Question about a Mosfet

[Madshock]

Hi all,

I have a question that i need help on, I understand there is a need for a resistor on a mosfet shorting the Gate and Source. i just cant seem to get the facts right on the need for it even after reading post after post... So i hope someone could tailor to my questions...

Read somewhere that the resistor is there just so that the mosfet will not open its gate itself due to overload? If so, does that mean if i am below the gate voltage, example, 0-4.5V gate and im using 3.7V, I will not need the resistor? Or will it just not open at all due to the low voltage im feeding it?

what resistor would i need to use on a 5v or a 3.7? I tried to source for surface mounted resistors but i cant seem to find any based on Scubabatdan's use on it.. can i use regular 1/4W through hole ones?

I got this samples Power MOSFET - Power MOSFET Transistor - CSD16325Q5C - TI.com
and i was thinking if i should add in the Mosfet on my 5v Variable Box mod just to shrink the button.

i think i just got information overloaded...

 

[bigblue30]

I have a question that i need help on, I understand there is a need for a resistor on a mosfet shorting the Gate and Source.

This is correct.

Read somewhere that the resistor is there just so that the mosfet will not open its gate itself due to overload?

The resistor is there to make sure the gate is at ground when the switch is open. With the resistor, the MOSFET is off whenever the switch is open.

Without the resistor, the MOSFET would stay on after you closed the switch, and would stay on as long as there is battery voltage on the drain and the source was at ground.

In other words….once the MOSFET is turned on by the switch…..It would stay on until the battery died….even if you opened the switch if the resistor was not there.

If so, does that mean if i am below the gate voltage, example, 0-4.5V gate and im using 3.7V, I will not need the resistor? Or will it just not open at all due to the low voltage im feeding it?

The Threshold Voltage for the TI MOSFET you linked below is 1.1 V ….Not 0-4.5

That means that at 0 volts the MOSFET is off, and at 1.1 or above it is on.

what resistor would i need to use on a 5v or a 3.7?

A 47k will work

I tried to source for surface mounted resistors but i cant seem to find any based on Scubabatdan's use on it.. can i use regular 1/4W through hole ones?

Yes

I got this samples Power MOSFET - Power MOSFET Transistor - CSD16325Q5C - TI.com
and i was thinking if i should add in the Mosfet on my 5v Variable Box mod just to shrink the button.

 

[Madshock]

Thank you BB! I think this clears most of my queries!

Is it a must for 47k resistors? I have some 10k, 470, 100 Ohm resistors with me. Im not too sure on the calculations for this. If not then i will need to make a trip to get them 47ks.

 

[bigblue30]

Thank you BB! I think this clears most of my queries!

Is it a must for 47k resistors? I have some 10k, 470, 100 Ohm resistors with me. Im not too sure on the calculations for this. If not then i will need to make a trip to get them 47ks.

Madshock,

Look at the circuit below. On the leg with the switch and resistor.... you will see that when the switch is closed, the full battery is across the switch and resistor. You want to limit the current through the switch, this will increase the life of the switch contacts.

47k works good here.

 

[NuMoNiC]

What determins the value of the resistor? Is there a formula? I have been using the 10K and 47Ks but never noticed much of a differance. Just trying to get a better idea of usefullness vs. overkill. Thanks guys!

 

[Rocketman]

In lieu of buying another 30 cent resistor of the correct value the 10000 ohm one you have will work.
The current that flows through the 10000 ohms won't hurt the switch, waste too much energy, and will pull-down the gate so the MOSfet turns off when you want it off. The three criteria for resistor selection is this application.

(same answer blue, just spelled different)

 

[WillyB]

I too am a bit confused. Using Ohms Law, 5V with a 10k resistor shows the current at 0.5mA. I may be looking at this wrong, but most of the little MadVapes switches folks use are rated at 50mA, 100 times that.



Rocket,

What's incorrect about the 10k one?

 

[Rocketman]

Yup, "won't hurt the switch".

Probably missed the "won't" and the later "will"

 

[Rocketman]

Ok, grammar in my post was hard to follow.
Let's try a different explanation.

How about a forward biased LED shunted by a 47K resistor in series with a 470 ohm resistor between the gate and source.

Nope, ambiguous again and one variation will burn out the LED.

How about a 47K resistor shunted by a forward biased LED in series with a 470 ohm resistor between the gate and source.
Still ambiguous, but they all will work>

By George, I think we've got it this time. One sentence pull-down resistor and LED description.

 

[MahBohMan]

why in the picture ...are the drain source and gate so clearly MISIDENTIFIED. If the source is in the correct position per this drawing....the drain and gate are incorrectly identified. Maybe its just me, but when wiring a real world application, when you purchase a mosfet chip, they are clearly marked pins 1, 2, 3....1=gate, 2= drain, 3= source. This seems to be true for ALL manufacturers, so if theres something different going on with this particular circuit, or if there is a reason why the drain is wired to the gate pin please explain. By the way....if the source pin per this drawing is correctly identified, you are lookin at the BACK of the mosfet ship if you were to lay it on a table and have the pins match the drawing(shiny side facing you). PLEASE CLARIFY THANKS

 

[Madshock]

I hope im right. but the Diagram shows the correct. think of it this way. If you are using a 3 pin Mosfet. You will only be shorting the Pin 1 and 3 with a resistor. there is no connection between the drain and gate until u flip the switch at all. The diagram is basically a "standard" mosfet connection diagram. which u cannot use it as a diagram to mimic the physical mosfet line for line.

 

[WillyB]

why in the picture ...are the drain source and gate so clearly MISIDENTIFIED.

They aren't. Do some research and learn to understand the symbols used in schematics.

A variety of symbols are used for the MOSFET. The basic design is generally a line for the channel with the source and drain leaving it at right angles and then bending back at right angles into the same direction as the channel.

That is a pretty generic representation, here's some for N-channel specific ones.

 

[duby]

Taking a moment to back up Willy and Blue here.
The diagram is correct. That is how you draw a Mosfet in a diagram. It is generally assumed that you will be able to translate it and properly wire pins.
Mosfets are drawn like that, because thats what a mosfet looks like.

It actuality they are a bit more blobly looking than that, but you'll never notice with the naked eye. (Probaly won't even see them for that matter)

Blue made a diagram that is correct for any mosfet, not one specific mosfet. There is an expectation that someone reading the drawing know what they are doing.
Perhaps Blue was incorrect in that expectation, but the drawing is correct, and Quite beautiful I might add. Blue I've really become impressed with your artistic skills they are improving. =)

 

[bigblue30]

Thanks guy. I really need to spend more time here.

 

[AttyPops]

I just finished a 5v mod with a 4 pin regulator. This is probably redundant, but stating it another way may help others reading this.

I found that thinking about the control pin circuit as a high-resistance logic gate helped me... So for my other fellow electronics noobs reading the above:

Given the diagram in post #4 (good work Rembrandt/Blue) one must assume that the control pin on the MOSFET is sensing logic "high/low" and IS HIGH IMPEDANCE (in other words.... it is itself a high-value resistor). Not much current flows into the pin; it's just sensing "high/low" (that threshold >= 1.1 voltage for high). Thus, when the switch is closed, the full current is not going through the chip, and is limited by the pull-down resistor to ground. When the switch is open (no current), the pin is pulled to ground by the resistor.

So I guess the question on the value of the resistor is... how much current does the pin actually need to sense/activate? Not much. It's a logic gate. High value resistors protect the button/wires.

And the cool thing for me was MV included a resistor with the MOSFET (thanks hoogie! You da man!). lol. I didn't even check the value of it (I'll bet it was the 47k).

Anyway, once I got that whole "high-impedance" thing strait in my head, I realized how it worked (or so I think) and understood that another resistor limiting the current to the pin is not necessary. Closing the control switch won't fry the MOSFET (at least, due to the control pin current).

Hope this is at least in the ball-park and explains it (yet) another way.

 

[WillyB]

... And the cool thing for me was MV included a resistor with the MOSFET (thanks hoogie! You da man!). lol. I didn't even check the value of it (I'll bet it was the 47k).

Anyway, once I got that whole "high-impedance" thing strait in my head, I realized how it worked (or so I think) and understood that another resistor limiting the current to the pin is not necessary. Closing the control switch won't fry the MOSFET (at least, due to the control pin current).

Didn't think MadVapes even sold mosfets. I assume you mean the 4 pin Fairchild.

I didn't even check the value of it (I'll bet it was the 47k).

MV just supplies 470Ω for everything. That those resistors work does not mean that there aren't better choices. It may be keeping the Gate from floating but a higher value may be more effective in limiting the current through the switch.

On another note the datasheet for MV's 4pin Fairchild calls for a value of greater than 47uf for the output capacitor. Yet MV shows in his schematic a value of 10uf which is what he has for sale.

 

[AttyPops]

It's a power regulator - KA378R05 - 3.0 amp /w a Fairchild logo. Yup....they sell em.

The resistor was 37K ohm, I think. I need a battery for my meter, or I'd test it. The damn thing is so small I put it under a magnifying glass... looks orange/violet/orange. Maybe a microscope and better light..... lol. Yellow/violet/orange would be 47K. Anyway, it's above 470. I know the 3rd band is not black/brown.

My real quandary is weather or not I really need the capacitors. C0 and C1 look optional in the datasheet, but maybe I'm reading it wrong. I don't really care about ripple for powering an atty. For a microcontroller, or other sensitive devices, I would want caps on it. Then again, I wouldn't want a 3.0 amp regulator in that case.