Its not his chair anymore. Its her chair now.

Cats do be like that

Miller induced turn on is one of the reasons why gate drive transformers are often used with the SSTC circuits you find online, since they apply negative voltage at turn off.

Yes. Besides SMPS, it decreases the gain of high frequency amplifiers

@@mrjohhhnnnyyy5797 Is that why the current gain goes down at higher currents?

I already put gate current amplifiers in it. They will be tested in the next episodes :).

The cat is the best.

@rpbajb You understand cats? Wow :-)

Best way to start understanding this stuff is to try it for yourself imo, set up a breadboard and connect one of those little SMPS transformers as an inductive load and watch as the gate waveforms change from nice clean square waves to horrible messes!

If you keep on watching these type of videos, eventually stuff starts to sink in (well that's my plan anyways...!)

Agree

The miller effect will be a bit worse at 300V, but not 10x worse. That's because the reverse transfer capacitance goes down with CE voltage. Or if you mean the small step on the upper trace, that's when the internal diode stops passing a current and the transistor starts passing the current instead (the current changes direction).

@@DiodeGoneWild After following you content for so long, your simultaneous broad and depth of knowledge still amazes me

@@DiodeGoneWild The small step seems to be where both switches are of, so the diode should start conducting, and the step should be downward not upward?!, I've seen this waveform so many times and it still annoys me :)

Maybe, but with a load and no snubber, it would destroy the transistors. A 15k resistor wouldn't have any effect. You have to pull the gate voltage down via a very low resistance resistor, just several ohms.

@@DiodeGoneWild Yup inductive load on circuit without snubber would probably made magical puff :), btw You should have autotransformet to start up device like this - it save a lot of transistors ( and time ), but add quite big inductance so current draw can be affected. Im not telling 15k resistor will turn off transistor it give pass for small current and made unintentional transistor capacitance charge less possible (mostly by Miller), proper way is just another transistor (ommitting gate resistor) to pull down gate but it require separate logic to keep transistor slew rate and clamp miller, modern IGBT drivers have integrated miller clamp circuit. Second thing is 18V zenner diode to safe gate from overvoltage.

@@DiodeGoneWild There are some gate drive ICs which bypass the gate drive resistor once turn off has taken place, effectively a miller clamp.

You can decrease the gate resistor to a lower resistance so the gate driver can push more current into the gate to switch it quicker. Also helps if to use a higher gate voltage (15V for power mosfets that have 18V to 20V gate breakdown voltage). The more energy you can dump into the gate, the faster the transistor will turn on. I don't have much experience with IGBTs, but to turn them off faster, you need a slightly negative voltage (from what I recall many years ago). Since you using a inductive load I would recommend that you include a snubber circuit to attenuate voltage spikes from inductance leakage, as well as gate to source (N-channel) TVS diodes so that voltage difference on the gate does not exceed the gate breakdown voltage. The faster you turn on\off your transistors it will result in bigger inductance leakage spikes, which can damage your transistors. So if you circuit is currently working now, you *might* run into issue if when you change your circuit for faster switching. Other options include using MOSFETs with lower Gate charge which will need less gate energy to switch. Generally low Rds MOSFETs have a higher Gate charge. Low Rds perform best with low switching frequency applications, but no so great for high switching frequencies like a Tesla coil. For full H-bridge circuits you can delay the low side Mosfets turn on so that your high side MOSFETs turn on first, then turn on your low side MOSFET, this helps reduce the switch loses. When you try to turn on both the high-side & lowside MOSFETs at the same time, both transistors have high impedence until the reach full conduction., But if you turn on your high side first it can reach full or near full conduction before you turn on the low side, which reduces your switching loses. IIRC I turn on the high side because if your using a gate transformer or a pumped high side driver, there likely is a finite amount of energy you can pump in before either your gate transformer saturates, or the amount of energy stored in the high-side driver cap. That said, I presuming your using all N-Channel MOSFETs for your H-Bridges and not a P\N-Channel bridge, thus either need a pump charge gate driver or gate transform for high-side N-Channel MOSFET. FWIW: Generally I use high-side gate transformers and direct low side, as I never had any good sucess using pumped charged h-bridge drivers for high-power applications. Also its been a while since I did any power switching projects so my information & my memory on all this may be a bit fuzzy.

@@guytech7310 Today i did some measurements and i concluded that the signal from my ne555p timer had weird shape. The pulse width wasn't 50% so one bridge side was working more than the other and also the waveform was a square wave but with a big notch on the beginning. I will try to make a different version of the signal generator.

Fancy seeing you here. I love your channel. 😄

Do you mean exploding transistors? :). Ever since I started using snubber networks, they don't explode.

@@DiodeGoneWild Mostly MOSFETs turning back on when they shouldn't be, ends up pinging the primary and leading to extra heating and upsetting my lossless snubbers ( :

But His Accent is 1000% better than great Scott

and his cat is really funny

@@PCbolt17 yeah.. GS uses autotune his voice. Listen to it using earphones with full volume and it becomes annoying

No way this slav is as annnoying and as bad at soldering like that overinflated ego lil' nonhetero

I started with some books in the 1990s, then the internet of course, and my experiments and experiences, reverse engineering of countless devices, and I put all the fragments together...

@@DiodeGoneWild you are a genius!! From "Samir"🤗🤗

@@DiodeGoneWild Did you ever edit anything on Wikipedia?

@@playingwithplasma2440 where is Samir?

@@matid8453 Probably a man is using the name of a woman.

10 times the voltage won't make 10 times the bump. The Miller capacitance goes down with CE voltage.

@@DiodeGoneWild Interesting. Presumably due the depletion region getting wider for a BJT? Would that still be true for for a MOSFET and is it also true for an IGBT?

I'm even thinking about IRS2453D. But let's finish the halfbridge first.

I see you everywhere.