Isolation/high side driver (beyond bootstrap caps) techniques, like gate transformers, problems at different PWM% etc. Using simple totem pole drivers - WAY CHEAPER than IC gate drivers etc. Things to look out for like Drain capacitive coupling when switch very fast (i.e. not destroying the gate). I know all these things, but, looking at what your channels shows, it might be of interest to your other viewers.
thanks. Yeah I thought about isolated drivers but decided it requires its own dedicated video. Interesting, I never actually thought about a totem pole being cheaper that a driver, I will look into it. As for the drain-gate capacitive coupling I thought this was taken care of by the protection on the gate...resistor plus zenner. What else can you do to stop drain-gate capacitance problem?@@stevenbliss989
Yeah the zener will take care of it, IF NEEDED. Zener or TVS diodes are the only practical ways I know of. But as I said only in very highspeed situations, and even then it may not be an issue given the capacitance voltage divider of the gate between the source and drain where the source capacitance always dominates. I really should have said, MY MISTAKE HERE -SORRY, ...very high speed and VERY HIGH VOLTAGE (on the drain, at hundreds of volts) switching situations.
I came across an ULTRA FAST gate driver IC using high voltages on the GATE, many years ago (sorrry do not remember the model/manufacturer - and it was EXPENSIVE!). What it did was switch up using 30V but only until the gate reaches 12v, then it stop FAST to prevent gate damage. It does this to charge the gate capacitor very fast (remember the gate has a 1-3 ohm resistance). It also used a -ve (user set up to -25V) to turn the gate of VERY FAST. The turn on current was not fixed, and was in fact set to max Charge, NOT drive current, using a programming resistor. Yep it was a very exotic, ...and VERY expensive (something like US$20's I think?). Anyway, if you may have come across it in your studies.
Brilliant content! Are you interested in testing out some of the high voltage mosfets from China? Their rds on is pretty good but the gate charge is much higher and that requires a strong gate drive design. Might be something you're interested in?
Well its job is to pull down the mosfet gate, so why add extra resistance to that by putting it on the left of the gate drive resistor. But since the gate drive resistor is usually so small it doesn't make much difference.
It’s a little more complicated, yes FETs are voltages device’s but a typical power FET has around 2000pf of Gate capacitance, it takes current to charge the gate capacitance to reach a certain voltage level, so turning on An FET is a Hybrid operation, gate capacitance charging ( Current, dynamic current charging, and a static voltage once the Gate Capacitor is charged, once the gate capacitor is charged only a very small continuous gate current would be needed to Compensate any gate capacitor leakage current. This is why analog electronics is so much fun!
You know the standard joule thief circuits? These circuit use current applied by the trigger coil to turn the transistor on and off. These circuits don't work with mosfets unless you use very high turns for the trigger coil
To add to the Complexity of designing Power FET circuits: the faster you turn them on the less time they spend in the resistive reign, less resistive losses and less heat generated, but there is a faster rising edge producing more High Frequency EMI. Some of LTC’s switching regulators give the ability to control the rise time of the FET Turning on! Another trade trade off to be made between EMI and temperature rise!
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
Excellent presentation
Very useful and clearly presented.
Thanks
Great basics, but I was hoping for more :( ...THUMBS UP! :)
thanks. Out of interest, what else were you looking for?
Isolation/high side driver (beyond bootstrap caps) techniques, like gate transformers, problems at different PWM% etc.
Using simple totem pole drivers - WAY CHEAPER than IC gate drivers etc.
Things to look out for like Drain capacitive coupling when switch very fast (i.e. not destroying the gate).
I know all these things, but, looking at what your channels shows, it might be of interest to your other viewers.
thanks. Yeah I thought about isolated drivers but decided it requires its own dedicated video. Interesting, I never actually thought about a totem pole being cheaper that a driver, I will look into it. As for the drain-gate capacitive coupling I thought this was taken care of by the protection on the gate...resistor plus zenner. What else can you do to stop drain-gate capacitance problem?@@stevenbliss989
Yeah the zener will take care of it, IF NEEDED. Zener or TVS diodes are the only practical ways I know of. But as I said only in very highspeed situations, and even then it may not be an issue given the capacitance voltage divider of the gate between the source and drain where the source capacitance always dominates. I really should have said, MY MISTAKE HERE -SORRY, ...very high speed and VERY HIGH VOLTAGE (on the drain, at hundreds of volts) switching situations.
I came across an ULTRA FAST gate driver IC using high voltages on the GATE, many years ago (sorrry do not remember the model/manufacturer - and it was EXPENSIVE!).
What it did was switch up using 30V but only until the gate reaches 12v, then it stop FAST to prevent gate damage. It does this to charge the gate capacitor very fast (remember the gate has a 1-3 ohm resistance). It also used a -ve (user set up to -25V) to turn the gate of VERY FAST. The turn on current was not fixed, and was in fact set to max Charge, NOT drive current, using a programming resistor. Yep it was a very exotic, ...and VERY expensive (something like US$20's I think?). Anyway, if you may have come across it in your studies.
Brilliant content! Are you interested in testing out some of the high voltage mosfets from China? Their rds on is pretty good but the gate charge is much higher and that requires a strong gate drive design. Might be something you're interested in?
Thanks for the quick drill down. Short and sweet. These were all N channel correct?
thanks. Yes, everything I said refers to n channel mosfets.
Just out of curiosity : is it not preferable to have the 10k pull down resistor on the left side of the gate resistor R? Great video. Thanks
Well its job is to pull down the mosfet gate, so why add extra resistance to that by putting it on the left of the gate drive resistor.
But since the gate drive resistor is usually so small it doesn't make much difference.
Mosfet doesn't use current, it's a voltage device
2:00 He literally explains that the gate of a FET is capacitive. Show me a capacitor that doesn't require current to charge it.
It’s a little more complicated, yes FETs are voltages device’s but a typical power FET has around 2000pf of Gate capacitance, it takes current to charge the gate capacitance to reach a certain voltage level, so turning on An FET is a Hybrid operation, gate capacitance charging ( Current, dynamic current charging, and a static voltage once the Gate Capacitor is charged, once the gate capacitor is charged only a very small continuous gate current would be needed to Compensate any gate capacitor leakage current.
This is why analog electronics is so much fun!
You know the standard joule thief circuits? These circuit use current applied by the trigger coil to turn the transistor on and off. These circuits don't work with mosfets unless you use very high turns for the trigger coil
When I worked at HP Labs we used a transistor to drive the gate current to turn on the Power FETS quickly for the motors in our Robots.
To add to the Complexity of designing Power FET circuits: the faster you turn them on the less time they spend in the resistive reign, less resistive losses and less heat generated, but there is a faster rising edge producing more High Frequency EMI. Some of LTC’s switching regulators give the ability to control the rise time of the FET Turning on! Another trade trade off to be made between EMI and temperature rise!