What does negative voltage mean for MOSFET?

In summary, the conversation is discussing the use of a MOSFET with a Raspberry Pi to control a circuit. The individual is having trouble choosing which type of MOSFET to use and is confused by the negative voltage values in the data sheet for a possibility. A summary of potential solutions is provided, including using a P-channel MOSFET with very low threshold voltage or using a BJT (NPN) to drive the MOSFET. The conversation also explores the option of using a 2 input multiplexer or a half dozen NAND gates to create a basic multiplexer. It is recommended that the individual have a good understanding of how MOSFETs work before attempting to use them in their circuit.
  • #1
andrew700andrew
11
0
Hi, I'm planning on using a mosfet with my raspberry pi to controll a circuit, using the GPIO pins which can be set to +3.3V or GND but I'm having problems with choosing the type of MOSFET I will need. I've found the data sheet for a possibility here
http://www.diodes.com/datasheets/ZVP3306A.pdf
But I'm confused as to why all of the figures are negative. E.g. The Gate-Source threshold voltage is -1.5V. Does this mean I cannot use it wih my raspberry pi? What does negative voltage even mean in this case? I'm not sure how you can have a negative becuase doesn't that just mean the circuit is connected In reverse? So as you can see I'm rather confused and would really appreciate it if someone could explain what the negative values for voltage (and current) mean practically.
Thanks.
 
Last edited:
Engineering news on Phys.org
  • #2
No you cannot use this MOSFET. This MOSFET is a P- channel one. And you need a N- channel type one with very low threshold voltage (logic level). Or use a BJT (NPN) that will drive the MOSFET.
 
  • #3
Jony130 said:
No you cannot use this MOSFET. This MOSFET is a P- channel one. And you need a N- channel type one with very low threshold voltage (logic level). Or use a BJT (NPN) that will drive the MOSFET.

I need to use a P channel becuase the MOSFET will be controlling a PWM data line connected to a servo, so the load will always be on the GND side of the MOSFET. Is it only this specific one I cannot use, or P channels in general? Also, could you explain why this one is not going to work? Thanks for replying!
 
  • #4
andrew700andrew said:
I need to use a P channel becuase the MOSFET will be controlling a PWM data line connected to a servo, so the load will always be on the GND side of the MOSFET. Is it only this specific one I cannot use, or P channels in general? Also, could you explain why this one is not going to work? Thanks for replying!
Can you provide a sketch of the circuit (with voltage levels) that you have in mind?
 
  • #5
NascentOxygen said:
Can you provide a sketch of the circuit (with voltage levels) that you have in mind?

http://imgur.com/UwkOqvX

The 300k ohm resistor represents the servo and the two "raspberry pi GPIO" on the side are the pins I will use to controll which device can send PWM signals to the servo. This will either be the RX (radio reciever) which will continuously transmit a PWM signal or the raspberry pi which will only sometimes send a PWM signal, but when it does it will use the other two GPIO ports to disconnect the RX, then connect it's self to the servo data line and take over controll. At all other times the RX should have control of the servo. Also, I don't think the RX always uses a set max voltage for the PWM (depends on the voltage supply to the RX) so that's why I've written between "3.3 and 5v".
 
  • #6
I guess what I'm also asking is are there an alternatives to create a circuit which allows me to use two PWM signals to drive one servo with a way to select which source I use, if the MOSFET idea is bad.
 
  • #7
The MOSFET idea is not necessarily ruled out, it just hasn't been considered sufficiently thoroughly (speaking only for myself!). Do you have IC sockets you can solder to? There are available plenty of MOSFETs-on-chips and already fabricated data switches, I'll be surprised if something can't be found to do the job; the CD4016 or CD4066, for example. http://www.jaycar.com.au/products_uploaded/ZC4066.pdf
 
Last edited by a moderator:
  • #8
andrew700andrew said:
I guess what I'm also asking is are there an alternatives to create a circuit which allows me to use two PWM signals to drive one servo with a way to select which source I use, if the MOSFET idea is bad.
Look into using a 2 input multiplexer to drive the gate of your FET.
 
  • #9
donpacino said:
In your diagram you show the raspberry pi PWM on the source of the FET. That is wrong. the pwm signal should come through the gate of the FET. The same should occur with the signal from the receiver. You do NOT want to pull power from your pi or receiver.

In which case, will something like this work?
http://imgur.com/vmcR53y
The diodes will prevent the current shorting to ground at the GPIO. The problem is still this negative voltage thing and also, in the simulation I'm only getting 2.8V over the servo resistance even when is is really high. Is this due to the diodes? Also, I'll check out the other devices later.
 
  • #10
andrew700andrew said:
In which case, will something like this work?
http://imgur.com/vmcR53y
The diodes will prevent the current shorting to ground at the GPIO. The problem is still this negative voltage thing and also, in the simulation I'm only getting 2.8V over the servo resistance even when is is really high. Is this due to the diodes? Also, I'll check out the other devices later.
Do you know how a mosfet works?
 
  • #11
I reccomend using nascent's solution, as it is a one package version of my solution, but you really shouldn't be doing this if you don't have a grasp on how fets work.
 
  • #12
I recommend using nascent's solution, as it is a one package version of my solution, but you really shouldn't be doing this if you don't have a grasp on how fets work.
 
  • #13
donpacino said:
In your diagram you show the raspberry pi PWM on the source of the FET. That is wrong. the pwm signal should come through the gate of the FET. The same should occur with the signal from the receiver. You do NOT want to pull power from your pi or receiver.
I don't see the problem. The control signal goes to the gate, the controlled signal is fed S to D.

But you multiplexer idea is good, using a half dozen NAND gates to make a basic multiplexer would be a fruitful exercise and is all that's needed.
 
  • #14
you should NEVER power a servo through the raspberry pi, it should come directly from the power source. in addition it is not a good design practice to have the gate of a ptype fet at a higher potential than the source
 
  • #15
donpacino said:
you should NEVER power a servo through the raspberry pi, it should come directly from the power source. in addition it is not a good design practice to have the gate of a ptype fet at a higher potential than the source
The servo input is 300kΩ, he's not powering the servo.
 
  • #16
donpacino said:
you should NEVER power a servo through the raspberry pi, it should come directly from the power source. in addition it is not a good design practice to have the gate of a ptype fet at a higher potential than the source

Ok, The 2 input multiplexer looks right, well at least the truth table on them seems to do what I'm looking for.

And the Raspberry pi won't be 'powering' the servo, the servo has 3 leads, one for +5V other GND and the third is the PWM control which is fine at 3.3V and can be controlled by the Pi, as I have already made it work.
 
  • #17
ohhh if that's the case then you don't need a fet at all. just the mux should do it
 
  • #18
I've found this http://www.ti.com/lit/ds/symlink/sn54ahc157.pdf to use as a MUX.

However, I'm confused about a few details:
1) What is the "Input clamp current" because I'm worried about the negative current there.
2) Do I need some sort of resistor on the Vcc to limit the current through the MUX. It says the continuous current though Vcc or GND is +-50mA, is this what I need to aim for (in which case I also need to know the resistance of the MUX itself through Vcc to calculate a current limiting resistor value) or is this what the current will be if I connect the MUX directly to a 3.3V supply and GND (i.e. no resistor is required)
3) Also, what are the (Ioh) High-level output current and (Iol) Low-level output current values mean? The High level is -4mA and the Low-level is +4mA, yet earlier in the datasheet it says that the Continuous output current is +-25mA?

Thanks.
 
  • #19
1. that is in the absolute maximum rating table. that means with vi<0 the input current will clamp at 20 mA
2. no you do not. like i said that is the max current draw without damaging the part will be +-50 mA. you should be able to calculate your current draw from the load to see if you lie within that range.
3. like i said the above table is absolute maximums for the part. The below is recommended maximums for specific operating conditions.
 

1. What is a MOSFET and how does it work?

A MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) is a type of transistor commonly used in electronic devices. It works by controlling the flow of current between the source and drain terminals using an electric field created by a voltage applied to the gate terminal.

2. How is negative voltage applied to a MOSFET?

Negative voltage is applied to a MOSFET by connecting the negative terminal of a power supply or battery to the source terminal of the MOSFET, and the positive terminal to the drain terminal. The gate terminal is then used to control the flow of current between the source and drain.

3. What does it mean when a MOSFET has negative voltage?

When a MOSFET has negative voltage, it means that the source terminal has a lower voltage than the drain terminal. This creates a reverse bias on the channel between the source and drain, causing the MOSFET to behave differently than it would with positive voltage.

4. What are the effects of negative voltage on a MOSFET?

The effects of negative voltage on a MOSFET can include a decrease in the channel current and an increase in the threshold voltage. It can also cause the MOSFET to switch off faster and have a lower resistance in the off state.

5. Can negative voltage damage a MOSFET?

Yes, negative voltage can potentially damage a MOSFET if the voltage applied exceeds its maximum rating. It is important to follow the manufacturer's specifications and properly design the circuit to prevent negative voltage from causing damage.

Similar threads

Synchronous MOSFET Driver not functioning when voltage already present
    • Electrical Engineering
Replies
2
Views
807
Looking for the right MOSFET transistor
    • Electrical Engineering
Replies
6
Views
1K
2n3904/2n3906 driving mosfet gate
    • Electrical Engineering
Replies
5
Views
3K
Is there a way to keep an LED turned on even when the power source is off?
    • Electrical Engineering
Replies
2
Views
2K
Simple understanding of mosfets
    • Electrical Engineering
Replies
6
Views
1K
What Are the Two Types of Threshold Voltage (Vth) in MOSFETs?
    • Electrical Engineering
Replies
1
Views
3K
Engineering Determining if voltage or current is positive or negative with sources
    • Engineering and Comp Sci Homework Help
Replies
10
Views
4K
Body effect for NMOS transistor
    • Electrical Engineering
Replies
11
Views
3K
What is a Negative Voltage Regulator and How Does it Work in Audio Amplifiers?
    • Electrical Engineering
Replies
6
Views
3K
What is the Proper Voltage to Open a 350v Rated Power N Channel Mosfet?
    • Electrical Engineering
Replies
1
Views
2K

Hot Threads

Recent Insights

Back
Top