Question about MOSFET Saturation region

[Dextrine]

Ok, so I've read around and have become quite confused with, not only the terminology about whether a device is in linear or saturation or ohmic or active region, but now my whole concept of mosfets is on the brink of breaking down!

This is my understanding.

In the attached picture, the maximum current that can flow is 1A. The higher Vgs, the higher the POSSIBLE current (if the load were smaller for example). Therefore, if I have a high Vgs, then the current through the mosfets will vary as a function of my input voltage which is NOT what I want??

I know I need the resistance across the mosfets to be as small as possible to minimize conduction losses, it's just confusing where exactly that happens. I thought the Rdson is the smallest, the mosfet is fully on, in the saturated (active) region.

However, this makes less and less sense to me as time goes on. If someone could give any definitive answer as to how the mosfets in my example should be biased to give the lowest conduction loss.

Thanks for any help.

 

[Svein]

1. MOSFETs do not go into saturation.
2. MOSFETs act a voltage-controlled resistors. Look for R_DS(on) in the specification sheet. You will get a set of values, corresponding to different values of V_GS

 

[Dextrine]

Ok, I think I figured it out. If you want to use a MOSFET as a switch, you must have VGS be as high as possible in order to keep VDS as low as possible and thus minimize conduction losses. This is called the triode, or ohmic, or linear region.

Operating the device as a voltage controlled current source is NOT what you want if you want to use it as a switch. This is called the active, or saturation region.

Can anyone confirm this?

and also, Svein, I'm not sure what you mean by mosfets do not go into saturation, there is always a clearly marked saturation region.

 

[phyzguy]

1. MOSFETs do not go into saturation.

MOSFETs certainly do go into saturation. I think the thing that is confusing you is that the terminology is confusing. Look at the two attached images from Wikipedia. While they look similar, the region on the right of the MOSFET curve is called saturation, while the region on the left of the BJT curve is called saturation.

 

[Dextrine]

MOSFETs certainly do go into saturation. I think the thing that is confusing you is that the terminology is confusing. Look at the two attached images from Wikipedia. While they look similar, the region on the right of the MOSFET curve is called saturation, while the region on the left of the BJT curve is called saturation.

So would you say my second post has me understanding the subject correctly?

 

[phyzguy]

So would you say my second post has me understanding the subject correctly?

Yes, I think you have summed it up well.

 

[Svein]

Svein, I'm not sure what you mean by mosfets do not go into saturation,

I have checked the current definitions on the net, and I have seen that they have introduced "saturation" as a description of the "pentode" region of the MOSFET. However, it is very different from the saturation phenomenon in bipolar transistors. A bipolar transistor in saturation has a V_CE that is independent of the base current, a MOSFET in "saturation" has an I_DS that is independent of V_DS.

A "rule of thumb" might be: A bipolar transistor in saturation acts as a constant voltage generator, a MOSFET in "saturation" acts as a constant current generator.

 

[phyzguy]

I have checked the current definitions on the net, and I have seen that they have introduced "saturation" as a description of the "pentode" region of the MOSFET. However, it is very different from the saturation phenomenon in bipolar transistors. A bipolar transistor in saturation has a V_CE that is independent of the base current, a MOSFET in "saturation" has an I_DS that is independent of V_DS.

A "rule of thumb" might be: A bipolar transistor in saturation acts as a constant voltage generator, a MOSFET in "saturation" acts as a constant current generator.

Agreed. That's why I said the terminology is confusing.