3
\$\begingroup\$

How do I calculate the V(ds) voltage loss which a MOSFET introduces into the load circuit given a specific load and V(gs) voltage?

I assumed I could use the V(ds) vs. I(d) diagram for that, but I don't find an intersection for my assumed I(d) current. Am I doing something wrong, do I misinterpret the diagram, or does it mean that the MOSFET is not usable for the task?

Here are the conditions:

The load circuit is a 12V DC circuit with a 12W load, i.e. if it was directly connected to a 12V battery it would draw 1A. For the remainder let's assume it is a 12 Ohm resistive load. The MOSFET is a IRF610 and the gate is driven by a 5V general purpose output (a µC).

The datasheet of the IRF610 gives us

  • V(ds,max) = 200 V
  • I(d,max) = 3.3 A
  • V(gs,max) = +/-20 V
  • V(gs,th) = 4.0 V
  • R(ds,on) = 1.5 Ohm @ V(gs) = 10V and I(d) = 2A.

From those specs I assume the MOSFET would switch on, if 5V were applied to the gate, as 5V > V(gs,th). Moreover it should no overload as 5 < V(gs,max), 12V < V(ds,max) and 1A < I(d,max).

In order to calculate the voltage drop over the MOSFET, I might use R(ds,on), but R(ds,on) = 1.5 Ohm is measured for the test conditions V(gs) = 10V and I(d) = 2A. In my case V(gs) is only 5V and the current is at most 1A reduced by whatever resistance the MOSFET introduces into the load path.

Anyway, I calculate R(total) = R(load) + R(ds,on) = 12 Ohm + 1.5Ohm = 13.5 Ohm and hence I get 888mA load current, 1,33V drop over the MOSFET and 10.6V remaining for the actual load.

I thought I could be more precise and had a look at the V(ds) vs. I(d) diagram (datasheet, p.3, top-left). I thought this is what this diagram is for. So I used the V(gs)=5 curve, but it never intersects with I(d)=1A. The curve is already horizontal in that area. Does this mean that the V(ds) will be "infinity" (or 12V in my case), i.e. the MOSFET blocks?

\$\endgroup\$
4
  • \$\begingroup\$ Every beginner trips over this at some point. The threshold voltage is not some magic voltage above which the MOSFET is entirely on; the threshold voltage is the voltage at which the MOSFET just barely begins to turn on, and it takes a good few volts more (exactly how much depends on other parameters, most importantly transconductance) to turn it on fully. For switching applications, you should always choose a MOSFET with a rated Rdson at a Vgs equal to or lower than the gate voltage you intend to use, as those are the voltages where it's designed to be operated. \$\endgroup\$ Commented Oct 10 at 18:27
  • \$\begingroup\$ @periblepsis No there was not a particular reason. It was the first MOSFET which my local dealer had in stock and which matches most of the obvious criteria at first glance. And then I had closer look and had my doubts. Is there any good "search engine" to find MOSFETs by desired ratings? Even Mouser is not really good. \$\endgroup\$ Commented Oct 10 at 19:00
  • \$\begingroup\$ @Hearth Thank you for the tip (using a MOSFET which operates at the designed values). See previous question: What "search engine" do you recommend? Or as you know my constraints: Do you can recommend a particular MOSFET? Through-the-hole packing is mandatory. \$\endgroup\$ Commented Oct 10 at 19:00
  • 1
    \$\begingroup\$ @user2690527 Digikey's parametric search includes drive voltage, both lowest and highest with rated Rdson. I also just have a couple FETs I use by default for different purposes; the Si2318 is good for medium power loads at 5 V drive, and the DMG2302 for low voltage but medium current loads at 3.3 V drive. \$\endgroup\$ Commented Oct 10 at 19:10

2 Answers 2

Reset to default
6
\$\begingroup\$

Look at this image from your data sheet and you'll see that the MOSFET of choice is unsuitable for a load current of 1 amp with a Vgs voltage of 5 volts: -

enter image description here

Does this mean that the V(ds) will be "infinity" (or 12V in my case)

No, it means it will be typically drawing around 300 mA and probably getting quite warm.

I assumed I could use the V(ds) vs. I(d) diagram for that, but I don't find an intersection for my assumed I(d) current. Am I doing something wrong, do I misinterpret the diagram, or does it mean that the MOSFET is not usable for the task?

Your analysis is correct and I'd choose a better MOSFET if I were you. If you operate the gate at a higher volt there would still be a lot of power dissipation. There are plenty of MOSFETs that will fit the bill.

Load line analysis: -

enter image description here

  • I've traced over the MOSFET characteristic in red that applies to Vgs = 5 volts
  • I've drawn a blue line that represents the load 12 Ω
  • Where they cross is 3.6 volts. That's the operating point
  • So, when Vgs is 5 volts, the drain voltage will be 3.6 volts at 300 mA
\$\endgroup\$
8
  • \$\begingroup\$ How did you come up with the conclusion that it will draw 300mA and get quite warm? Which values from the the datasheet do I have to look at to see that? Maybe you could extend your answer by that, because it would be really helpful for understanding how to draw the correct conclusions from a datasheet. \$\endgroup\$ Commented Oct 10 at 18:54
  • \$\begingroup\$ @user2690527 look at the x-axis and reckon on the MOSFET DS voltage being somewhere between 1 volt and 10 volt and look at the current along the y-axis where I wrote 300 mA in light blue. The question is can the DS voltage ever, ever be less than 1 volt given that the load is 12 ohm? That answer is unambiguously never. \$\endgroup\$ Commented Oct 10 at 19:44
  • 2
    \$\begingroup\$ @Andyaka Do you mean the IRLZ44N? The IRFZ44N's threshold voltage is only guaranteed to be below 4 V, so 5 V isn't really enough to fully turn on a worst-case device. The IRLZ44N is the logic-level equivalent, designed to be driven at 5 V. \$\endgroup\$ Commented Oct 10 at 20:52
  • 2
    \$\begingroup\$ @Hearth bingo, I got the number wrong. Thanks. I mean the IRLZ44N. \$\endgroup\$ Commented Oct 10 at 20:57
  • 1
    \$\begingroup\$ @Andyaka A lot of effort went into this answer, I wish I could +2 \$\endgroup\$ Commented Oct 11 at 5:00
2
\$\begingroup\$

To help understand the MOSFET's operation, here's the LTspice sim of a simple circuit to control the load's current with a MOSFET close to the IRLZ44N's characteristics (with a Vgs nominal threshold of 1.9V):

Note that the MOSFET goes from full OFF to full ON over a Vgs range of about 0.2V from 1.9V to 2.1V.

The MOSFET dissipated power (red trace) goes from zero to 3W maximum, where the voltage across the transistor equals the voltage across the load (as per the Maximum Power Theorem), and then goes down to near zero again when the transistor is fully on.

enter image description here

\$\endgroup\$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.