Questions about serial and parallel connection of capacitors

[psychepool]

The result of serial connection and parallel connection of capacitors is always slightly confusing.
I will ask you a question with an example.


Q1.
Below is the Orange OR15's Treble EQ cap.

q1.jpg

There are two 470pF caps connected in series, what is the value in this case?
What I know is that if connect capacitors series, the value is halved.
But in this case, since it acts as a filter, it may be thought that there is no difference if two of the same band filters are attached in succession.
When operating as a band filter, I am curious about what value is applied when the capacitors are serialized.


Q2.
Below is the first stage coupling cap of the 3-channel amplifier that I current making.

q2.jpg

Signals are transmitted from the first tube to three channels through the 47n coupling cap.
To give the 2nd channel a little tightness, an additional 2.2n capacitor was added in series, and a jumper was connected with a switch to enable on/off. The effect was pretty good.
In this case, since the band of the 2.2n cap is narrower than that of the 47n cap, is it the same result as applying the 2.2n cap in the end? Or is there another formula?
Compared to the difference in value, it seems that there are not so many band(low end) changes.
If the final band value is decided by the narrower capacitor, I think that it is theoretically correct to decided by the value by 470pF after the 2.2n cap. However, there was a noticeable difference when applying this 2.2n cap.
When a capacitor operates as a band filter, I am curious about how the band value is determined when different values ​​are connected series.


Q3.
Below is the circuit diagram of Friedman BE100's near FAT switch (it looks like the circuit of the early version).

q3.jpg

It seems that the band of the signal is controlled by switching 22n in parallel to the 2.2n capacitor.
In this case, when the switch is connected, is the band determined by the sum of the two capacitors? Or is the band decided on the larger of the two?


Q4.
Until now, it has been a question about the result of capacitor connection as a band filter.
From now on, I would like to ask a question about when electrolytic capacitors are applied to power supplies.

Below is the power supply of Dumble ODS.

q4.jpg

I have seen a lot of cases where the first capacitor is configured like that.
Is it the result of simply reducing the value by half?
What's the difference between connecting one 50uF capacitor of adequate breakdown voltage?


Q5.
Again, this is a question related to the PSU electrolytic capacitor.
Below is the electrolytic capacitor on the preamp side of JCM800 2203 Reissue.

q5.jpg

It seems that a multi-can type of electrolysis is used. What will be the capacitance value and breakdown voltage if two capacitors are used in parallel like this?
And what's the difference between using one capacitor with has same result?


I am gaining a lot of knowledge through your good answers. It is very helpful in making amps.
I am always feeling grateful.
Please answer.

[thetragichero]

series/parallel capacitors combine the opposite of series/parallel resistors, so two capacitors in series will always be smaller than the smallest of the two Ctot = 1 /(1/C1 + 1/C2)
for parallel they just add, so Ctot = C1 + C2

[thetragichero]

for parallel resistors (or series capacitors) i just use this: http://www.1728.org/resistrs.htm

[ChopSauce]

If you think of a capacitor as composed of two conducting plates (1) separated by a dielectric (2) - as schematised - then things might become more obvious.

- when capacitors are in parallel, the side by side plates ressembles a bigger plate, with surface the sum of the surfaces, and capacities sum: C=C1+C2

- when caps are in series, the dielectric is - roughly speaking - composed of two halves of capacitors and then comes (3) the C = (C1*C2) / (C1+C2) but the voltage they can handle doubles(4).

The answer two your questions is a sequel, but Q2: the connection between caps being shared with R7, they can't be considered in series. They are just two single caps.

Also note :

- for Q3: the three nodes connected to SW9/2 are one single node.

- for Q4: I'm sorry I didn't really catch the need for the 220k/3W resistors, though I'm pretty sure there's some good one.

- for Q5: the voltage the "compound" can handle should be that of the (5) least capable of the two
___

(1) that is each point of the plate being at the same potential

(2) roughly a non conducting material

(3) that is 1/C = 1/C1 + 1/C2

(4) as does the thickness of the dielectrics

(5) weakest dielectric

[pompeiisneaks]

For the q4, that's done so that it handles higher voltages. Yes if you have a half sized cap that is rated for over the max B+ then you can just use that. For example on my dumble modded bassman I'm building, I have 600VDC rated caps that are 50uF. So I've just not done two 350V caps at 100uF in series.

If you're doing PS caps in series the resistors are there to help ensure a slow loading of both caps at once and to provide bleeders to the circuit as well on power off so the caps drain. the second one is just a benefit. If you leave both caps w/o balance resistors, it can end up loading improperly and have problems.

If you measure voltage at the top you'll see B+ if you measure in the middle of the two you'll see b+/2 or half b+

~Phil

[psychepool]

Thanks for the detailed answers!
I have additional questions.


In the case of Q1, I understood that it would be 235pF by the formula.
If so, using 220pF instead of two 470pF would have similar results. Then, why did the two connect in series to form a circuit?
Since eq is connected to the plate, a higher voltage flows compared to the eq attached to the cathode follower, so is it to be prepared for that? Is it a choice that has nothing to do with sound?


And I did not understand exactly the case of Q2. Is it 2.1n by the formula? Or does it follow the smaller of the two caps?

[thetragichero]

my guess is this was done to allow fine tweaking on the tone stack. also could be because 470pf are a standard part value found everywhere
you are correct, 2.1nf. you can imagine quite a difference between 47nf coupling cap and 2.1nf! switching this way is smart because they're should be no pops from switching potential differences