Transformer/Capacitor question

[George L.]

Hello eveyone,

I have two simple questions, I would really appreciate if you could help me.

If you connect two capacitors together in series that have the same ratings, will the voltage double. Example - You have two 100uF caps rated at 25V each. If connected in series, could you charge it to 50V, and the capacitance would still be 100uF ???

2. Also, if you connect two transformers together in the followong way, what would be the result... Two identical step-up transformers with a 4:1 ratio, the primary coils are connected in parallel and the secondary coils in series. Would the ratio of the entire set up be 8:1, or would it just not work ???

I would really appreciate it if someone could clear this up for me.

Thanks,

George

 

[Dean Huster]

In the case of your series-connected capacitors, George, the voltage rating will be double, however, the capacitance value will be half, assuming indentical caps.

In the case of the transformer, the output voltage will be the sum of the two secondaries or it will be zero! I depends upon the phasing of the connections of the secondary and primary. If like primary wires are connected together, the primary will be in-phase. If the secondaries are connected in series with "unlike" ends connected and the other "unlike" ends used as the output, you will effectively have an 8:1 step-up ratio. But be very careful doing this, for you can get into some pretty high voltages. Also, the transformer's voltage ratings (e.g., secondary-to-primary differential) could be exceeded and the transformers will arc over between windings. That's the down-side of it all.

Dean

 

[ljcox]

In the case of your series-connected capacitors, George, the voltage rating will be double, however, the capacitance value will be half, assuming indentical caps.

Dean

As Dean said, this is true for identical capacitors. But all components have a tolerance. So one is likely to be greater than the other. eg. one may be 104 uF and the other 98uF.

The charge Q on both will be equal, therefore the voltages will be different.

Q = CV = C1 V1 = C2 V2. Thus V1 = C2 V2 / C1.

So if C2 > C1, then V1 > V2.

Therefore the voltage across C1 could be greater than the allowable voltage.

Len

 

[Nigel Goodwin]

If you're putting capacitors in series in this way, you MUST fit voltage balancing resistors across them - otherwise they will have different voltages across them - and one will most probably fail within a fairly short time.

 

[lord loh.]

If you're putting capacitors in series in this way, you MUST fit voltage balancing resistors across them - otherwise they will have different voltages across them - and one will most probably fail within a fairly short time.

I do not figure this part out. Why would the voltage across each of the capacitors be different? And why would one fail?

 

[ljcox]

If you're putting capacitors in series in this way, you MUST fit voltage balancing resistors across them - otherwise they will have different voltages across them - and one will most probably fail within a fairly short time.

I do not figure this part out. Why would the voltage across each of the capacitors be different? And why would one fail?

See my post, the one before Nigel's.

Len

 

[Nigel Goodwin]

I do not figure this part out. Why would the voltage across each of the capacitors be different? And why would one fail?

See my post, the one before Nigel's.

Yes, exactly right!.

The two capacitors will never be exactly the same value, so one will always have more voltage across it than the other - the balancing resistors help to keep the voltage the same - and also provide a discharge path for when the power is unplugged!.

 

[George L.]

thanks for helping everyone,

here is what I wanted to do with the series capacitors:

Instead of buying a 32 Microfarad 4500V energy storage capacitor from https://www.amazing1.com/capacitors.htm. The capacitor cost $170 :shock:....

I though I could make my own energy storage cap by connecting ten 330V 150uF photo flash capacitors caps in series for a total for 3300V cap. As I have just learned, it is not that simple. I need voltage balencing resistor that need to be able to handle high voltage(is it optional or is it necessary for performance), also will the capacitance of this set up be 15uF if each cap is 150uF (150 / 10 = 15)???

Please help,

George

 

[Nigel Goodwin]

As I have just learned, it is not that simple. I need voltage balencing resistor that need to be able to handle high voltage(is it optional or is it necessary for performance), also will the capacitance of this set up be 15uF if each cap is 150uF (150 / 10 = 15)???

It's essential, with out the resistors I very strongly expect the capacitors to fail within a VERY short time! - with ten in series I would suspect that minutes would be all they would last! - and they could well fail quite dangerously!.

And yes, 10 in series would give 15uF.

However, I don't know how well it they would replace the single capacitor you gave the link to, it talks about >4000A pulse capability - your capacitors in series will have vastly less current capability, probably in single figures?.

 

[lord loh.]

I very strongly expect the capacitors to fail within a VERY short time! - with ten in series I would suspect that minutes would be all they would last! - and they could well fail quite dangerously!.

Why would it fail? Can't it be avoided if a safe margin could be maintained between the rated maximum voltage and the voltage that is to be developed between the capacitors in series.

Also I was wondering what happened to destroyed capacitors? DO they start to conduct? or do they become insulators? :?:

 

[zevon8]

If one capacitor failed, probably as a short, it would raise the potential on the rest of the string, causing the rest of the string to fail in quicker and quicker succession.

Think of a string of fire crackers with progresively shorter fuses. Pretty much how the caps will react to this situation also.

Differences in ESR between the capacitors would cause problems during discharge also, some caps would see higher losses, causing them to over heat. ( I get the impression you will want to rapidly discharge them )

 

[Nigel Goodwin]

Why would it fail? Can't it be avoided if a safe margin could be maintained between the rated maximum voltage and the voltage that is to be developed between the capacitors in series.

They would fail because you're exceeding their voltage ratings - electrolytic capacitors have very wide voltage tolerances (often -25%/+50%) this will cause one capacitor to have too much voltage across it and go S/C. In turn this will cause excessive voltage across the next capacitor which will go S/C, in a matter of seconds (or minutes?) ALL the capacitors will have failed - probably quite spectacularly?.

As I've repeadedly said, it's ESSENTIAL to have voltage balancing resistors across capacitors in series to prevent this problem!.

 

[George L.]

thanks for the help,

what resistance and what rating should the votage balencing reistors have ??? (radioshack is my main source of supplies).

Here is my set up:

9 total caps, all rated at 120uF 330V (photoflash)
(3 sets) three caps wired in seires for a total of 990V at 40uF. Three of these are connected in series to make a 120uF, 990V capacitor. Voltage balencing resistor across each capacitor.

Will this work?

Please help,

George

 

[George L.]

sorry,

I messed up on my last post... I said "Three of these are connected in series to make a 120uF, 990V capacitor. Voltage balencing resistor across each capacitor."

I meant: Three of these are connected in parallel to make a 120uF, 990V capacitor. Voltage balencing resistor across each capacitor

George

 

[ljcox]

George,
To make up 120 uF 990V using 120 uF 330 V caps you need 9 connected in series parallel with 3 voltage balancing resistors.

Len

 

[lord loh.]

Okay...

I was wondering of one more thing...

If I charged several capacitors parallely, and discharged them by connecting them in a series, What would happen?

1. They shall provide V1+V2+V3+…
2. The capacitance shall be reduced.
3. The charge have already been placed on the capacitors. (V=Q/C) but C is reduced…

All this seems confusing…

 

[ljcox]

Okay...

I was wondering of one more thing...

If I charged several capacitors parallely, and discharged them by connecting them in a series, What would happen?

1. They shall provide V1+V2+V3+… Correct
2. The capacitance shall be reduced. Yes and no. The capacitance of each capacitor is not changed. Each cap retains its charge. But the overall capacitance is reduced.
3. The charge have already been placed on the capacitors. (V=Q/C) but C is reduced… If all C are equal, then the total charge is Q and the cap is C/3. So the voltage is Q/(C/3) = 3Q/C which is 3V


All this seems confusing…

 

[George L.]

What reistance value should the voltage balencing resistor have.

Also, Ljcox, this doesn't make sense:

George,
To make up 120 uF 990V using 120 uF 330 V caps you need 9 connected in series parallel with 3 voltage balancing resistors.

Len

If I put nine in series the votage rating would be 4,770V :?

George

 

[ljcox]

George,
I would use 100k.

I did not say "series" I said "series parallel".

I try to draw it.
_________
|_|_|
|_|_|
|_|_|_____

Where each vertical line is a capacitor and each horizontal line is a wire.

I did not show the resistors

Len

 

[George L.]

ok, I get it.......

Do the resistors need to be able to handle high voltage???
I would like to use the 100K 1/2W resistors from radioshack because it is 2 minutes from my house.

Also, this is my first time using voltage balencing resitors for caps. Where do I put them??? I remmember you said I need three.

Please help,

George