Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

On resistance and High switching speed of FET

Status
Not open for further replies.
Hi all,

I have seen the following features in one of the FET specification sheet as

Low On Resistance
High switching speed.

Could somebody please explain me what is the advantage of low on resistance, what is the general range of this resistance? Also please explain me in simple terms the switching speed the parameters involved in switching speed and how to measure from the data sheet?

Thanks in advance,
Regards,
Satya
 
Unless you are using a mosfet in it's linear mode, you are using it as a switch. The on resistance becomes a point of power loss. Low on resistance mosfets are a few tens of milliohms, but some are down around a milliohm. Many of the new mosfet packages have been specifically designed to add as little packaging resistance as possible.

Switching speed is also related to on resistance. A mosfet cannot turn on or off instantly. They change from their low on resistance to almost infinate resistance. The faster the transition time, the less power is wasted during the change. Switching speed is based on turn on - turn off times, and are usually measured in nanoseconds.
 
Could somebody please explain me what is the advantage of low on resistance,
The lower the On resistance, the less power dissipated in the FET while it is on. The less power dissipated in the FET, the more efficient your design is (less wasted heat), and less you have to worry about heat sinking and/or active cooling.

what is the general range of this resistance?
I don't know, look up a few datasheets to get an idea.


Also please explain me in simple terms the switching speed the parameters involved in switching speed and
While the FET is in transition from OFF to ON or vise versa, it is neither fully on nor fully off. During this period of time, massive amounts of power are dissipated in the fet. So, you want this switching time to be as fast as possible so that these massive amounts of wasted power are minimized. Factors that contribute to switching time are mainly gate capacitance, and whatever circuit resistance present at the gate. The gate of a FET can be thought of as a capacitor, and it takes time to charge up this capacitance before the FET turns on. There is also miller plateau but I don't really think it's important to go into.


how to measure from the data sheet?

The only thing you can measure from a datasheet is lies. "datasheet coolaid" - don't drink it. Datasheets often give fantastical ratings based on theoretical things and hairbrained calculations. The biggest one I'm aware of is the amperage ratings. For example, I've seen FETs in TO-220 package that advertise the maximum current as being in excess of 120A, yet it's in a package with legs smaller than toothpicks. Common sense applies; if it sounds too good to be true, it probably is.
 
...................................
The only thing you can measure from a datasheet is lies. "datasheet coolaid" - don't drink it. Datasheets often give fantastical ratings based on theoretical things and hairbrained calculations. The biggest one I'm aware of is the amperage ratings. For example, I've seen FETs in TO-220 package that advertise the maximum current as being in excess of 120A, yet it's in a package with legs smaller than toothpicks. Common sense applies; if it sounds too good to be true, it probably is.
I believe that's a little harsh. ;)

Generally the data sheet is where you go to get all the desired device operating characteristics. Manufacturers (at least the reputable ones) normally do a fairly good job of characterizing their devices so that you can properly use them.

Often a problem occurs in incorrectly using a device at the "Absolute Maximum Ratings" values. Just be aware that you don't want to operate at or near those ratings. Typically you should be at no more the 50-75% of the maximums for good reliability.
 
I once asked an engineer from one of the big mosfet companies about how they come up with the current rating. He told me that it was a theoretical calculation based on the current that would run the die at 175C when the case is held at 25C on an infinite heatsink.

So, it is a totally useless number for any kind of practical design.
 
I once asked an engineer from one of the big mosfet companies about how they come up with the current rating. He told me that it was a theoretical calculation based on the current that would run the die at 175C when the case is held at 25C on an infinite heatsink.

So, it is a totally useless number for any kind of practical design.
Hence the reason to significantly derate the maximum ratings, as I mentioned. It still gives a relative rating with which to compare transistors.

So, out of curiosity, how would you determine the maximum current rating? :confused:
 
I believe that's a little harsh. ;)

Generally the data sheet is where you go to get all the desired device operating characteristics. Manufacturers (at least the reputable ones) normally do a fairly good job of characterizing their devices so that you can properly use them.

Often a problem occurs in incorrectly using a device at the "Absolute Maximum Ratings" values. Just be aware that you don't want to operate at or near those ratings. Typically you should be at no more the 50-75% of the maximums for good reliability.

+1 what ChrisP58 said. Also, the ratings they show pertain only to the silicon, and totally ignore the limitations and parasitic properties of the package. The datasheet specs are out to lunch when you try to apply them to real world. For example, IFR1405PBF: datasheet says 169A max @25C continuous, 118A max @ 100C continuous, and max pulsed current 680A. It's in a TO-220AB package. It says right there on the datasheet in the fine print that package limitation is 75A. Really? 75A?. My welder operates around 75A and it needs #2 thick copper cable to carry that. Am I really to believe that the cute little leg of a TO-220AB fet can carry the same amperage? I'm not comfortable with passing more than say 10A through it, and it would probably be pretty darn hot even at that. So I say datasheets are off by several orders of magnitude, and the thumbrule of 50-75% of max is not even a safe bet in some instances. My goal is only to make OP aware of some of the flagrant fouls in datasheets, and to be careful; I know they are not all this bad.
Hence the reason to significantly derate the maximum ratings, as I mentioned. It still gives a relative rating with which to compare transistors.

So, out of curiosity, how would you determine the maximum current rating? :confused:

I haven't gotten far enough to answer this from experience, but what I would probably do it test it out. Take that IRF1405PBF and put 10A through it in the worst (heat) anticipated conditions with the intended heat sink and monitor the temp of the package, heat sink, and legs. if it looks like it can take more, then give it more, until it blows up, and then go back a few amps and call that the *real world* max.
 
I choose a mosfet based on calculated temperature rise. If it is a static switch, then I^{2}R, and the thermal characteristics of the case and environment, are about all you need. If it is a PWM circuit, then you need to factor switching looses in as well.

If you choose a mosfet by current, then you will probably end up running it hot and will need a heatsink. I often find that I can spend a little more on a low loss mosfet, or put two in parallel, and save more than that cost by not needing a heatsink.
 
Last edited:
+1 what ChrisP58 said. Also, the ratings they show pertain only to the silicon, and totally ignore the limitations and parasitic properties of the package. The datasheet specs are out to lunch when you try to apply them to real world. For example, IFR1405PBF: datasheet says 169A max @25C continuous, 118A max @ 100C continuous, and max pulsed current 680A. It's in a TO-220AB package. It says right there on the datasheet in the fine print that package limitation is 75A. Really? 75A?. My welder operates around 75A and it needs #2 thick copper cable to carry that. Am I really to believe that the cute little leg of a TO-220AB fet can carry the same amperage? I'm not comfortable with passing more than say 10A through it, and it would probably be pretty darn hot even at that. So I say datasheets are off by several orders of magnitude, and the thumbrule of 50-75% of max is not even a safe bet in some instances. My goal is only to make OP aware of some of the flagrant fouls in datasheets, and to be careful; I know they are not all this bad.
.....
You've made a good point about the MOSFET maximum current ratings. Some are obviously seriously over-optimistic. So you really need to select the current rating based upon the transistor package power rating, the available heat sink, and the ON resistance of the transistor.

But I don't think you should give a blanket condemnation to data sheets as being all lies because of that. They still have a lot of good, reasonably valid information about the device characteristics which are useful for using the device as long as reasonable derating factors are used.
 
But I don't think you should give a blanket condemnation to data sheets as being all lies because of that. They still have a lot of good, reasonably valid information about the device characteristics which are useful for using the device as long as reasonable derating factors are used.
Ok, I'll concede, and admit that I don't have enough experience to have seen these attributes proven in real life.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top