abstract theory here, so not talking about a specific component: I need a switchable pull down/up resistor.
I don't want a plain resistor (which will dissipate more or less power all the time depending on its specs and the current I want to control).
The goal is to control a mosfet with a pwm, of which gate could be floating at the beginning, flowing unpredictable current because of possible high current spikes /ringing down the trace -> unpredictable behavior before stabilization.
But I don't want to pay the efficiency cost of a resistor, minimal as it could be.
I need very high efficiency for three 24/7 circuits (<20mA, <40mA and <200A).
Does that switchable resistor exists ? (except a resistor)
May two mosfets specced differently in series do the job?
Sorry, I have read your question several times and I don't know what you are asking for. Maybe someone else will get it but in case not please can you post a schematic showing what you want to achieve? Hand drawn and photographed is fine. How to post an image
To answer directly the question in the title a bipolar junction transistor is current controlled.
You could use P channel MOSFET to connect/disconnect pullup resistor, and N channel MOSFET to connect/disconnect a pulldown resistor, depending on whether your controlled MOSFET is allowing high side current (so you'd want the gate pulled high to turn it off) or low side current (so you'd want the gate pulled low to turn it off).
Say I have an arduino (could be any board).
My pin of interest is one of the pwm. It is connected to a mosfet for a purpose. I control the gate of the mosfet with the pwm.
I need to protect my the path from the pin to the gate from over current. The simple solution would simply be to add a resistor in front of the mosfet gate and call it a day.
BUT. Since I control my circuit power with the pwm pin, I can vary the voltage, and so the current at any moment. And resistors are not power efficient (not talking about if it matters or not, given different circuits constraints, I don't want that).
That resistor is only interesting at the beginning, when the arduino initializes, to reduce the risk of floating pin, ringing, EMI.
But after that stage, it just dissipates power/heat. That I would avoid.
If magic existed, that resistor would be there at the initialization stage of the circuit, disappear, and only reappear a few nanoseconds before it is needed in the circuit. That ressembles the mosfet operation.
are there a really current based version of the mosfet?
The amount of current into or out of the gate is tiny, so the power wasted in any resistor between the Arduino output and the MOSFET gate is also tiny. Not generally worth considering and way, way less than the power controlled by the MOSET and way, way less than the power consumed by the Arduino to control its output. What you are concerned about is not worth worrying about, or I have still not understood.
There might be some esoteric argument about whether an BJT is or is not current driven, for the entire time I have used them regarding them as current controlled has worked very well in my circuits, so that's the interpretation I will stick with. If you find treating them otherwise results in circuits that do what you want be my guest.
PerryBebbington:
The amount of current into or out of the gate is tiny, so the power wasted in any resistor between the Arduino output and the MOSFET gate is also tiny. Not generally worth considering and way, way less than the power controlled by the MOSET and way, way less than the power consumed by the Arduino to control its output. What you are concerned about is not worth worrying about, or I have still not understood.
There might be some esoteric argument about whether an BJT is or is not current driven, for the entire time I have used them regarding them as current controlled has worked very well in my circuits, so that's the interpretation I will stick with. If you find treating them otherwise results in circuits that do what you want be my guest.
As mentioned above.
This current might be .000001 Amps, 1u Amp, and drawn only for .001 second, 1 milli sec, or so each time the Arduino pin goes HIGH (for a common source N channel MOSFET).
Nothing to worry about when there is the Arduino controller taking current.
PerryBebbington:
The amount of current into or out of the gate is tiny, so the power wasted in any resistor between the Arduino output and the MOSFET gate is also tiny
All in all, you are right on the arduino. But the 200A circuit is a different matter. And I still have no solution for it. That current will definitely need cooling solutions anyway. So a resistor won't matter much in a pwm context.
jremington:
and mostly a waste of time to argue otherwise
Right, and I won't.
jremington:
Resistors of any sort, controlled or otherwise, convert electrical energy into heat with 100% efficiency.
(uncontrollable) heat is something I would like to avoid as much as possible, like the rest of us.
You all answered my questions and I would like to thank you for the time spent.
Moving on, my work is now to find the best pwm frequency with my circuits to reach the best electricity usage.
EDIT: A bit too late, but oh well, these were my thoughts...
I need a switchable pull down/up resistor.
I don't want a plain resistor (which will dissipate more or less power all the time depending on its specs and the current I want to control).
A current source with switched range?
I need very high efficiency for three 24/7 circuits (<20mA, <40mA and <200A).
AC or DC?
What voltage or voltage range?
For AC, low voltage and 200A ---> How about a phantom load?
Silly question, but why can't you simply use the off the shelf components currently available instead of
asking for some mythical "energy free " component ? As already explained countless times, the amount
of power dissipation to pwm a mosfet is minimal, and the power dissipated by the mosfet itself in switching
the load is pretty much unavoidable, especially when switching a 200A load. Is there some reason you have
mysteriously OMITTED the nature of the loads you are switching or why you are switching them ?
Usually that is the first piece of information presented by a poster:
ie
"I need to switch a _______ " and would like to reduce the power required to do so. Any ideas ?
as opposed to :
abstract theory here, so not talking about a specific component: I need a switchable pull down/up resistor.
I don't want a plain resistor (which will dissipate more or less power all the time depending on its specs and the current I want to control).
The goal is to control a mosfet with a pwm, of which gate could be floating at the beginning, flowing unpredictable current because of possible high current spikes /ringing down the trace -> unpredictable behavior before stabilization.
But I don't want to pay the efficiency cost of a resistor, minimal as it could be.
I need very high efficiency for three 24/7 circuits (<20mA, <40mA and <200A).
Does that switchable resistor exists ? (except a resistor)
May two mosfets specced differently in series do the job?
arduinist1901:
The goal is to control a mosfet with a pwm, of which gate could be floating at the beginning, flowing unpredictable current because of possible high current spikes /ringing down the trace -> unpredictable behavior before stabilization.
But I don't want to pay the efficiency cost of a resistor, minimal as it could be.
If you use a MOSFET gate driver chip it will handle clamping the gate to the source when idle. For your large current examples you will need a MOSFET gate driver.
Unless you use a Logic Level Mosfet , your choice of gate driver will depend on the Mosfet specs.
If you use the wrong gate driver you'll be able to turn it on but not off.
raschemmel:
..of asking for some mythical "energy free " component ?
raschemmel:
As already explained countless times
raschemmel:
Is there some reason you have mysteriously OMITTED
Well, the tone of your post is definitely on the chill/relaxed side. I feel welcomed.
raschemmel:
the amount of power dissipation to pwm a mosfet is minimal, and the power dissipated by the mosfet itself in switching the load is pretty much unavoidable,
Fine, nobody said otherwise. because we are not talking about the mosfet itself, but the mosfet's gate resistor (between the pwm pin and the mosfet itself.
raschemmel:
but why can't you simply use the off the shelf components
this is what my post (title included) is all about, by describing the behavior I need.
raschemmel:
Usually that is the first piece of information presented by a poster:
first line of my post:
abstract theory here
raschemmel:
which tells us nothing about what or why
And that is because I focused my interest on that part of my circuit.
My friend, your tone is in line with what you understood, but dissonant with what I asked for.
Our colleagues here answered pretty much all my points already and I thank them, again.
Our colleagues here answered pretty much all my points already and I thank them, again.
They didn't say anything that wasn't immediately obvious from a Google Search. In fact , the end result
is that your question was academic and the answer is "no there is not" (a magical component).
The Question
abstract theory here, so not talking about a specific component: I need a switchable pull down/up resistor. I don't want a plain resistor (which will dissipate more or less power all the time depending on its specs and the current I want to control).
Reply#1: "What are you asking for ?" #2:Use mosfets to switch the resistor. (the resistor being "a plain resistor" of course) #3: "Does that switchable resistor exist ?" #4: Your reply #5: "Gate resistor power dissipation is negligible" #6: "BJTs are current controlled" #7: "Current through resistor is negligible" #8: Your reply "You answered all my questions." (Really? What happened to the theoretical magic
component ? Which one of the above questions answered that ? Could it be that you realized
that it was not really a "practical" question and that there was no "practical" solution ?)
Do the math. Switching the gate resistor isn't going to save you much because the I*V for
the resistor is going to be very small, as mentioned . In short, the question was not really
answered. It was ignored, as it should be. In the future, you would do better to simply say:
"How much power is dissipated in a mosfet gate resistor and can it be minimized ?" (and leave
out the 'magical' part. That way it sounds like a normal reasonable question, rather than 'pure
theoretical physics', which this forum topic is not really intended for)
#9: "What do you want ?" #10: Origin of word "transistor" #11: Your reply #12: My comment #13: "Use a gate driver IC"
That's pretty much the whole lot.
A resistor is technically a load, like any other. It's not a stretch to conclude that if a mosfet can be
used to switch a larger load, it can be used to switch a smaller load, although in practice, I think
there may be some issues with switching very small loads. (like you might need a gate drive resistor
for the mosfet switching the gate drive resistor ?) Not sure about that , actually. I'll leave that to you.
So what's my point ?
"Abstract theory here" is hardly appropriate for a "General Electronics" forum topic.
If you were an actual real genuine hobbyist or person with a project, then the appropriate approach
would be "I have some mosfets I need to switch. Can I reduce the power consumed by the gate resistors ?" Is this possible and if so , how ?"
Doesn't that sound more reasonable ? (as opposed to "Twilight Zone here ...")
So much aggressivity.
"The whole lot" was enough to answer my question actually. And I follow their advice in my circuit.
But since I am a fake, really doesn't matter of what exactly, since I am not an "actual real genuine hobbyist or person with a project", I will leave you that point: you are right. May that relax you. (but I am still a hobbyist with a project, mind you, hence my questions).
So much aggressivity.
"The whole lot" was enough to answer my question actually. .
I don't see any "aggressivity". If you ask a nonsense question on an international forum you should expect to be called on the carpet. All I see is me pointing out that none of the answered pertained to the mythical 'magic resistor'. You got the same answers you would have got if you asked the normal question the way
I suggested and left out the abstract theory and 'not a plain resistor' part.
And I follow their advice in my circuit
And what advice is that because I didn't see any circuit. Where is the circuit ? What advice did you follow ?
You see, if you were just up front about everything in the beginning and asked a normal question you wouldn't wind up like this ...this being you talking about a circuit that was'hypothetical',
claiming to have followed advice that you clearly have not specified by posting a circuit indicating how you applied
what you learned from your post. Some people never come clean. I have been neither rude nor aggressive.
All I have done is call you out for posting a bunch of nonsense instead of asking a simple straightforward
question.
Other members following this post would be wondering why what was previously 'hypothetical' is now real and why you are unwilling to share it
on an OPEN SOURCE forum ?
