# Mosfet doesnt fully turn off

So im using a mosfet as a switch to supply power to a servo, move the servo, then switch off. I have the pins wired as follows:

Gate: to pin 9 Drain: to negative of servo Source: to negative of 6v AA battery pack.

I have a resistor that starts at the gate and connects to source. source is also grounded.

When the mosfet is turned on it has a voltage across G/S of about 50mV. When its turned off there is a 4v across the two pins. the servo sees 2v across +/- when the mosfet is off. pin 9 has 0v when its supposed to be off so it is turning off.

EDIT: when i unplug the servo the voltage goes to 0.

Everything works as expected though but its important to have the mosfet off when not in use so that the batteries are saved.

What type of FET are you using?

When the mosfet is turned on it has a voltage across G/S of about 50mV. When its turned off there is a 4v across the two pins.

That is the wrong way round.

Grumpy_Mike: What type of FET are you using? That is the wrong way round.

Nchannel

No, what type of MOSFET exactly - part number, datasheet, anything with numbers

IRFP260n

You need a logic level FET that one takes 10V to switch it.

Also you need to make the servo control pin into an input before you power it off otherwise you get parasitic powering.

Grumpy_Mike: You need a logic level FET that one takes 10V to switch it.

ok i have other mosfets but my question is then why does it still work as expected except for the problem explained above?

the data sheet says its Vgs is 4v though?

I also have these mosfets: nte2376 IRF510

The datasheet says the threshold voltage is 4V. That's nothing to do with switching on, its entirely about switching OFF. You need to be significantly above the PLATEAU voltage for the device to be properly on. Datasheets always show a typical plateau graph (gate charge v. Vgs).

Logic level devices typically have a threshold voltage as 0.5 to 1.0V and plateau about 2.5V.

Non-logic level devices have 2--4V threshold, 4--6V plateau, and are meant to be driven at 10 or 12V.

The plateau in the gate charge curve is where the inversion layer in the channel mainly builds up - this is the layer that carries the current and needs to be fully developed to get the rated Rds(on).

Are either of the other two i posted logic level?

How can you tell if they are logic level?

shouldnt 0v still turn off that mosfet though?

Those are not logic level.

IRL540 is a good logic level MOS fet to have on hand. At 5V VGS, ID is at maximum. For a logic level.

0V will turn off a N channel

.

LarryD: 0V will turn off a N channel

then why when I put 0v to the gate do I see those voltages across the servo/mosfet? It seems as though Im using the wrong mosfet but that that is not the problem. If 0v turns off the mosfet i shouldnt see a voltage across the drain and source pins.

I want to say again that everything is working as I want it to except when its turned off there is that voltage which i worry will drain my batteries when the purpose of the mosfet was to turn off the power to the servo to save power.

I will try to get a logic level soon(holiday weekend kills shipping times :( ) but do you think that that is the cause of the positive voltage across the servo and mosfet?

cbrunnem: but do you think that that is the cause of the positive voltage across the servo and mosfet?

cbrunnem: When the mosfet is turned on it has a voltage across G/S of about 50mV. When its turned off there is a 4v across the two pins. the servo sees 2v across +/- when the mosfet is off. pin 9 has 0v when its supposed to be off so it is turning off.

Grumpy_Mike: That is the wrong way round.

So, you're certain that you have your pins etc. identified right? You really ought to sketch out and post your "circuit" - verbals don't cut it.

Sorry for the crudeness http://postimg.org/image/wqflqaicb/

Your original post states "gate to pin 9" but your dwg has gate to pin 11 (D11) and you show "5" to D9 (for some reason}. Or is that "S" for servo signal? I don't know if that's such a great application: the servo with a switched Gnd and signal still being connected (and Gnd referenced.) That would, in my opinion, be better done via "high-side switching" instead. [I do know that getting the drawing right can sometimes be a real pain.]

[quote author=Runaway Pancake link=msg=2242477 date=1432258460] Your original post states "gate to pin 9" but your dwg has gate to pin 11 (D11) and you show "5" to D9 (for some reason}. Or is that "S" for servo signal? I don't know if that's such a great application: the servo with a switched Gnd and signal still being connected (and Gnd referenced.) That would, in my opinion, be better done via "high-side switching" instead. [I do know that getting the drawing right can sometimes be a real pain.] [/quote]

yes the drawing is correct. i needed to use pin 9 for the servo so i would be able to use PWM on the other pins because the servo library messes with that. im not using pwm for this mosfet but another part of the project.

Can you further explain "high-side switching"?

Why do you need to switch the servo power supply?

JimboZA: Why do you need to switch the servo power supply?

so that the servo is not constantly drawing power. once completed this will be 24/7 operational on a battery. the servo might run upwards of 10 times a day for a half second each. inbetween i want the servo drawing no power. plus the servo twitches and gets warm when its powered and holding position.

EDIT: I see now i need an IRL not IRF mosfet. Im going to frys and they sell NTE mosfets so im assuming these are all suitable:? http://www.nteinc.com/Web_pgs/LL_MOSFET.html

servo.detach() might have the same effect... the 50Hz repeated holding pulse stops so there's no power consumption for holding position. (Although I guess there may be a tiny current in the power line, I'm not sure on that; worth testing if you didn't already.)

JimboZA: servo.detach() might have the same effect... the 50Hz repeated holding pulse stops so there's no power consumption for holding position. (Although I guess there may be a tiny current in the power line, I'm not sure on that; worth testing if you didn't already.)

Hmmmm this could work.... i was unaware of this.

http://www.nteinc.com/specs/2900to2999/pdf/nte2987.pdf is that a suitable mosfet for what im trying to do if servo.detach doesnt work? ill buy bulk from sparkfun since they are way cheaper but im impatient and dont want to wait till mid next week.

I wish there was an easy way to measure current...

It's true, with 4V, your mosfet will turn ON poorly. You need a logic-level mosfet.

But, that doesn't explain your problem with turning off.

When the mosfet is turned on it has a voltage across G/S of about 50mV. When its turned off there is a 4v across the two pins. the servo sees 2v across +/- when the mosfet is off. pin 9 has 0v when its supposed to be off so it is turning off.

This bit from the OP still doesn't make sense.