 00:38:25.900 align:middle line:84% So the diode cannot turn on or carry current until it has 00:38:25.900 --> 00:38:27.240 align:middle line:90% zero-voltage drop. 00:38:27.240 --> 00:38:31.390 align:middle line:84% And similarly, a diode cannot turn off or block the voltage 00:38:31.390 --> 00:38:32.970 align:middle line:90% until the current becomes 0. 00:38:32.970 --> 00:38:36.030 align:middle line:90% 00:38:36.030 --> 00:38:37.610 align:middle line:90% Cool. 00:38:37.610 --> 00:38:40.520 align:middle line:84% So do you have any questions about those numbers? 00:38:40.520 --> 00:38:42.350 align:middle line:90% Yeah, go ahead. 00:38:42.350 --> 00:38:45.280 align:middle line:84% AUDIENCE: Does [INAUDIBLE] capacitor or conductor 00:38:45.280 --> 00:38:46.867 align:middle line:84% make the resonance problem worse? 00:38:46.867 --> 00:38:47.450 align:middle line:90% XIN ZAN: Yeah. 00:38:47.450 --> 00:38:49.350 align:middle line:84% That's, actually, a very good question. 00:38:49.350 --> 00:38:53.090 align:middle line:84% So let's say, because we have a parallel-- 00:38:53.090 --> 00:38:55.490 align:middle line:84% we have a series inductance here. 00:38:55.490 --> 00:38:57.740 align:middle line:84% And during some transient, we still 00:38:57.740 --> 00:38:59.460 align:middle line:90% have this parallel capacitance. 00:38:59.460 --> 00:39:01.280 align:middle line:84% And for this diode, there also could 00:39:01.280 --> 00:39:02.860 align:middle line:90% be some parallel capacitance. 00:39:02.860 --> 00:39:04.260 align:middle line:90% They're still resonant. 00:39:04.260 --> 00:39:06.360 align:middle line:84% That is something we cannot avoid. 00:39:06.360 --> 00:39:08.990 align:middle line:84% So in the textbook, there's a chapter talk 00:39:08.990 --> 00:39:11.570 align:middle line:84% about how we choose this inductor value 00:39:11.570 --> 00:39:14.090 align:middle line:84% and how we choose this resistor value. 00:39:14.090 --> 00:39:18.720 align:middle line:84% And similarly, we still have the parasitic stray inductance here. 00:39:18.720 --> 00:39:21.170 align:middle line:84% And during some case, this capacitor 00:39:21.170 --> 00:39:22.810 align:middle line:90% will still resonate with this. 00:39:22.810 --> 00:39:25.850 align:middle line:84% So this is something we cannot avoid. 00:39:25.850 --> 00:39:28.250 align:middle line:84% But we can have some circuits, or elements, 00:39:28.250 --> 00:39:30.080 align:middle line:84% like how we choose this capacitance 00:39:30.080 --> 00:39:33.710 align:middle line:84% and how we choose this resistor to improve this. 00:39:33.710 --> 00:39:36.860 align:middle line:84% And also, if you imagine a second-order system, 00:39:36.860 --> 00:39:38.280 align:middle line:90% we can have critical damped. 00:39:38.280 --> 00:39:39.330 align:middle line:90% We can have overdamped. 00:39:39.330 --> 00:39:40.560 align:middle line:90% Or we can have underdamped. 00:39:40.560 --> 00:39:43.440 align:middle line:84% So we can selectively purposely put some small resistor 00:39:43.440 --> 00:39:44.740 align:middle line:90% to dampen that out. 00:39:44.740 --> 00:39:48.000 align:middle line:84% But this is something we need to pay off about that. 00:39:48.000 --> 00:39:50.470 align:middle line:84% That's, actually, a very good question. 00:39:50.470 --> 00:39:52.030 align:middle line:84% And here, I want to still mention 00:39:52.030 --> 00:39:54.610 align:middle line:84% about the snubber is we improve the switching 00:39:54.610 --> 00:39:56.500 align:middle line:84% loss or the turn-on and the turn-off energy 00:39:56.500 --> 00:39:58.570 align:middle line:90% inside of the switch itself. 00:39:58.570 --> 00:40:03.520 align:middle line:90% But we lose energy here. 00:40:03.520 --> 00:40:05.800 align:middle line:90% And we lose energy here. 00:40:05.800 --> 00:40:09.100 align:middle line:84% And the total energy we lost may be even more. 00:40:09.100 --> 00:40:10.690 align:middle line:90% But why we want to do this? 00:40:10.690 --> 00:40:13.300 align:middle line:84% Because we want to keep the switch iv 00:40:13.300 --> 00:40:16.030 align:middle line:84% curves inside of the safe operating area. 00:40:16.030 --> 00:40:19.030 align:middle line:84% We want to have a safe di/dt or dv/dt. 00:40:19.030 --> 00:40:20.680 align:middle line:90% So that's the motivation. 00:40:20.680 --> 00:40:23.350 align:middle line:84% And as I said before, like for the snubbers, 00:40:23.350 --> 00:40:25.090 align:middle line:90% we use it for IGBT. 00:40:25.090 --> 00:40:26.740 align:middle line:90% We use it for thyristor. 00:40:26.740 --> 00:40:28.270 align:middle line:90% We use it for the BJT. 00:40:28.270 --> 00:40:31.030 align:middle line:84% And currently for the MOSFET, we generally 00:40:31.030 --> 00:40:32.830 align:middle line:84% don't use snubber for the MOSFET. 00:40:32.830 --> 00:40:34.600 align:middle line:84% But from this, we still introduce 00:40:34.600 --> 00:40:36.820 align:middle line:84% this part of the content because you can learn 00:40:36.820 --> 00:40:38.270 align:middle line:90% the beauty of a capacitor. 00:40:38.270 --> 00:40:40.190 align:middle line:84% You can learn the beauty of a diode. 00:40:40.190 --> 00:40:43.100 align:middle line:84% So the diode is only one directional device. 00:40:43.100 --> 00:40:45.610 align:middle line:84% So you can imagine, for future, when you design a circuit, 00:40:45.610 --> 00:40:48.850 align:middle line:84% how to use a diode to control your current flow paths. 00:40:48.850 --> 00:40:52.170 align:middle line:90% 00:40:52.170 --> 00:40:54.930 align:middle line:84% So another thing I want to say, because we still have some time, 00:40:54.930 --> 00:40:57.660 align:middle line:84% is I want to introduce the gate drive. 00:40:57.660 --> 00:41:01.080 align:middle line:84% As we said before, for this tr and tf-- 00:41:01.080 --> 00:41:03.950 align:middle line:90% where is that? 00:41:03.950 --> 00:41:07.590 align:middle line:84% For this tf or tr, what do they depend on? 00:41:07.590 --> 00:41:10.920 align:middle line:84% They depend on the devices, or they depend on the drivers. 00:41:10.920 --> 00:41:13.910 align:middle line:84% So imagine we have a huge, or a big, device. 00:41:13.910 --> 00:41:16.040 align:middle line:84% It may take longer time to finish the turn-on 00:41:16.040 --> 00:41:17.540 align:middle line:90% and the turn-off transient. 00:41:17.540 --> 00:41:19.490 align:middle line:84% But if we have a powerful driver, 00:41:19.490 --> 00:41:21.530 align:middle line:90% maybe we can reduce that time. 00:41:21.530 --> 00:41:24.800 align:middle line:84% So I want to introduce a small bit of the gate driver 00:41:24.800 --> 00:41:30.240 align:middle line:84% is like sometimes when we want to drive a MOSFET. 00:41:30.240 --> 00:41:34.310 align:middle line:84% So this is the voltage source, which is a gate drive. 00:41:34.310 --> 00:41:38.260 align:middle line:84% So we can have different paths to turn on 00:41:38.260 --> 00:41:43.830 align:middle line:90% and to turn off of the MOSFET. 00:41:43.830 --> 00:41:47.520 align:middle line:84% So during the turn-on, we can select a relatively bigger 00:41:47.520 --> 00:41:49.410 align:middle line:84% resistor to charge this capacitor. 00:41:49.410 --> 00:41:52.080 align:middle line:84% For turn-off, we really want it to be fast, 00:41:52.080 --> 00:41:53.780 align:middle line:84% so we can have a small resistor here. 00:41:53.780 --> 00:41:56.630 align:middle line:90% 00:41:56.630 --> 00:41:59.665 align:middle line:84% So do you have any further questions about today's lecture? 00:41:59.665 --> 00:42:06.620 align:middle line:90% 00:42:06.620 --> 00:42:09.960 align:middle line:90% We still have some time. 00:42:09.960 --> 00:42:13.210 align:middle line:84% So if not, that will be end for today's lecture. 00:42:13.210 --> 00:42:15.080 align:middle line:90% Thank you. 00:42:15.080 --> 00:42:17.000 align:middle line:90%