EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: salfter on August 14, 2018, 04:39:44 am
-
I've been working on getting a 3D-printer interface board (a RAMPS-FD v2.2 (https://github.com/bobc/RAMPS-FD/tree/dev), if it matters) up and running...finally got the bed-heater circuit working, which had been giving me trouble for some reason, and verified the rest of the board works. (I ended up desoldering and resoldering everything in that part of the board. Maybe I had a bad joint previously, though I also ended up replacing a possibly-bad LED.)
While debugging this part of the board, I needed to see if an IRLB8743 MOSFET was any good, or if I had somehow killed it. First thing I ran across was variations of this:
https://www.homemade-circuits.com/how-to-check-mosfet-using-digital/ (https://www.homemade-circuits.com/how-to-check-mosfet-using-digital/)
Basically, you use a multimeter in its diode-test mode: red to drain, black to source. Shorting gate to drain should switch it off, while touching the red probe to the gate and back to the drain should switch it on, and these conditions should be readable on the meter...but mine did nothing. Fail...or was it?
I then ran across this test circuit:
https://electronics.stackexchange.com/questions/319946/should-i-immediately-return-mosfets-shipped-in-non-esd-protective-packaging/319948#319948 (https://electronics.stackexchange.com/questions/319946/should-i-immediately-return-mosfets-shipped-in-non-esd-protective-packaging/319948#319948)
Just a couple of resistors, a switch, an LED, and a power supply...had it knocked together in no time, and this time the LED was blinking off when I pushed the button and turning back on when the button was released, just as described. It looked like maybe it was working after all, so I started resoldering things one component at a time, checking as I went. Once I remembered to reattach the board to the power supply (doh!), it started working like it should. Awesomesauce.
So...that first test of the FET with a meter. Should that have worked, or is there something weird about this type in particular that means that test won't work? The IRLB8743 is a rather beefy device in a TO-220 package that can switch up to 30 amps at up to 32 volts. I was using the diode-test function on an Extech MN26T; the manual says this about the diode-test function: "Test current of 0.3mA maximum, open circuit voltage 1.5V DC typical." Looking at the IRLB8743 datasheet, the gate-threshold voltage is quoted as 1.8V (typical). It's now looking like I'm answering my own question, but just to make sure I'm reading things right, is it the case that the multimeter test for MOSFET functionality isn't going to work for this particular type of MOSFET?
-
Most of the time when mosfets fail, they fail shorted. I usually check between S and D to see that I get the body diode drop in one direction and an open circuit in the other. Then check between G-S and G-D to make sure it isn't shorted there either.
-
It's now looking like I'm answering my own question
Yes. I can vouch for this test; it has always worked for me (although I have probably only done it a few times; rare need for me to test a FET not in a circuit). But my DMM puts out closer to 3V on the diode test.
I can bet that James_s suggestion is probably going to be more useful to you in the future.
-
I forgot to mention that the diode check on the meter can charge up the gate and turn on the mosfet, which then apears to have a D-S short. You can short G-D together for a moment to discharge and turn it off.
-
Depends on the Vgs(th) of the mosfet. "Logic" mosfets with low switching voltage threshold will work. Others may require something like 10V which the diode test will likely not provide
-
IRLB8743 has Vgs(th) of 1.8V typ. so your diode test mode 3V will be able to turn it on fine.
-
I've found that even normal mosfets will be turned on by the diode test. They probably won't saturate, but they don't have to in order to appear shorted to a multimeter.
-
It depends on the max Vgs threshold and your particular DMM's max diode test range voltage.
A PP3 battery + battery clip with a red LED and a 680R resistor in series, to a pair of clip leads, used in addition to your DMM gives you a lot more options to confirm the device under test actually operates. When testing MOSFETs, it will provide >7V gate bias while you test Rds on the DMM, and indicate gate shorts. For BJTs, use it to feed the collector, and use the DMM diode test to feed the base. It also will illuminate LEDs.