Go Down

Topic: Transistor Question - Gate threshold voltage not providing max flow? (Read 4160 times) previous topic - next topic

dmayman

Hi there,

I have a quick question regarding MOSFET transistors. Here's the piece in question: 2N7000_D26Z

In the datasheet it says the gate threshold voltage is .8 min, 3 max, and 2.2 average. When I try supplying 5v to the gate and the source, I only get 2.3v at the drain. When I give the transistor 12v at the gate and 5v and the source, I get the full 5v at the drain.

I'm wondering why this transistor doesn't provide full connectivity between source and drain when the gate is given 5v? Isn't that more than enough according to the spec sheet? I tried this with another transistor I have (511-STP16NF06L) and have the same results.

Here's a little schematic of my test condition in case this changes anything.




Thanks!

CrossRoads

You're not using it correctly.
N-channel parts like to sink current.
Use a 150 ohm resistor between arduino pin and the gate.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

winner10920

Perhaps 5v-2.3 = 2.7(convienently the voltge drop of a generic led), so maybe you re not metering it the way in which you would expect the full voltage

dmayman


You're not using it correctly.
N-channel parts like to sink current.
Use a 150 ohm resistor between arduino pin and the gate.


Thanks for the reply. I think I might have fried my uno while testing this. It was getting very hot after sending 12v straight from the Vin to the gate of that transistor. Is it possible something was overloaded with current because I didn't use a resistor between the power source and the gate? Right now the on LED and L LED illuminate when I plug it in but I can't upload and the reset button doesn't affect it at all. I guess this is a lesson in how not to use the Vin. It didnt occur to me the kind of current I was driving to the gate, I guess I thought it had internal resistance? Does this add up or does it sound like another issue?

Thanks again for your help!

dmayman

To clarify - it stopped working after testing my initial circuit, not the circuit you sent.

CrossRoads

Yes, you if allow an output pin to output >40mA for a length of time, it will damage the pin, maybe some around it, and possibly the whole chip.
If the uC is warm to the touch, that's usually a good indication of a damaged chip.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.

fungus


I have a quick question regarding MOSFET transistors. Here's the piece in question: 2N7000_D26Z

In the datasheet it says the gate threshold voltage is .8 min, 3 max, and 2.2 average. When I try supplying 5v to the gate and the source, I only get 2.3v at the drain. When I give the transistor 12v at the gate and 5v and the source, I get the full 5v at the drain.


The voltage at the gate might be 5V relative to ground but it's not 5V relative to the transistors source pin, which is what matters. You have a LED between the source pin and ground which raises the source pin up by a couple of volts.

Golden rule of NPN transistors: The load (ie. the LED) goes above the transistor in the circuit.

No, I don't answer questions sent in private messages (but I do accept thank-you notes...)

dhenry

Quote
It was getting very hot after sending 12v straight from the Vin to the gate of that transistor.
Quote


You were essentially powering the mcu at 11.3v. In experiments like that, you should have a small resistor there to just limit the current.

To make you feel better, avrs can survive 13v for a couple of seconds, at about 50% survival rate. at 12v, 80% of them can survive.

So there is a good chance that your arduino is actually ok. Those little fellows are not that fragile.

dhenry

This actually points to some interesting way to power your mcu: they can be powered from an input pin.

In some cases, that can be quite handy.

MarkT

The threshold voltage is the voltage at which the device turns off, not on. More specifically it is usually quoted at a few 100 _micro_amps.  Ignore Vthr completely for switching applications, the way it works is this:

If Vgs = 0V the device is off.

If Vgs > Vgs(on), then it is fully on, where Vgs(on) is the value quoted in the datasheet under the Rds(on) entry.

The datasheet tells you the on-resistance for a given Vgs - any voltage less than that is not guaranteed to turn it as fully on (or at all).

Logic-level MOSFETs will have something like "Rds(on) = ... (Vgs=4.5V)".  This means the on-resistance is guaranteed for logic level drive.
Non-logic level MOSFETs only have Rds(on) quoted for  Vgs=10V.

Sometimes values for Rds(on) are given for both 4.5V and 10V, this is a logic-level device basically.

Typical values for Vthr are much less than Vgs(on), about 1/4 of it in fact.  There is another voltage, the plateau voltage, at which most of the current switching happens - this is somewhere between Vthr and Vgs(on).  Datasheets have a graph of typical gate charge v. gate voltage - have a look at it and see the plateau.  If you provide a Vgs thats too low you may fall below the plateau voltage and the device will conduct only a small fraction of its rated current.

You only ever use common-source circuit configuration for switching in practice.
[ I won't respond to messages, use the forum please ]

dhenry

Quote
If Vgs = 0V the device is off.


For enhancement mosfets. Depletion mosfets can conduct at 0v Vgs.

Quote
You only ever use common-source circuit configuration for switching in practice.


Or for high speed switching.

MarkT

Depletion-mode power MOSFETs are as rare as hen's teeth - never seen a circuit using one, counts as highly exotic to me.
(Depletion-mode RF signal MOSFETs are another matter)

Show me a high-speed power switching circuit using a source-follower then?  Again must be very niche.
[ I won't respond to messages, use the forum please ]

dhenry

Quote
Depletion-mode power MOSFETs are as rare as hen's teeth


On digikey, they list 112 depletion mosfet. 27 of which rated 1amp or higher.

Quote
- never seen a circuit using one,


Nelson Pass has been quite busy using them.

Quote
counts as highly exotic to me.


They are not typical or even rare. Not sure abou thighly exotic.


Runaway Pancake

I think that "source-follower" operation is appropriate only with JFETs.

2N7000 is a MOSFET, but the OP's dwg uses an NPN symbol.
Next time, ground the Source and place the LED and resistor between +V and Drain.
"Hello, I must be going..."
"You gotta fight -- for your right -- to party!"
Don't react - Read.
"Who is like unto the beast? who is able to make war with him?"

Go Up