Sciencemadness Discussion Board

Super Capacitors

Blue Matter - 24-7-2013 at 11:46

I have 16 2.5v 2600 farad super capacitors I have had a while and I really don't need as many as I have so I am thinking of letting them go to some people who could use them for interesting things. These are really fun to play with all you need is a small power supply to charge them (higher amps faster charge).

I am selling these for cheaper than you can get anywhere on the internet only 18 dollars each plus shipping. they all work just have some writing on them and small smudges and scrapes.

http://imgur.com/uU9vGr0,70T4bPQ,cLp4vCW#0
http://imgur.com/uU9vGr0,70T4bPQ,cLp4vCW#1

[Edited on 24-7-2013 by Blue Matter]

watson.fawkes - 24-7-2013 at 16:08

I have four of these. Been meaning to make a capacitor discharge spot welder with them. Perhaps someone else will get to that first.

Blue Matter - 24-7-2013 at 16:40

I think 6 is sufficient for spot welder.

watson.fawkes - 24-7-2013 at 16:57

Actually, for what I want one for, which is tiny stuff (lead wires on cartridge heaters, for example), just one would do insofar as current delivery goes. I'll be using two in series because I want better efficiency in the charging circuit.

Blue Matter - 24-7-2013 at 17:17

I used 8 at one time in series it was really awesome but the sparks were filling up the entire garage they are really quite fun to play with.

Harristotle - 15-8-2013 at 02:28

I can vouch for this guy - my two arrived today.

I am very happy with these -they look awesome, about 500-600ml in volume each. They look like just what I need to weld different types of thermocouples.

Cheers.
H

Mr.Chemical1 - 10-9-2013 at 16:47

Do you have any left? If so would you happen to have an idea how much shipping to Canada would be?

AndersHoveland - 10-9-2013 at 18:07

I thought about connecting a long string of many small super capacitors, to potentially get a higher voltage high energy storage capacitor bank for a rail gun. Connecting them in series means they can discharge faster, important if one wants a fast rise pulse.

woelen - 10-9-2013 at 22:51

Connecting them in series makes the internal resistance larger, not smaller!
Connecting capacitors in series is bad practice, because it easily leads to breakdown of all capacitors in the string.

Imagine a string of three capacitors, which each of them are rated for 10 V maximum and break down above 12 V and all of them have the same capacity. Now suppose initially these are uncharged and you put 3 of them in series and then connect a 30 V power supply. Quickly you will have appr. 10 V across each of the capacitors.

Now we have the real world and the capacitors have some leaking with a high resistance. One capacitor has leaking resistance R, the other has R/2 and the third also has R/2. This is well possible. Capacitance is specified, leak resistance is not specified, it just is "high". When the 30 V power supply is connected to this string of capacitors, then you'll see that in the long run you get 15 V across the capacitor with internal resistance R and 7.5 V across the other two. The capacitor with 15 V across it will break down and become conductive. Then the voltage across the other two will become 15 V as well and these will also break down.

You can overcome this problem if you put so-called bleeder resistors in parallel to the capacitors, such that these are still very large, but much smaller than the internal resistance of the capacitors. The disadvantage of the use of such bleeder resistors is that the capacitor chain discharges more quickly when power is taken away (sometimes this may be an advantage, especially in high voltage circuits, because it makes the circuit less dangerous).

WGTR - 29-9-2013 at 15:15

Quote: Originally posted by woelen  
Connecting them in series makes the internal resistance larger, not smaller!
Connecting capacitors in series is bad practice, because it easily leads to breakdown of all capacitors in the string.

Imagine a string of three capacitors, which each of them are rated for 10 V maximum and break down above 12 V and all of them have the same capacity. Now suppose initially these are uncharged and you put 3 of them in series and then connect a 30 V power supply. Quickly you will have appr. 10 V across each of the capacitors.

Now we have the real world and the capacitors have some leaking with a high resistance. One capacitor has leaking resistance R, the other has R/2 and the third also has R/2. This is well possible. Capacitance is specified, leak resistance is not specified, it just is "high". When the 30 V power supply is connected to this string of capacitors, then you'll see that in the long run you get 15 V across the capacitor with internal resistance R and 7.5 V across the other two. The capacitor with 15 V across it will break down and become conductive. Then the voltage across the other two will become 15 V as well and these will also break down.

You can overcome this problem if you put so-called bleeder resistors in parallel to the capacitors, such that these are still very large, but much smaller than the internal resistance of the capacitors. The disadvantage of the use of such bleeder resistors is that the capacitor chain discharges more quickly when power is taken away (sometimes this may be an advantage, especially in high voltage circuits, because it makes the circuit less dangerous).


Ultracapacitors are partially self-balancing, because the leakage gradually increases along with the voltage across them. Leakage due to faradaic reactions (from too much cell voltage) will gradually destroy the electrodes and/or electrolyte, however.

I use active balancing networks when I design with them. In theory, all that is needed to balance each pair of cells is one op-amp and some resistors. Each network overlaps the previous one, enabling one to balance the entire series string of cells. For the 2600 cells, a circuit that can source and sink > 1 amp is adequate. This means that the output of the op-amp will have to be buffered with a pair of power transistors.

With such a circuit one can achieve quiescent current that is much less than would be attained if using balancing resistors.

Woelen is right about series resistance. The time constant of the cells themselves does not change when adding cells in series. However, if the load resistance is considerably greater than the series resistance of the cells, then Anders would be right. Doubling the number of capacitors in series would divide your time constant in half for a given load.

As an aside, woelen, I've always enjoyed looking through your website over the years.

Also, these cells made by Maxwell are used in electric vehicles, and even when removed from service are still very, very, good.

[Edited on 30-9-2013 by WGTR]

Blue Matter - 30-9-2013 at 11:09

I have some other capacitors I was thinking about making into a rail gun but never had the funds to complete it. It was going to be about 3kj with around 40 capacitors. If anyone is interested I have some 250v 3300mf capacitors much better suited for a rail/coil gun application.