How to Test a MOSFET Transistor with a Multimeter
As an Amazon Associate we earn from qualifying purchases.
How to test a MOSFET transistor with a multimeter comes down to three checks on a loose part: the gate should not conduct to the drain or source, the body diode between source and drain should conduct one way only, and the channel should stay off until the gate is charged. Mains circuits can kill, so this guide covers loose parts and powered-down boards, never live wiring. Panel or service work belongs to a licensed electrician, and your meter’s CAT rating must suit any circuit it touches.
Quick answer
- Take the part out of the circuit and use diode test mode. Fluke calls Diode Test mode “almost always the best approach” for junctions.
- Gate to drain and gate to source should read OL in both directions, because the gate sits behind an insulating oxide layer.
- With the gate shorted to the source, the source-to-drain path should read like a diode one way and OL the other way.
- A pass shows an intact gate and a working body diode. It does not measure threshold voltage, on-resistance or voltage rating.
Electrical safety before you start
Work on mains circuits can kill. The Klein MM400 manual says: “Failure to observe these warnings can result in severe injury or death.” It also urges caution above 25 V AC RMS or 60 V DC, because such voltages pose a shock hazard. This guide gives no steps for live boards, outlets, panels or service wiring. If your part sits on a board that plugs into the wall, unplug the board first, and leave any power-supply section to a qualified technician or a licensed electrician.
Stored charge is the second hazard. Fluke notes that voltage may be present in a circuit because of charged capacitors, and the Fluke 287/289 manual says to disconnect circuit power and discharge all high-voltage capacitors before testing diodes. Your meter’s CAT rating comes from its manual or listing; the Klein MM400 manual, for example, states “Safety Rating: CAT III 600V”. A meter whose paperwork lists no CAT rating should not be assumed fit for household mains.
Static is the hazard to the part. A University of Washington lab handbook says MOSFETs can be very static sensitive. It advises keeping the leads in black conductive foam whenever possible and touching a grounded object before handling the part.
What the meter is actually checking
In Kuphaldt’s open textbook the MOSFET gate is a metal or polysilicon layer on a silicon dioxide insulator. A positive bias on it forms a channel between source and drain, and a normally off MOSFET is called an enhancement mode device. The gate draws no continuous current, only a surge to charge its capacitance, and a Case Western lab sheet calls the gate current practically zero.
That gives a meter three things to see. The gate is insulated, so a good gate shows no conduction to either other pin. The channel is off at zero gate bias, so it should not conduct. And both datasheets we opened draw an integral reverse p-n junction diode between source and drain. One contributor to a repair FAQ places this diode’s anode at the source and its cathode at the drain in an N-channel part, and the University of Washington handbook finds the diode on a 2N7000 passing current from source to drain.
Read the datasheet before you probe
Pin order lives on the datasheet, not on the part. The Infineon IRFZ44N sheet labels its TO-220 pins G, D, S. The Vishay IRF540 sheet labels the part an N-channel MOSFET and lists its thermal resistance as “Maximum junction-to-case (drain)”, which suggests the metal tab is the drain. That is our reading of the label, so confirm it on your sheet. The last two rows of the table are our own arithmetic.
| Item | IRFZ44N (Infineon) | IRF540 (Vishay) |
|---|---|---|
| Drain-source voltage rating | 55 V | 100 V |
| Continuous drain current at 25 °C case | 49 A | 28 A |
| Maximum on-resistance at VGS = 10 V | 17.5 mΩ at 25 A | 0.077 Ω at 17 A |
| Gate threshold voltage at 250 µA | 2.0 V min, 4.0 V max | 2.0 V min, 4.0 V max |
| Gate leakage, maximum | 100 nA at 20 V | 100 nA at 20 V |
| Body diode forward voltage, maximum | 1.3 V at 25 A | 2.5 V at 28 A |
| Sheet test current ÷ 1.5 mA meter current | 25 ÷ 0.0015 = about 16,700 | 28 ÷ 0.0015 = about 18,700 |
| Gate resistance implied by leakage (20 V ÷ 100 nA) | at least 200 MΩ | at least 200 MΩ |
Three things follow. First, 1.5 mA is the maximum diode-test current in the Klein MM400 manual, so the meter drives the body diode with roughly 17,000 to 19,000 times less current than the sheets use. The display will not match 1.3 V or 2.5 V; Kuphaldt says diode-check readings run low “because the current provided by the meter is of trivial proportions”. Compare your reading with a known-good part of the same type.
Second, a gate leakage limit of 100 nA at 20 V works out to at least 200 MΩ, and the Klein resistance table tops out at 40.00 MΩ, so a healthy gate should read OL on every range. Third, the on-resistance values of 0.0175 Ω and 0.077 Ω are below one display step on the Klein’s lowest range (400.0 Ω, 0.1 Ω resolution), and its stated accuracy adds 5 digits, or 0.5 Ω. A multimeter cannot confirm either value, which the sheets also state at VGS = 10 V.
Can the meter turn the MOSFET on?
That depends on the meter’s test voltage against the gate threshold. The University of Washington handbook warns that a resistance-mode test voltage “may be insufficient to turn on a pn-junction”. Kuphaldt adds that some diode-check meters use under 0.3 V in resistance mode. The last column is our own subtraction against the 2.0 V to 4.0 V window.
| Meter | Mode | Stated figure | Against the 2.0 V to 4.0 V window |
|---|---|---|---|
| Klein MM400 | Diode test | 3.0 V DC open circuit, 1.5 mA max | 1.0 V above the minimum, 1.0 V below the maximum |
| Fluke 287/289 | Diode test | 3.1 V range | open-circuit voltage not stated in the pages we read |
| mas838 meter | Resistance | 3.2 V maximum open circuit | 1.2 V above the minimum, 0.8 V below the maximum |
A meter near 3 V sits mid-window. Some parts will start to conduct and others will not, and the sources do not say which. Do not read a missing turn-on as a failure.
Test an N-channel MOSFET step by step
These steps are for a loose enhancement-mode N-channel part. They combine the university lab methods, the Fluke diode procedure and one hobby-level check from a repair FAQ, which is marked.
Prepare
- Touch a grounded object, and keep the part in conductive foam until you probe it.
- Find the datasheet, confirm the channel type and note the pin order.
- Set the dial to diode test. On the Klein MM400 it shares a position with capacitance, and you press SEL once.
- Touch the leads together: the display should read about 0 V. Hold them apart and it should show OL.
Step 1: discharge the gate
Short all three pins together for a moment with a bare wire. The University of Washington handbook suggests wrapping fine copper wire around the leads for this purpose.
Step 2: check that the gate is isolated
Take four readings: red on gate with black on source, the leads swapped, red on gate with black on drain, the leads swapped. The handbook describes the gate as the lead that conducts to neither other lead in either polarity, and a repair FAQ says to verify infinite resistance from gate to both other pins. A good gate reads OL four times. Any other reading points to a damaged gate oxide or an internal protection diode, which some parts include, so check the sheet.
Step 3: discharge the gate again
Short the gate to the source. A contributor to the repair FAQ writes that if you measure the drain after measuring the gate, you will find it conducts, and another says the gate capacitance stays charged and keeps the FET on. The handbook adds that with gate and source shorted, negligible drain-source conduction marks an enhancement-mode part.
Step 4: read the body diode
Keep the gate shorted to the source. Put the red lead on the source and the black lead on the drain. A good body diode shows a forward drop, and Fluke’s 287/289 manual says a typical junction drops 0.5 V to 0.8 V. Then reverse the leads. Fluke says the meter displays OL when a good diode is reverse-biased.
Step 5 (optional, hobby-level): charge the gate and watch the channel
This check comes from contributors to a collected repair FAQ, and the university pages we opened do not describe it for a plain multimeter. With the black lead on the source, touch the red lead to the gate for a second, then move the red lead to the drain. A channel that has turned on reads low, and the reading can persist because the gate holds its charge. One contributor says most FETs come on at about 2 V on the gate. Finish by shorting the gate to the source. If nothing happens, the meter’s test voltage may simply sit below the part’s threshold.
Identify an unmarked part
The handbook says the gate is the lead that conducts to neither of the others, and the diode direction then separates source from drain. It adds that a true four-terminal MOSFET shows source and drain as symmetrical, and that the body diode polarity would then identify the body terminal and the channel type.
Reading the results
| Red lead | Black lead | Healthy N-channel reading | What a different reading suggests |
|---|---|---|---|
| Gate | Source or drain | OL | A low reading points to a damaged gate oxide or a protection diode |
| Source or drain | Gate | OL | As above |
| Source | Drain, gate shorted to source | A forward drop, like a diode | OL both ways resembles an open diode, per Fluke; with the beeper on, a continuous tone below 0.1 V on the Fluke 289 means a shorted junction |
| Drain | Source, gate shorted to source | OL | A low reading means the channel is on or drain and source are shorted |
A repair FAQ contributor writes that the usual failure mode is gate-to-source and drain-to-source shorts together, “everything connected together”. That pattern shows as low readings in every row of the table.
What a diode-mode pass does not prove
- Threshold voltage. The GWU lab finds it with a small circuit, adjusting the supply until an ammeter reads 1.00 mA of drain current.
- On-resistance. Both sheet values sit below one display step, as shown above.
- Gate strength. The sheets rate the gate at 20 V and the Klein applies 3.0 V, which is 15 percent (3.0 ÷ 20).
- Switching behavior. A repair FAQ contributor says you really need a scope to check the drive circuit.
In-circuit checks and P-channel parts
Take the MOSFET off the board whenever you can. Fluke says it may be necessary to remove one end of a diode from the circuit to test it, and Kuphaldt explains that parallel-connected components change the reading. Dedicated board testers exist, and our roundup of PCB testers covers them, but a clean in-circuit reading is not proof of a good part.
For a P-channel part, Kuphaldt notes the symbol’s arrow points the other way. We found no page that walks through a P-channel meter test, so this is our inference: reverse every lead position above and confirm the body diode direction against the datasheet symbol. For bipolar parts, see our guide to an NPN transistor test.
Mistakes to avoid
- Measuring the drain right after the gate. Leftover gate charge can keep the channel on, so short the gate to the source first.
- Using resistance mode on a low-voltage meter. It can show a healthy body diode as open, and Kuphaldt names the under-0.3 V case.
- Expecting the datasheet voltage on the display. The sheet figures come from tens of amps, not milliamps.
- Trusting lead colors on an analog meter. Kuphaldt says many make the black lead positive in resistance mode.
Our guide to using a continuity tester covers the beep test; for a junction, diode test remains the better mode.
How we chose what to include
Each check here comes from an opened page that describes it. The gate, body diode and normally-off checks come from the university and textbook pages. The diode-mode behavior comes from Fluke and the Klein, Fluke and mas838 manuals. The gate-charge check is the one item that rests on collected contributor notes, and the text says so. We opened twelve pages that loaded. Washington University’s MOSFET wiki failed a certificate check, an older RepairFAQ mirror address did not resolve, and the Vishay IRF9540 sheet and the onsemi 2N7000 file returned errors, so none of them counts. Every ratio and difference here is our own arithmetic from a stated figure.
What we could not verify (what the sources do not say)
The pages we opened give no body diode range at meter current, so no source tells you the exact number to expect. They do not state the open-circuit voltage of the Fluke 287/289 diode test, only its 3.1 V range. They do not describe a P-channel meter procedure or an IGBT check, and the sheets do not say how a particular meter will behave on a particular part. Datasheet figures belong to the two named parts and can differ on other MOSFETs.
Frequently asked questions
Can I test a MOSFET without removing it from the board?
Other components sit in parallel with the part and can mislead the reading. Remove the part or lift its leads for a trustworthy result.
Why does a good MOSFET read OL between the gate and the other pins?
The gate sits on an insulating oxide layer, so no DC current flows through it. Both datasheets cap the leakage at 100 nA.
What reading should the body diode give?
Expect a forward drop near the 0.5 V to 0.8 V that Fluke gives for a typical junction, and compare it with a known-good part. The datasheet’s 1.3 V and 2.5 V come from much higher current.
Can I use the resistance range instead of diode test?
Fluke calls resistance mode a fallback for meters without diode test, and the handbook warns its test voltage may be too low.
Does a good body diode mean the MOSFET works?
No. It shows one junction conducts one way, while the channel, threshold, on-resistance and voltage rating remain unmeasured.
Bottom line
Switch everything off, take the part out, short its pins to discharge the gate, then check that the gate is isolated and the body diode conducts one way. Treat that as proof of two intact structures and nothing more.
Sources
- https://people.ece.uw.edu/darling/eefacrbd/EE331LE3rev7.pdf (opened 2026-10-21)
- https://workforce.libretexts.org/Bookshelves/Electronics_Technology/Electric_Circuits_III_-_Semiconductors_(Kuphaldt)/02%3A_Solid-state_Device_Theory/2.10%3A_Insulated-gate_Field-effect_Transistors_(MOSFET) (opened 2026-10-21)
- https://workforce.libretexts.org/Bookshelves/Electronics_Technology/Electric_Circuits_III_-_Semiconductors_(Kuphaldt)/03%3A_Diodes_and_Rectifiers/3.02%3A_Meter_Check_of_a_Diode (opened 2026-10-21)
- https://www.infineon.com/assets/row/public/documents/24/49/infineon-irfz44n-datasheet-en.pdf (opened 2026-10-21)
- https://www.vishay.com/docs/91021/irf540.pdf (opened 2026-10-21)
- https://www.fluke.com/en-us/learn/blog/digital-multimeters/how-to-test-diodes (opened 2026-10-21)
- https://assets.fluke.com/manuals/287_289_umeng0100.pdf (opened 2026-10-21)
- https://data.kleintools.com/sites/all/product_assets/documents/instructions/klein/MM400_1390110ART_WEB.pdf (opened 2026-10-21)
- https://www.repairfaq.org/sam/semitest.htm (opened 2026-10-21)
- https://www2.seas.gwu.edu/~ece20/fall05/lab_ex7.html (opened 2026-10-21)
- http://engr.case.edu/merat_francis/eecs245/labs/lab_7.pdf (opened 2026-10-21)
- https://akizukidenshi.com/goodsaffix/mas838.pdf (opened 2026-10-21)
