How do capacitors charge in a series?

[alpine2beach]

I was wondering... exactly how do capacitors charge in a series? Let me clarify what I mean. Let's say that I have 5 capacitors in a series. How do the ones in the middle gain a charge if they are not connected to the actual power source? I know that this is probably a stupid question, but this has been bothering me for a long time now and I've looked everywhere, yet I still cannot seem to find an answer for WHY capacitors can charge in a series.

 

[Mbert]

Why couldn't they? Perhaps to understand the case for 5 capacitors, you need to first look at how 1 capacitor can charge by looking at the electric field. Since this is a transient problem, you would probably get initially a very high gradient of electric potential near the high-voltage plate (suppose we have a 2 parallel plates capacitor, just to simplify and understand) that would gradually go down as the capacitor charges (I'm guessing). If the potential of the low-voltage plate is not at fixed potential (e.g. grounded), this mechanism will raise the low-voltage plate electric potential. If you put many capacitors in series, the high-voltage plate of the next capacitor gets the potential (and gradient) of the low-voltage plate of the previous and also begins to charge. This means that the capacitors in the beginning of the chain get charged faster and is usually why we insert shunt resistors (to equilibrate the charge in each capacitor, or also for discharge purposes).

Hope this helps,

M.

 

[alpine2beach]

Why couldn't they? Perhaps to understand the case for 5 capacitors, you need to first look at how 1 capacitor can charge by looking at the electric field. Since this is a transient problem, you would probably get initially a very high gradient of electric potential near the high-voltage plate (suppose we have a 2 parallel plates capacitor, just to simplify and understand) that would gradually go down as the capacitor charges (I'm guessing). If the potential of the low-voltage plate is not at fixed potential (e.g. grounded), this mechanism will raise the low-voltage plate electric potential. If you put many capacitors in series, the high-voltage plate of the next capacitor gets the potential (and gradient) of the low-voltage plate of the previous and also begins to charge. This means that the capacitors in the beginning of the chain get charged faster and is usually why we insert shunt resistors (to equilibrate the charge in each capacitor, or also for discharge purposes).

Hope this helps,

M.

Thank You! Finally! Anyways, I just tried to charge a capacitor bank with 5 capacitors in it. When I measured the voltages they were as follows from first to last: 6.52v 6.24v -.93v .406v -.209v Is this attributed to not putting resistors in (the inverting of voltages makes me think not)? It is charged by 2*6v batteries. All caps are 35v @ 1000uf. Is this normal?

 

[Mbert]

Yes, you should put resistors in parallel to equilibrate the charges. The values should be large enough to draw only a small current in steady-state and the rated powers (which shouldn't be a problem at your voltage) should be calculated to withstand that current and voltage. Note that if you have big capacitors (such as in a switch-mode power supply or amplifier), you usually place shunt resistors to discharge the capacitors within a certain amount of time when the circuit is switched off (to avoid getting shocks when servicing).

M.

 

[alpine2beach]

Yes, you should put resistors in parallel to equilibrate the charges. The values should be large enough to draw only a small current in steady-state and the rated powers (which shouldn't be a problem at your voltage) should be calculated to withstand that current and voltage. Note that if you have big capacitors (such as in a switch-mode power supply or amplifier), you usually place shunt resistors to discharge the capacitors within a certain amount of time when the circuit is switched off (to avoid getting shocks when servicing).

M.

Thank you very much!