# What precisely is $R_{ds(on)}$?

I keep getting conflicting answers on this and it's absolutely killing me. Some say $R_{ds(on)}$ is the contact resistance in the FET (for instance, make many FETs of varying channel lengths, turn them full on and measure the resistance and the residual resistance at zero channel length is $R_{ds(on)}$). Others say it's the slope of the $I_{ds}$ $V_{ds}$ curve in the linear regime when the FET has a typical gate voltage applied and is at a sensible temperature.

I also can't honestly find an answer on whether you use a FET as a switch in the linear or in the saturation regime!

• R_ds is the linear relationship between V and I. R_ds(on) is a specific value of R_ds when the fet is in saturation. You're just getting confused because you think R_ds = R_ds(on) – I. Wolfe Feb 3 '15 at 20:02

To answer your second (sort-of) question first, you would normally operate a MOSFET in the ohmic (aka linear or triode) mode when it is used as a switch. The voltage drop across the MOSFET is more-or-less proportional to the drain-source current.

In the saturation region, the current through the MOSFET is more-or-less independent of the drain-to-source voltage- it 'looks' like a constant current source or sink once the voltage across it is great enough. Image from Wikipedia.

Rds(on) is the slope of the above curve (well to the left of the red curve). As you can see, the slope depends on Vgs. It also depends on temperature, which is typically stated to be Tj = 25°C.

Here is how a typical small high-performance MOSFET behaves:

The Rds(on) resistance is fairly constant up to several amperes. It also has a strong temperature dependency, so at 150°C it may be 70% higher than at room temperature. And, like most specs, the typical curves don't represent the guaranteed limits. The difference between typical and the worst-case (hot) Rds(on) may be as much as 2.5:1 so it is better to be conservative when specifying parts.

It's the resistance of the channel from drain to source when the FET is switched on. There may be several different values for R_ds mentioned in the datasheet for different operating regions of the FET(cutoff, ohmic, saturation,...). And as helloworld922 commented; resistance is the relation of voltage to current.

Datasheets always use a lot of abbreviations, but then don't describe them. This is because there is a standard associated with most of the measurements. Each has a description and either you're expected to know it, or the engineers building datasheet are so marinated in the concepts they don't realize they're not speaking English.

Texas Instruments, and I'm sure others, have published a translation guide. It's more like a tourist travel book, it gives most of the basic translations, enough for the more complex ones to at least be guessed at with some confidence.

For instance, the closest matching translation for your question:

$r_{on}$ - On-State Resistance

JEDEC – The resistance between specified terminals with input conditions applied that, according to the product specification, will establish minimum resistance (the on-state) between those terminals.

TI – The resistance measured across the channel drain and source (or input and output) of a bus-switch device.

Both the JDEC standard definition and TI's interpretation of that definition are given.