Measuring a MOSFET’s Miller Plateau - Workbench Wednesdays
HTML-код
- Опубликовано: 20 сен 2024
- Turning on a MOSFET takes more than knowing the threshold voltage. A special event occurs when a FET turns-on, which is called the Miller Plateau. The gate voltage sticks to the threshold voltage while the drain opens up. In this video, James shows how to measure the Miller Capacitance, using an oscilloscope and custom MOSFET test board: bit.ly/3iEMH5v
Engage with the element14 presents team on the element14 Community - suggest builds, find project files and behind the scenes video: bit.ly/2MFMG0v
Visit the element14 Community for more great activities and free hardware:
Tech spotlights: bit.ly/2KLz0TS
Roadtest and Reviews: bit.ly/2KH4fj3
Project14: bit.ly/2wPnajx
I watched several videos about Miller capacitance, but I've enjoyed this one the most
Is there a follow to this? As a brief demo of the Miller plateau, it's okay, but what most folks really want to see are some concrete examples. It's reasonable to skip the "bunch of maths", but then why not simply connect a multi-turn pot of about 250Ω as a gate resistor and show how you would optimise the switching for a given mosfet and load? You could easily use the Arduino to switch the mosfet faster and faster in order to see how it heats up and demonstrate the upper limit of the gate drive resistor.
A good explanation. Something for me to consider (better) than I have in the past. Thanks.
A unique video that experimentally explains what you only see in diagrams. Thank you for this excellent channel.
I'm using MOSFETs in a new design for the first time ever, so this was useful! Gonna be measuring some switching voltages tomorrow 🥰
Great video
Very nicely explained!! Really understood the Miller plateau well!! Thanks a lot!
Good video for basic understanding of MOSFETs.. good video.
Thanks! 👍
This is a good course. Thanks!
Thank you very much sir, it couldn't be clearer.
Interesting to see slowly turning it on versus slamming it on, both with probing the curcuit and radiated energy. Would have really liked to see the results with different resistor values. That active probe most certainly would give infinitely more accurate results compared to everyday 1/10x probes, which would be even more beneficial to see. Does rise and fall times listed in the datasheet coorelate completely with the input and output capacitance? What other things would you consider for rise and fall times to optimize switching speed? Look forward to a follow up video on this subject
Thank you for the comments. I provided a response here: www.element14.com/community/docs/DOC-95249/l/workbench-wednesday-25-measuring-a-mosfet-s-miller-plateau#comment-268120
Awesome video!
Thank you!
There is one thing I dont understand yet,
on 7:51 the gate (orange) seems to be instantly charged over a 3k resistor, but the mosfet is reacting with an time offset, did you maybe insert a picture taken with the probe on the wrong side of the charging resistor, directly on the signal generator?
Since the resistor + gate is significantly higher impedance than the signal source, it doesn't matter which "side" of the resistor is probed because it is in series with the gate. There is a ramp to the gate's turn-on but it isn't visible because of the scope's timebase.
@@bald_engineer
Thank you for your time :)
Im just wondering why there is a "slow" ramping in the first experiment (orange graph 4:28) with 0 Ohm resistor and in the second part (orange graph 7:51) there is no ramping up on the gate but the resistance is increased to 3kOhm. If there would just be a capacitor of 800(E-12) F charging over 3k Ohm, the RC time base should be 2,4E-6 and we should see charging for around 3x time base but the rising is instantly. (ignoring the gate drain capacity, but it should slow it down further more)
Thats why I think, that the probe maybe was on the wrong side of the gate resistor.
Sorry if im wrong, im just a student :D
@@angerwurst1860 That is the entire point of the video. Without a gate resistor, the Miller Effect causes the shelf which results in switching losses (among other issues.) Additionally, the ringing generates EMI. As I said in the video, the rated capactiances are not "capacitors." They are dynamic characteristics. The gate's capacitance and the gate-drain capactaince CHANGES as the drain turns on. It's the Miller Effect.
@@bald_engineer
Sorry that im a bit annoying.
Okay, I understand the EMI and ringing problem without a gate resistor, but for me there seem to be a logical problem.
With 3k3 there is a time period in the video where the gate voltage is maxed but the mosfet is not reacting.(7:51)
If the gate really finished charging instantly and is not dropping or ringing in this time period, why does the mosfet do nothing for a "long" time?
IF the measuring probe is directly on the signal generator, the orange graph just shows the input signal and not the true gate voltage with its ramping up, that could explain the time where the mosfet is letting no current flow and doing nothing for over 1us after the orange graph stops rising.
I just build a small circuit with an ESP32, MCP1407 mosfet driver, 2N7000 mofet, 3k gate resistor and 1k load resistor, Im on the same time base as you and I see my expected slow ramping up of the gate voltage, I also can identify a small miller plateau in this setup when the drain-source voltage starts changing.
If i connect the probe to the other side of the gate resistor, I see exactly the same behavior like in the video.
I just really really want to understand it, because I think that I can learn a lot from your video and im confused because the shown measurement seem to have a logical inconsistency for me.
When the drain-source voltage is decreasing, then the gate drain capacitor should slow down the further rising of the gate voltage because of evil miller, but in the video U(GS) is maxed and stable and U(DS) is sill sleeping.
Edit again: I red it again, you said it does not matter on wich side of the resistor the probe is.
If you charge the gate by switching on, current is flowing, I*R=U. You have a voltage over this resistor.
The voltage on a capacitor can not jump, the entire voltage is on the resistor for a short time.
You can measure 5V = (U(R)= 4,9V + U(GS)=0,1V)
You would measure 5V over the entire charging time of the gate if you measure on the wrong side.
If the charging is finished, the current is 0, the voltage over the 3k resistor is 0 and in this stable state, then you can assume that the measured voltage is the true gate voltage.
Greetings from Germany.
@@angerwurst1860 Okay. I see your point now and you are correct. I should have had a probe point between the resistor and gate. Someday, I will re do the measurements...
What's the make/model of the hall effect current probe? Thanks.
R&S RT-ZC15B. It is a re-badged Hioki with custom interface for R&S scopes. It is the same probe that all teir 1 scope manufacturers use/sell.
It kind of failed to actually explain what happens but hey, so long as the giant power pro box is in shot that's that matters to the producers.
Yeah, he barely mentions "miller capacitor" and that's it.
Hi, I need help, this is a digital temperature sensor on the pellet burner electronics. Everything works well but I need a little diversion on electronics. The thing is that at some point I need to tell the electronics that some temperature has been reached and if creativity has not been reached. How to influence that temperature via a sensor. There are three wires on the sensor: green, brown and white. Green is electronically connected under one connector and brown and white are connected together under the other connector. So, three wires, two connectors. Thanks.
and how does that remotely relate to the topic? go find a forum.
Hello brother Ilike your video thank you very much
Noice
no first ?, ok I'll "transmit" it
Is there no better way to explain how mosfet works ? this is fast and complicated.
maybe this is not a explanation of how mosfets works...
Great video