MOSFET Gate Drive Resistor Selection - Part 1: Turn On
Вставка
- Опубліковано 26 чер 2024
- In this video, we will describe how to quickly select a "near optimum" value of your Gate Drive resistor in a MOSET driven switch mode power supply during turn on.
The methodology described here should give you a good comprise between fast switching speed, low EMI and low time to market without difficult measurements.
For detailed mathematical analysis and accurate equations, we highly recommend the excellent paper by By Mr. Laszlo Balogh, "Fundamentals of MOSFET and IGBT Gate Driver Circuits" available from TI: www.ti.com/lit/ml/slua618a/sl... - Наука та технологія
UA-cam's algorithm recommended your videos (esp. the Miller Plateau video) right when I was working on a related problem on a design. Very serendipitous...and your videos are great!
Me too. I was watching eevblog about PCB current trace
The real fun starts when switching inductive loads 😉
That's actually where most issues are in practice.
😂😂
Didn't you pay attention? The circuit is a flyback converter (4:20) which is really the most inductive as you can get.
@@gaynzz6841 Yes, I was a bit unclear. Referred more to a simple step down. Normally you will never see a Miller plateau in such cases.
They put also ferrite bead too. One small ring ferrite bead in wire .
Great videos Professor!!!
Greetings from Mexico
Thank you ,
It is really very useful.
Thank you, Learned something new today.
Your teaching skill is awesome. Thank you professor. Love you to see from Bangladesh 🇧🇩.
awesome video!, I wish you could of talk about adding diode in parallel to resistor and see it's effect on the scope. You can also combination of diode and resistor to control the turn-on and turn-off speed.
Very clear and helpful.
So... in your example, you end up with a 5ohm resistor...
How would you cslculate what power rating that resistor needs to be?
Thanks - great explanation.
Great work 👌
Very nicely explained and demonstrated.
Excellent, thanks.
Nice talk. Thanks
Fantastic explaination
Thank you!
Nice explanation , Thanks
Очень понравилось видео! Нужный полезный практический урок. Спасибо Вам!!
Thank you very much sir
Very good
So we need to look at the gate to soyrce vootage and make a oerfect shape like you shown by choosing different gate resistor
Hey, I thought I was the only one to "design by tweak" I tried the "math" way and the values always turned out wrong so ended up tweaking the values anyway!
I must admit I was hoping for a bit more than "try different resistor values, and see what looks best on a scope". But that's fair enough advice in most cases, and you did supply the link to a more detailed analysis.
He links to the paper that describes how to calculate the gate drive resistance provided you can measure the parasitics.
@@ferrumignis Yes, indeed, as I remarked. Most of the time, of course, you'll have an estimate of the maximum parasitic inductance when you know the length and size of the leads/tracks between the driver and the mosfet. That's sufficient to determine the minimum value of resistor needed to damp the LCR circuit.
Wow thanks so this should be done under load...if I'm trying to rive a dc traction motor should I do it under load or off the ground freewheeling with the scope
Please clarify where is the oscillo probe connected. To the mosfet gate?
Thanks a lot for the great video. Is it possible to use a whiteboard instead please? I understand that this is a weird request. But the marker on paper makes a sound that is hard for some people (not everyone) to tolerate. I am sorry for this strange comment. Thank you again. you rchannel is amazing.
😢mee too
Sir, We have to keep underdamped position to the waveform and need to keep less than 10 ohm resistor.
Great experiment. And what will be if you use 0 ohm resistor with the ferrite bead on the gate pin of the MOSFET?
Did you try it ???? Let me know if you have done it
Hi. Does this apply to inductive loads as well as to the logic level signal? is there any formula for gate resistance ?
Is there a way without using a o scope?
dosent the calculation affect fall time
is the ringing coming from the parasitic inductance?
My gate driver circuit burnt. What could be the reason?
One question remains: how can a small 100ohm resistor make such a difference in connection with the Mohm gate resistance of the mosfet?
Does the oscilloscope's probe capacitance (e.g. 1pF) not change the gate signal too much? Or is the gate capacitance order of magnitude bigger than the 1pF of the probe?
Gate capacitance is hundred or hundreds picofarad at least typically and oscilloscope probe is 12-25 picofarad (fine one) so it doesn't change much. If you work with more powerfull transistors it is neglisible. It is only consern at rare cases.
As real examples, a power mosfet like the IRL530 has a typical input capacitance around 1nF and the IRFZ44 around 2nF, while for the small-signal 2N7000, it's just 60pF max.
@@volodymyrzakolodyazhny7740 Thanks
@@RexxSchneider Thank you
as a 10:1 passive probe is typically around 11 pF, it won't load the drive path very much, relative to the gate capacitance of a big mosfet, but really loads a smaller device, where an active probe becomes indispensable!
Also, a typical passive probe, if you use a "ground lead" often adds a resonance at 100-125 MHz.
probing is always raising its head as a tricky factor in switching analysis...
But what if you're just switching on/off a load (no fast switching requirements) ?
Wouldn't a high resistor value be preferable, in order to get lower emi emissions?
Then a higher then optimum will be prefered I guess. Here another problem may arise: your load may be placed at some relatively large distance and you'll have a high transients with ringing on a wire inductance. Then yes - it is more reasonable to to make it as slow as possible.
The problem then becomes the amount of time the mosfet spends in its linear region. If you are sinking a high current with a high voltage supply, you will need to make sure that the power dissipated during the switching is within limits for the device and its heatsink.