Power mosfet basics

[indulis]

...but I can tell you right now that it has a large gate capacitance so that it can be mutually coupled with the inductance of the circuit, therefore you will be able to "tune" the circuit using the correct mosfet capacitance and as larger power applications have larger inductances (due to transformers and coils in general) you need a larger gate capacitance to match, this is why the larger power mosfet have such incredibly large gate capacitances compared to control mosfets.

What are you talking about???

 

[Da$ymbolGuy]

Look guy, I dunno who you think you are but maybe you should read up a little about MOSFETs before you just start talking trash, I've been working in the Power industry for 20 years and I've used MOSFETs on electric trains and power plants before you were even a sparkle in your daddy's eye so why don't you back off and let the big boys talk

 

[indulis]

Oh sure... and your a telepathic that can post an answer without knowing what the original question was. If you have only been in the power industry for 20 years, then you have a ways to go before you have put in the time I have.... might I suggest you get a book on semiconductor physics.

Fact: Gate capacitance is a parasitic of MOSFET construction.
Fact: If you want to switch a MOSFET fast you don’t want capacitance…. I=C*dv/dt be it in a low side switch or a SMPS.... tuning, ya sure!!!

 

[Thunderchild]

I have no idea what the question is because it's been edited out for some crazy reason but I can tell you right now that it has a large gate capacitance so that it can be mutually coupled with the inductance of the circuit, therefore you will be able to "tune" the circuit using the correct mosfet capacitance and as larger power applications have larger inductances (due to transformers and coils in general) you need a larger gate capacitance to match, this is why the larger power mosfet have such incredibly large gate capacitances compared to control mosfets.

I hope this helped.

who let you out of the mental institute ?

 

[smanches]

The only thing I can think of is he's talking about the output capacitance being tuned with the load.

Other than that, as already stated, the gate capacitance is directly related to the size and shape of the channel in the mosfet. The more surface area the channel has, the larger the gate capacitance. They have been able to reduce the capacitance a good deal by using different channel shapes, like V channels.

 

[Thunderchild]

I think he just being a pillac and needs banning ! you don't talk like that on your first and second post and no mature member here is allowed to have such an attitude either.

 

[Da$ymbolGuy]

I'm just trying to help out and one of your members jumps down my throat, you can hardly blame me for getting defensive over such outlandish accusation to my professional image. I think that "indulis" punk needs a hose down to cool his jets and maybe we can have a proper discussion and you actually learn something.

 

[indulis]

...one of your members jumps down my throat, you can hardly blame me for getting defensive over such outlandish accusation to my professional image. I think that "indulis" punk needs a hose down to cool his jets and maybe we can have a proper discussion and you actually learn something.

All I asked in response to your original post was:

What are you talking about???

How exactly does that "jump down your throat"?

And you come back with:

Look guy, I dunno who you think you are but maybe you should read up a little about MOSFETs before you just start talking trash, I've been working in the Power industry for 20 years and I've used MOSFETs on electric trains and power plants before you were even a sparkle in your daddy's eye so why don't you back off and let the big boys talk

And:

I think that "indulis" punk needs a hose down to cool his jets and maybe we can have a proper discussion and you actually learn something.

"...big boys talk " & "...that "indulis" punk..." and what would you call that? And for the record, my father is 83 so you have that "sparkle in your daddy's eye" backwards.

As you imply to be so knowledgeable, why not provide some references that would suggest anything that's been posted in this thread is wrong ? Then perhaps explain to everyone what problems are associated with swithching a MOSFET on in 20nS and how gate capacitance plays a roll in trying to do that...

 

[mneary]

I have no idea what the question is because it's been edited out for some crazy reason but I can tell you right now that it has a large gate capacitance so that it can be mutually coupled with the inductance of the circuit, therefore you will be able to "tune" the circuit using the correct mosfet capacitance and as larger power applications have larger inductances (due to transformers and coils in general) you need a larger gate capacitance to match, this is why the larger power mosfet have such incredibly large gate capacitances compared to control mosfets.

I hope this helped.

It didn't help because it's incorrect.

The internal gate capacitance is not intentionally designed into power MOSFETs. It is an outcome of large gate areas placed on thin insulating barriers. Large gate areas are the result of wide channels created to reduce resistance.

The answers given before your arrival are correct.

 

[Da$ymbolGuy]

Tell me where I said anything was wrong in this thread and I'll go right ahead or how about you stop trying to instigate something here like some Internet tough guy with a chip on your shoulder mate

 

[smanches]

I have no idea what the question is because it's been edited out for some crazy reason but I can tell you right now that it has a large gate capacitance so that it can be mutually coupled with the inductance of the circuit, therefore you will be able to "tune" the circuit using the correct mosfet capacitance and as larger power applications have larger inductances (due to transformers and coils in general) you need a larger gate capacitance to match, this is why the larger power mosfet have such incredibly large gate capacitances compared to control mosfets.

I would just like to know the references to this information. I have been studying mosfets for years now and have never heard mention of this once. I don't want to be missing anything. I can't even imagine how the gate capacitance and load inductance even interact with each other. Please explain further.

 

[audioguru]

I think that the rude nOOby uses Mosfets to power toy trains and uses an inductance at the gate to tune with the gate capacitance for some strange reason.
A moderator should explain to him how Mosfets are made and how they work then ban him.

 

[ljcox]

So with everythign considered now why is the MOSFETS chosen over the regular bipolar transistors? Or is there actually a situationw here you would use one over the mosfet for a power switch?

Power BJTs need considerable current to saturate them. This complicates the situation as you need at least one other BJT to drive the power BJT and a significant amount of power is consumed.

Whereas a MOSFET only needs energy initially in order to switch it on and off rapidly.

The other option is the IGBT. It combines the advantages of the MOSFET ie. very high gate resistance and the low saturation voltage of the BJT.

 

[smanches]

So where would you want to use BJTs instead of MOSFETs?

 

[Claude Abraham]

The gate to source capacitance is the mechanism by which the device operates. Of course it is desirable to minimize Cgs, but it is the reason FETs work. Increasing the area results in larger Cgs w/ smaller Rdson. You can trade 1 for the other. I usually end up with a FET whose Rdson is not the lowest for a given footprint. I switch anywhere from 150 kHz to 450 kHz. Switching losses are usually greater than conduction loss. Hence the lowest Rdson device is not what I use.

But I wouldn't call Cgs a "parasitic" which we incur due to construction. The MOS gate structure is intentionally designed to be capacitive in order to transport charge across an insulated barrier. "Parasitic" implies unintended. The Cgs is intended, albeit we sure wish we could get the same performance with lower Cgs. So I have to design my gate driver networks with enough current drive to switch the gate fast enough to manage losses, keeping them small. The FET must remain cool. Make sense?

 

[mneary]

So where would you want to use BJTs instead of MOSFETs?

I would use a BJT in a motor drive if the power was extremely low (<30V @ <300mA) and cost was critical. I also would look at a BJT if my available drive voltage was 5V or less, since logic level MOSFETs are even more expensive. At higher RF frequencies, the high capacitance of a MOSFET gate is larger and more variable than a BJT base; and is very hard to match.

 

[mneary]

Hi Everyone, thankyou for your help

Ssam87, please put your original question back into post #1 of this thread.

In the future, the answers without the question will make little sense to people who come here. Even in the present, replies are already taking the form of guessing what the original question was.

 

[ljcox]

So where would you want to use BJTs instead of MOSFETs?

Each case has to be considered based on factors such as cost, heat dissipation, frequency, etc.

I would use BJTs for relatively low current switching, say < 2 Amp. A BJT switching 2A needs a base current of 200 mA. For example, if you are driving an NPN BJT from a 5 Volt device, the device must be able to source at least 200 mA. Otherwise, you need a buffer of some kind.

The base resistor will dissipate about 4V * 0.2A = 0.8 Watt.

 

[ljcox]

Tell me where I said anything was wrong in this thread and I'll go right ahead or how about you stop trying to instigate something here like some Internet tough guy with a chip on your shoulder mate

The problem is that no-one understood your point.

It may be valid, but how are we to know unless you provide more info.?

The question asked by indulis could have been a little more diplomatic, but all he was asking was for more info.

There is no point in becoming defensive and abusive.