In DIY driver, you can add NPN transistor as an emmitter follower to boost current pumped into the gate, so that rise time decreases without lowering resistor value much. There is one more thing you have to be careful about, when switching on the mosfet discharging the gate, it forces negative G-S voltage on driven mosfet gate, with high drain voltage you can easily exceed max G-S voltage and kill the transistor. Isolated drivers can be built around an optocoupler, but they are not that fast and have a few pitfalls as well. So the best option is just sticking to dedicated ICs.
High side means that the transistor is placed between the load and the positive supply. Low side instead is when the transistor is connected to ground. In this configuration we have the disadvantage that the load is connected positive even when off.
I've never used traditional bootstrap circuits for MOSFET driving, but 50kHz is a relatively low frequency and if I had to guess I'd say that the simple approach of using bootstrap circuits could work. But I don't know your specific situation. The switching won't be particularly fast because (if I'm thinking about the same circuit as you) the gate is switched on through a pull-up resistor, so your MOSFET will heat up more than with a dedicated driver. Good luck!
@@5VLogic yeah I am working on my FYP of designing an EPS for cubesat, the first step is to build an mppt which uses buck converter so I want the mosfet inside buck converter to switch on atleast 50khz for minimum power losses, the big problem is it is not giving me the required output voltage because the mosfet is not working properly. That's why i needed to know. BTW thanks for quick reply and help! 😍
I think it'd be useful also to take a bit of a wider step back and answer why we'd use an n-FET as a high-side switch, given all this extra complexity, rather than use what seems on paper to be a far simpler solution of just using a p-FET instead.
Thanks for the comment. It's true, in fact, actually two assumptions were made in the video: that a P channel FET can't be used and that a low side switch is also not employable. For lower power applications the P FET is the simple solution. Higher performance devices will try to avoid them because of worse characteristics (like Rds) and their overall scarcity which limits choice.
@@5VLogic If you don't want us a P-Channel, also you must consider connect the load to the DRAIN and the SOURCE to GND. This way you apply 5V to gate and the switch ON happen!
In simple words, it's not very practical to use a MOSFET which will turn itself on when gate is grounded because it's usual practice to keep things pulled to ground by default. Also at voltages more than VGS rating, you need a negative voltage regulation to supply a pmos with safe gate voltage. So for application below 15v, P MOS can easily be used, but not beyond that. Inefficiencies are another reason
I think the catch is the difference between t d(off) and t d(on). The t d(on) is always different from t d(off). Usually td(off) is longer which makes both fets overlap for something like 20ns.
I really don’t see how you get 24 V on the top of the capacitor. Seems to me that you were simply putting ~12 V on both sides when the upper Mosfet turns on.
Most people it would be B, but in my case, I reckon D, because I am super talented like that. :D EDIT: I never would have guessed C, so I learned something new today :)
P MOSFETs simply suck. They're slower, waste more power and bit expensive. Also they need to be constantly pulled up to keep them off. Driving them requires gate voltage 12-15 v lower than supply voltage, which is again a problem because when you're using them at say, 50v, you need 45v signal on gate to turn them on safely. So P channel MOSFETs are not convenient enough.
@@swapnilkumar9363all those things use to be true but with newer parts you can get now they have greatly improved to where the difference is unnoticeable until you go to 100s of amps or volts. Where in the past the Rds difference was often 10-100x higher there is usually only a 2-4x now on the newer parts which is within the difference you see from different manufacturers for the same generation parts
@@michaelcummings7246 if P channel have gotten better, N channel's have too. No matter how better P channel MOSFET gets, it will never be close to N channel because laws of physics aren't going to change. N channel has electrons as charge carriers and P channel has holes, so there will always be more losses due to electron - home recombination.
@@swapnilkumar9363 you're complaining about all the wrong things and these days Pch FETs are almost as good an N for practical applications like this, and the selection is massive.
@@VEC7ORlt I am a relative beginner and yes, yours was/is my response on one level, but then I was grateful at having had an inadvertent master tutorial on the problems involved, which appeals to my curiosity. But not every beginner would be able to critique the video, I get it.
@@vincentwhite7693 start reading data sheets and application notes, there are plenty of those, varying from simple to PhD grade, and they are way better than these vids.
In DIY driver, you can add NPN transistor as an emmitter follower to boost current pumped into the gate, so that rise time decreases without lowering resistor value much. There is one more thing you have to be careful about, when switching on the mosfet discharging the gate, it forces negative G-S voltage on driven mosfet gate, with high drain voltage you can easily exceed max G-S voltage and kill the transistor. Isolated drivers can be built around an optocoupler, but they are not that fast and have a few pitfalls as well. So the best option is just sticking to dedicated ICs.
So if i just connected to the bulb to +12 volts and the source to ground will it fully turn on ?
Sir what is meant by high side and low side mosfet
High side means that the transistor is placed between the load and the positive supply.
Low side instead is when the transistor is connected to ground. In this configuration we have the disadvantage that the load is connected positive even when off.
can i use boostrap for 50Khz frequency? or should i use the IC?
I've never used traditional bootstrap circuits for MOSFET driving, but 50kHz is a relatively low frequency and if I had to guess I'd say that the simple approach of using bootstrap circuits could work.
But I don't know your specific situation. The switching won't be particularly fast because (if I'm thinking about the same circuit as you) the gate is switched on through a pull-up resistor, so your MOSFET will heat up more than with a dedicated driver.
Good luck!
@@5VLogic yeah I am working on my FYP of designing an EPS for cubesat, the first step is to build an mppt which uses buck converter so I want the mosfet inside buck converter to switch on atleast 50khz for minimum power losses, the big problem is it is not giving me the required output voltage because the mosfet is not working properly. That's why i needed to know. BTW thanks for quick reply and help! 😍
Well explained!
I think it'd be useful also to take a bit of a wider step back and answer why we'd use an n-FET as a high-side switch, given all this extra complexity, rather than use what seems on paper to be a far simpler solution of just using a p-FET instead.
Thanks for the comment. It's true, in fact, actually two assumptions were made in the video: that a P channel FET can't be used and that a low side switch is also not employable. For lower power applications the P FET is the simple solution. Higher performance devices will try to avoid them because of worse characteristics (like Rds) and their overall scarcity which limits choice.
@@5VLogic If you don't want us a P-Channel, also you must consider connect the load to the DRAIN and the SOURCE to GND. This way you apply 5V to gate and the switch ON happen!
@@jlsmonte Of course that is better way to do it. But sometimes we can't do that (for example in half-bridge and 2 switch forward power supply)
In simple words, it's not very practical to use a MOSFET which will turn itself on when gate is grounded because it's usual practice to keep things pulled to ground by default. Also at voltages more than VGS rating, you need a negative voltage regulation to supply a pmos with safe gate voltage. So for application below 15v, P MOS can easily be used, but not beyond that. Inefficiencies are another reason
I think the catch is the difference between t d(off) and t d(on). The t d(on) is always different from t d(off). Usually td(off) is longer which makes both fets overlap for something like 20ns.
I really don’t see how you get 24 V on the top of the capacitor. Seems to me that you were simply putting ~12 V on both sides when the upper Mosfet turns on.
You can use tlp250 or 350 it's very fast high side driver
I didn't know about those components, I just checked them out and it's a great suggestion, thanks!
Most people it would be B, but in my case, I reckon D, because I am super talented like that. :D
EDIT: I never would have guessed C, so I learned something new today :)
why not just use P mosfet
P MOSFETs simply suck. They're slower, waste more power and bit expensive. Also they need to be constantly pulled up to keep them off. Driving them requires gate voltage 12-15 v lower than supply voltage, which is again a problem because when you're using them at say, 50v, you need 45v signal on gate to turn them on safely. So P channel MOSFETs are not convenient enough.
@@swapnilkumar9363all those things use to be true but with newer parts you can get now they have greatly improved to where the difference is unnoticeable until you go to 100s of amps or volts. Where in the past the Rds difference was often 10-100x higher there is usually only a 2-4x now on the newer parts which is within the difference you see from different manufacturers for the same generation parts
@@michaelcummings7246 if P channel have gotten better, N channel's have too. No matter how better P channel MOSFET gets, it will never be close to N channel because laws of physics aren't going to change. N channel has electrons as charge carriers and P channel has holes, so there will always be more losses due to electron - home recombination.
@@swapnilkumar9363 you're complaining about all the wrong things and these days Pch FETs are almost as good an N for practical applications like this, and the selection is massive.
@@swapnilkumar9363 true, but completely misses the point - P channel is perfectly useful in most practical applications.
Whats with these useless and impractical circuits popping up everywhere?
It is valuable because if for nothing else it reinforces why things are done a certain way.
@@vincentwhite7693 its useless because noone is doing it this way, besides being a half-assed attempt at explaining it.
@@VEC7ORlt I am a relative beginner and yes, yours was/is my response on one level, but then I was grateful at having had an inadvertent master tutorial on the problems involved, which appeals to my curiosity. But not every beginner would be able to critique the video, I get it.
@@vincentwhite7693 start reading data sheets and application notes, there are plenty of those, varying from simple to PhD grade, and they are way better than these vids.