#### cmartinez

Joined Jan 17, 2007
7,223
I know what a Farad is, and what a Coulomb is. In fact, I've just finished reading both entries in the Wikipedia before coming here to refresh my memory, and to assure everyone that I've done my bit of research before posting my question:

Starting from the definitions:
One Coulomb = 1*F*V, or 1*A*s
One Farad = 1*A*s/V (among other definitions)

My question here is, why is it sometimes better to express things in Coulombs than in Farads and vice-versa? Why is a capacitor's capacitance better expressed in Farads? Why not use Coulombs instead?

I'm under the impression that it's easier for a scientist or engineer to visualize things depending on the chosen units. For instance, a Co2 laser's power is normally expressed in Watts, whereas a Ruby's is preferably expressed in Joules. That is because a Co2 laser has a continuous output, while a Ruby laser works in very short pulses. And although those pulses do span a certain amount of time, (and therefore its power can also be expressed in Watts) it's easier to visualize a Ruby laser's power as an instantaneous burst of energy. That's why Joules come into play.

Is that why a mosfet's gate capacitance is expressed in Farads, while a transistor's in Coulombs?

Last edited:

#### GopherT

Joined Nov 23, 2012
8,012
I know what a Farad is, and what a Coulomb is. In fact, I've just finished reading both entries in the Wikipedia before coming here to refresh my memory, and to assure everyone that I've done my bit of research before posting my question:

Starting from the definitions:
One Coulomb = 1*F*V, or 1*A*s
One Farad = 1*A*s/V (among other definitions)

My question here is, why is it sometimes better to express things in Coulombs than in Farads and vice-versa? Why is a capacitor's capacitance better expressed in Farads? Why not use Coulombs instead?

I'm under the impression that it's easier for a scientist or engineer to visualize things depending on the chosen units. For instance, a Co2 laser's power is normally expressed in Watts, whereas a Ruby's is preferably expressed in Joules. That is because a Co2 laser has a continuous output, while a Ruby laser works in very short pulses. And although those pulses do span a certain amount of time, (and therefore it's power can also be expressed in Watts) it's easier to visualize a Ruby laser's power as an instantaneous burst of energy. That's why Joules come into play.

Is that why a mosfet's gate capacitance is expressed in Farads, while a transistor's in Coulombs?

Sorry, whether kilograms, moles, BTUs of fuel equivalants, the closer and more familiar you get to the technology, the more inter convertable the units become.

Farads are independent of voltage when designing a filter, the lag is the same whether 5 or 50 volts.

Coulombs are a quantity of electrons. It takes about 95000 coulombs to make a mole, (from memory) - look it up to confirm. Independent of voltage.

Once you have coulombs and farads, it is like knowing the amount of water and the height and diameter of the storage tank it is in. The format of the storage tank determines the free-fall potential energy of the highest kilograms of water in the tank. You can calculate lots of things about the potential energy and so on.

Same for a capacitor and knowing the number of coulombs, you can figure out how many volts will be in the capacitor and the decay as that potential energy drained though a resistor or some other load.

#### GopherT

Joined Nov 23, 2012
8,012
Is that why a mosfet's gate capacitance is expressed in Farads, while a transistor's in Coulombs?
You flipped these, MOSFETS in Coulombs, BJT in farads.

#### cmartinez

Joined Jan 17, 2007
7,223
You flipped these, MOSFETS in Coulombs, BJT in farads.

#### Attachments

• 109.8 KB Views: 4

#### hp1729

Joined Nov 23, 2015
2,304

So both types of transistors have characteristics rated in Farads?

#### #12

Joined Nov 30, 2010
18,216
A farad is a measure of capacity, like a liter or a gallon. A coulomb is a measure of energy. How much energy can you stuff into a farad? It depends on how much voltage is associated with the current. Mosfets have...what? picofarads from gate to source? The mosfet datasheets also have a graph of how many coulombs of charge you must stuff into the gate capacitance in order to raise its voltage sufficiently.

#### GopherT

Joined Nov 23, 2012
8,012
No standard exists, apparently?

Irf540 uses nCoulombs.
2n3904 from fairchild lists pF

#### cmartinez

Joined Jan 17, 2007
7,223
No standard exists, apparently?

Irf540 uses nCoulombs.
2n3904 from fairchild lists pF
And therein lies the nature of my question...

#### cmartinez

Joined Jan 17, 2007
7,223
No standard exists, apparently?

Irf540 uses nCoulombs.
2n3904 from fairchild lists pF
Thanks for your fist answer, BTW, it's more or less what I expected... some things are easier to visualize using particular units.
Also, same thanks to you #12

#### ifixit

Joined Nov 20, 2008
650
Some thoughts on that...

If you're more concerned about the amount of energy stored in a capacitor then use Coulomb, i.e. photo discharge capacitor, but if only the capacitance value is of concern, i.e. active RC filter, then use Farads since you don't care about the energy... just the value.

If the laser application concerns energy dissipation (metal cutting) then Joules would be appropriate, however if it is a laser for communication then Watts are a better term to use.

MOSFET gate capacitance effects turn-on and turn-off time, however, the capacitance value by itself doesn't convey what the delay effect is well since internal gate resistances are not specified. A Coulomb value tells you the amount of energy that needs to be removed before the gate will go on or off. This allows you to pick a series gate resistance that operates the gate fast enough and terminates the signal well to minimise ringing.

Transistor base capacitance and resistance is low enough so that energy storage effects are not a concern for most applications.

Ifixit

#### WBahn

Joined Mar 31, 2012
26,398
They are measures of different things. A farad is a measure of how much charge a capacitance can hold for each volt of potential difference across it. It describes the structure and applies whether or not there is any charge stored there at all. A coulomb is a measure of how much charge is actually stored on that capacitance at a particular time (or under a particular set of conditions).

Neither is a measure of energy, though both can be a proxy in a given context. If I tell you that one capacitor has 1 mC of charge on it and another has 2 mC of charge on it, can you tell me which one has more energy stored in it? Of course not, since the energy stored on a capacitor is CV²/2. Since Q = CV, this can be written as Q²/(2C). So if the 1 mC charge is on a capacitor that has ten times the capacitance, the energy stored in it will be many times smaller than the capacitor charged to 1 mC due to the much higher voltage required to place the smaller amount of charge on the smaller capacitor.

As for transistor ratings, I've only seen capacitance values given since they are descriptive of the structure. But in cases where you have a relatively fixed voltage, such as the base-emitter junction of a BJT transistor, I can see it being convenient to describe the capacitance in terms of the amount of charge that is stored there at that operating voltage. In designing imagers we commonly did the same thing -- the pixel capacitance was often discussed in terms of the number of electrons (not coulombs, but the number of electrons) it could hold at it's saturation level (usually defined as the point at which it either became too nonlinear or the point at which charge would start to be lost through some mechanism).

#### crutschow

Joined Mar 14, 2008
26,077
To summarize:

A coulomb is a measure of charge, not energy.

A farad is a measure of charge capacity.

The gate of a MOSFET is often rated for both its capacitance and the charge required to operate the gate.
The capacitance is useful for small signal (AC) calculations at a single DC bias point.
The gate charge is useful when operating the device as a switch since it shows the total charge (no energy) that needs to be added or removed from the gate to switch the device fully on to fully off. The capacitance value is not accurate for that calculation since the capacitance varies with gate voltage.

#### Externet

Joined Nov 29, 2005
1,562
The size of a bucket is the capacity in Farads. The amount it is filled is the charge in Coulombs.

#### cmartinez

Joined Jan 17, 2007
7,223
The size of a bucket is the capacity in Farads. The amount it is filled is the charge in Coulombs.
That's a pretty good analogy.
Wouldn't it be better if we used a pressure tank instead of a bucket, so as to be able to include volts?

#### GopherT

Joined Nov 23, 2012
8,012
That's a pretty good analogy.
Wouldn't it be better if we used a pressure tank instead of a bucket, so as to be able to include volts?
Think of farads as the diameter of the bucket. After filling with charge, the height that the charge reaches is the voltage.

#### bertus

Joined Apr 5, 2008
21,063
Hello,

Perhaps a bit late, but have a look at the attached PDF.

Bertus

#### Attachments

• 97.5 KB Views: 16

#### cmartinez

Joined Jan 17, 2007
7,223
Think of farads as the diameter of the bucket. After filling with charge, the height that the charge reaches is the voltage.
mmmmhhhh... interesting. Yes, the higher the water level, the higher the pressure at the bucket's bottom.

I've always liked to think of electricity using hydraulic analogies. Although, of course, those analogies can only get you so far. There comes a point in which one has to take an intellectual leap and visualize things the proper way.
Thanks for the tip!

#### cmartinez

Joined Jan 17, 2007
7,223
Hello,

Perhaps a bit late, but have a look at the attached PDF.

Bertus
Excellent document.... just what I needed to refresh my knowledge. Thanks!

#### #12

Joined Nov 30, 2010
18,216
A coulomb is a measure of charge, not energy.
Yeah...I was having difficulty with words. Trying to keep it simple but it got so simple that it got wrong.