understanding common capacitor analog design paradigms

Thread Starter

rick5

Joined Jul 17, 2008
12
The funny thing with capacitors is that there appears to be absolutely nobody who fully understands how capacitors work in circuit, though most people on this forum know enough to get by and use them - and with SPICE tools you don't have to know very much as you can just play around until you get the waveform you want to see.

I reckon I could write a full book on capacitors and how to use them and do a great job of bluffing that I know how it works and everyone would believe I'm an expert.

I'm looking for resources that explain some of the common capacitor uses down to a level of electrons and electrostatic forces, for example with a coupling a capacitor between some arbitrary signal source of known output impedance and amplifier stage of known input impedance. How does that transmit the high frequencies of interest and block everything else?

Some people may think it's not appropriate to understand it down to this level in order to do good design.

The worst block to understanding is all this crap about "capacitors store charge".

One thing I've found in electronics is that you've got to break your head and bust your balls for years or decades before the understanding comes. Just having a degree is useless - no wonder everyone does software.
 

Papabravo

Joined Feb 24, 2006
21,159
The funny thing with capacitors is that there appears to be absolutely nobody who fully understands how capacitors work in circuit, though most people on this forum know enough to get by and use them - and with SPICE tools you don't have to know very much as you can just play around until you get the waveform you want to see.

I reckon I could write a full book on capacitors and how to use them and do a great job of bluffing that I know how it works and everyone would believe I'm an expert.

I'm looking for resources that explain some of the common capacitor uses down to a level of electrons and electrostatic forces, for example with a coupling a capacitor between some arbitrary signal source of known output impedance and amplifier stage of known input impedance. How does that transmit the high frequencies of interest and block everything else?

Some people may think it's not appropriate to understand it down to this level in order to do good design.

The worst block to understanding is all this crap about "capacitors store charge".

One thing I've found in electronics is that you've got to break your head and bust your balls for years or decades before the understanding comes. Just having a degree is useless - no wonder everyone does software.
None of my classmates ever expressed such radical cant. Where is this coming from?
 

beenthere

Joined Apr 20, 2004
15,819
I guess something like T = RC or Xc = 1/ωC would be of no use.

Why all the mystery about coupling caps?

Do the classic trick on yourself - charge up a capacitor of about 10 uF to 50 volts, disconnect it from the charging supply, and grab the capacitor firmly (the trick is to throw it to a buddy). The unpleasant sensation as it discharges through your hand is usually because the cap held a stored charge.

How would you explain it?
 

thingmaker3

Joined May 16, 2005
5,083
The funny thing with capacitors is that there appears to be absolutely nobody who fully understands how capacitors work in circuit, though most people on this forum know enough to get by and use them
I used to teach about capacitors and inductors. I will concur they are confusing for many folk.

I suggest reading through chapters 2, 13 through 15, and 17 & 18 of Donald P Leach's Basic Electric Circuits.

for example with a coupling a capacitor between some arbitrary signal source of known output impedance and amplifier stage of known input impedance. How does that transmit the high frequencies of interest and block everything else?
It doesn't. The capacitive reactance decreases as frequency increases. The lower the frequency, the more reactance, and the more the signal is attenuated. Said signal is never fully eliminated. It is said to be "blocked" when attenuated below 3dB.
 

Ratch

Joined Mar 20, 2007
1,070
rick5,

...The worst block to understanding is all this crap about "capacitors store charge".
Yes, capacitors don't store charge, they store energy. A capacitor has the same NET charge at 0 volts at it does at 100 volts. The number of charge carriers accumulated on one plate is the same as removed from the opposite plate for a net charge change of zero. This imbalance of charge causes a voltage to appear, which in turn creates a electrostatic field where the energy is stored. So one should say a capacitor is "energized", not "charged".

Ratch
 

Thread Starter

rick5

Joined Jul 17, 2008
12
Thanks for all these replies. I thought I had to make a radical post to get people's attention as there must be a million posts about capacitor questions already and nobody would answer if I just said "how does a capacitor work?", the might say rtfm. Unfortunately in electronics there isn't a user guide.
 
Last edited:

Wendy

Joined Mar 24, 2008
23,415
We're volunteers here all. No advertising, no one gets paid. If someone can help they do, we has source material in the same site in the form of a text book. If we don't feel comfortable we don't answer. It is that simple.

I replied to your rant, then deleted it, because this is a no flame zone, and I like it that way. We have a solid core of talent, but things will slip through the cracks. If it bothers you jump in and help.

This site allows things that no one (and I do mean no one) else allows.
 

Thread Starter

rick5

Joined Jul 17, 2008
12
I guess something like T = RC or Xc = 1/ωC would be of no use.

Why all the mystery about coupling caps?

Do the classic trick on yourself - charge up a capacitor of about 10 uF to 50 volts, disconnect it from the charging supply, and grab the capacitor firmly (the trick is to throw it to a buddy). The unpleasant sensation as it discharges through your hand is usually because the cap held a stored charge.

How would you explain it?
I would explain it as the potential energy stored in the electrostatic field between the plates is released into the holder as an electric current, but as Ratch said there is no net charge change in the capacitor as the same number of extra electrons that flowed onto the "charging plate" have been removed from the opposite plate.

I do understand how to use coupling capacitors, but I guess I'm just thinking down to too low a level of detail like on a physics level.
 
Last edited:

Ratch

Joined Mar 20, 2007
1,070
rick5,

I do understand how to use coupling capacitors, but I guess I'm just thinking down to too low a level of detail like on a physics level
Just think of capacitors as just another energy storage device. The capacitance relates the charge difference between the plates to the voltage across the plates. That is usually enough to know for circuit analysis. If want still more detail, then you have to get into dielectric constants and such.

Ratch
 

scubasteve_911

Joined Dec 27, 2007
1,203
It isn't practical to have such in-depth understanding, this is the separation from pure science and applied science.

If you understand how the capacitor acts, then you can apply them practically. If you understand how they work, you can produce better capacitors to sell to engineers. You cannot be the jack of all trades!

Personally, I have learned how capacitors 'actually' work from electromagnetic physics. This is part of my degree requirements.. But, to improve them, it's more about understanding what parameters to make better and then it becomes more of a chemistry problem. Dielectrics aren't simple and probably warrant their own degree..

Steve
 

Papabravo

Joined Feb 24, 2006
21,159
I've always considered the passive components R, L, and C as Linear operators. That's all you need to apply them in a circuit.
 

Thread Starter

rick5

Joined Jul 17, 2008
12
It isn't practical to have such in-depth understanding, this is the separation from pure science and applied science.
I think you've hit it on the head pretty well. I was coming round to this opinion myself, but I was interested in knowing for my own interest and also interested in seeing how deep other peoples' understanding is.

Perhaps if you're doing analog IC design you may need to go to that level of understanding, but for board level design it's not needed. Probably even for IC design you don't need to know either as you rely heavily on EDA software.
 

scubasteve_911

Joined Dec 27, 2007
1,203
Thanks Rick,

I noticed I might have 'down-played' how much understanding engineers need from a practical standpoint. I find that capacitors are incrementally learned by engineers over time. The most basic is value, then one realizes that there is a voltage level that will damage the capacitor. Then, it stems off into ESR, ripple current (power dissipation), leakage, etc. Then, of course, the chemistry of dielectric that is 'good' for these characteristics, ceramics: X5R, NP0, tantalums, electrolytics, polystyrene, mica, list goes on and on. It can be quite daunting for an engineer to learn, especially when we are only taught value in school.

Steve
 
Top