Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Capacitors in parallel

Status
Not open for further replies.

carbonzit

Active Member
Something else that's puzzled me for a good long while. May seem elementary, but maybe it isn't ...

Often one sees two caps placed next to each other, like across a power-supply rail to filter noise, spikes and ripple, like so:

**broken link removed**

It always strikes me that since the two caps are in parallel, what we really have is this:

**broken link removed**

since capacitances in parallel are simply added together (same as resistances in series). In other words, the small capacitance should be all but completely swamped or subsumed by the larger one.

But obviously this isn't the case. So isn't what we really have more a case of this?

**broken link removed**

or maybe even this?

**broken link removed**

In effect creating a "pi" filter. Otherwise, what would be the point of adding a little bitty capacitor next to a big old electrolytic, for example?

So enlighten me ...
 
Instead of a small inductor between the capacitors, figure a small resistor and inductor in series with the large electrolitic. They arn't good at high frequency transiets.
 
Like this?

**broken link removed**

Hadn't thought of that. So you're saying the "inductance" keeps high frequencies away from the larger cap? But the inductance must be very, very small. Hard to see how that could work.
 
I think all capacitors in practice have some resistance with it (higher with higher capacitor values), and not purely capacitive. So your circuits are not exactly equivalent.
 
Last edited:
A 1000uF electrolytic capacitor is excellent at low frequencies but is very poor at high frequencies.
A 0.1uF ceramic or film capacitor is very good at high frequencies but is very poor at low frequencies.
When both capacitors are in parallel then their combination is very good at most frequencies.
 
A 1000uF electrolytic capacitor is excellent at low frequencies but is very poor at high frequencies.
A 0.1uF ceramic or film capacitor is very good at high frequencies but is very poor at low frequencies.
When both capacitors are in parallel then their combination is very good at most frequencies.

Think..... A marriage made in heaven. :)

Ron
 
A 1000uF electrolytic capacitor is excellent at low frequencies but is very poor at high frequencies.
A 0.1uF ceramic or film capacitor is very good at high frequencies but is very poor at low frequencies.
When both capacitors are in parallel then their combination is very good at most frequencies.

I already know that.

Please explain why the two capacitances simply don't merge into one large one, which would totally negate the effects you describe. (A 1000 uF cap, as you point out, is ineffective at high frequencies.) That's the point I was trying to make; what is the actual effective circuit at work here?
 
Attached is a graph of 3 capacitors. (ceramic) It shows how capacitors resonate and become inductive at high frequencies.
22uf, 0.1uf and 0.01uf.
I have a bank of 330uF high current, low ERS, low ESL capacitors. (the best money can get) They resonate at about 200 to 300khz. I am slamming them with 320khz square wave current. At 1mhz the caps are not much good. They do not filter out 10mhz and above because they are to inductive. I have 2 of the 22uf ceramics, 2 of the 0.1 and some 0.01uf caps all in parallel. The smaller caps work well at high frequencys.
 

Attachments

  • cap.JPG
    cap.JPG
    102.8 KB · Views: 176
The low frequency and high frequency capacitors are in parallel. They filter away most frequencies very well and do not affect each other.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top