N Channel MOSFET Confusion Reigns

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Using the following circuit w/ 2N7000 w/o any V on the gate it is already ON and has 9.46mA Id? Tested the XSTR and swapped out several all w/ the same results? The M-Tester and PDF all show a Vth ~2.5Vg and it is on at 0Vg. I swapped it out for a FQP30N06L and it turns on a bit higher than what the tester and PDF give for Vth and I get a decent Ids vs. Vg curve. So once again I am confused...
tempsnip.jpg
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
I did and it gave me no output w/ Vg up to 10V but it does show a couple of Vg mA going somewhere?
 

Audioguru again

Joined Oct 21, 2019
6,672
What do you mean "without any voltage on the gate"?
You cannot leave the gate on a Mosfet floating, it must be driven with a voltage. If the gate is connected to 0V in your circuit then the maximum drain current is 1uA when it has a 48V supply and is at room temperature. The gate is a capacitor that will hold a voltage previously given to it for a long time if it is floating.
Your gate resistor to ground is 1000k which is 1meg ohms which should cause the gate to be 0V.

Are your Mosfets from China (ebay or AliExpress)? If they are they might be defective or fakes.
Did you make your circuit on a solderless breadboard that picks up every radio and TV station in town?
Transformerless power supply spraying strong ultrasonic signals everywhere? Try using a battery.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Argggh... Cannot get a pic under the 2Meg limit unless JPEG which the forum won't take. Hmmm... I re-edited to under 2Meg and it won't take it. The circuit does match the schematic. I am using 2 linear PSUs, one set 10V @ 0.7mA with LED on. 2nd PSU for Vg and up to 5Vg @ 0.00uA. Tested a pile of XSTRs. Most, even after grounding G, show R between D&S. I found a few that did not but after placing them in the circuit they do show R between D&S. They all test fine in the M-Tester and I even tried the older models to compare. All the same and show N_E_MOS and SGD for pins 123. Vt=2.64V. I am wondering about the components since I bought a lot of 50 from AliExpress and have been burned before.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
OK I finally got it to post. Note that I have 2 DMMs measuring Id and Ig inserted into the circuit. The Vg is the 2nd PSU for the gate.


MOSFET2 (2).jpg
 

sagor

Joined Mar 10, 2019
903
Hm, the right hand pin is supposed to be the Drain, connected to the LED. Left hand pin is supposed to be Source (ground).
At least based on the standard 2N7000 datasheets I've looked at. Am I reading them wrong?
Front (flat side) reads left to right as Source, Gate, Drain....
Try flipping the MOSFET 180 degrees??? (that is if you didn't blow them putting them in backwards...)
 

sagor

Joined Mar 10, 2019
903
Also, don't forget that the 2N7000 has a diode across the D-S junction, so if put in backwards, it will conduct all the time, typical of what you saw....
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
@sagor You are correct! And with the swap nothing... Same as I was having last night. Although you should be able to swap D&S with the change the LED is dead.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
@sagor BINGO!!! That was it! After moving it and double checking the wiring it works! Thanks for the sharp eyes!

Edit: It now comes on ~1.8VG @10uA Ids
 

Audioguru again

Joined Oct 21, 2019
6,672
The threshold voltage is when a Mosfet is barely turned on, and is almost turned off.
The datasheet for the 2N7000 guarantees an output current of only 1mA when the gate-source voltage is 3V. The datasheet guarantees that it turns on fully when the gate-source voltage is 10V.
Some 2N7000 Mosfets are more sensitive than the 3V above and will produce an output current of only 1mA when the gate-source voltage is 0.8V.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
One thing I did find out was that my VICI VC8145 bench DMM is terrible with uA. I had to drag out my Fluke since the VICI would not give a reading. I actually went inside it last night when I was having problems to check both fuses which were good. The display is for 00.000 mA but in the circuit I got nothing and it was blocking the current as if the fuse was blown. Will have to look into that a bit more...
 

TeeKay6

Joined Apr 20, 2019
573
One thing I did find out was that my VICI VC8145 bench DMM is terrible with uA. I had to drag out my Fluke since the VICI would not give a reading. I actually went inside it last night when I was having problems to check both fuses which were good. The display is for 00.000 mA but in the circuit I got nothing and it was blocking the current as if the fuse was blown. Will have to look into that a bit more...
Your VICI probes are plugged into "mA" and "COM"? You have checked the fuse hidden in the "mA" socket? You have the "mode" button for "mA" pressed?
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
@TeeKay6 AHA!!! Didn't know about that one. I found the one hidden in the cordset socket and the 20A on the back of the panel. And Yes, it was blown. Thanx! I was wondering why that socket looked different...
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Indeed! I was convinced everything was "wired" correctly and missed the D&S mixup, dyslexia strikes again... Thx again guys, I really appreciate the help I receive here.
 
Top