Building a 365kJ Capacitor Bank for Experiments | Abiscus.com

[sepulker]

I am pursuing a degree in Electrical Engineering with a minor in Physics, and I am almost done with my 4th out of 5 years ar Northeastern University (go SOX!) in Boston. I have a solid grasp of most theories related to the field.

However, exploiting my knowledge upon a physical device for the first time is different from using a pencil and a piece of paper. I am sure anyone will agree that there is no shame in double-checking yourself and questioning others with more experience.

I plan to begin experimenting with very high currents, and to do this I would like to build a capacitor bank. By luck I already have two Mallory 1800MF 450V capable electrolytic capacitors. Obviously, the energy I could store in these guys seems to be: (1/2)*(1.8)*(2)*(450)^2 which leads to 364.5kJ!

This seems like an awful lot of power, and also seems fansastically dangerous!

I plan on charging the bank with a rectified transformer (probably a microwave transformer since they are free from the junk yard) under a voltage divider for 450V. I will add safety devices like a complete 1/4inch plastic housing, bleeder resistors, and a voltage monitor, plus a huge red light that warns me not to go near the thing.

A few questions (since I have never made something like this before)

1. Is 364.5kJ correct?
2. What are the not-so-obvious dangers?
3. What can I use to discharge it safely?
4. Anything else that I should know/research before attempting this?
5. Once it's working, what devices can I use it for?
6. How would a Cockroft Walton voltage mult. config. affect energy and such?


If you would like to check out my website, it contains a lot of projects I have completed in the past:



www.abiscus.com


Thank you for your time!
-Nate

 

[Cliff_J]

Are you certain that its not microfarads instead of milifarads?

 

[sepulker]

Yeah, they definitely say 1800MF (and not 1800UF)

 

[Averagesupernova]

That doesn't mean that it isn't microfarads. I have seen caps printed that way. Physically they would need to be HUGE in order to have a capacity that large at a high voltage.

 

[willib]

Nate i have to ask , how large are they physically??
dimentions in inches?

 

[sepulker]

They are probably about 8 inches tall with a diameter of about 4 inches (which would make a lot more sense for a smaller capacitance)

So industry standards aren't as concrete as I thought?... some companies use an "M" and some use a "U" for microfarads?

 

[Averagesupernova]

If you still aren't sure why not just charge it up and measure the charge/discharge time? You don't have to charge it up to full voltage.

 

[sepulker]

Because I'm on vacation and I don't have them with me haha.

But you answered my question about the markings. Now I am almost 100% sure theyre microfarads. Thanks

SIDE QUESTION:
I don't want to spend time doing the analysis...

What are the differences in charging time, energy, etc of Cockroft-Walton voltage doubling configurations compared to other methods?

 

[Manchot]

Yeah, the largest capacitor I've ever seen sold is 5000 Farads, and it was quite expensive (since it was meant for industrial use).

 

[willib]

You also made a mistake with the formula 1/2CV^2..
unless i am mistaken your new Joule calculation is .5(0.0018)(450)^2 = 182.25 J

 

[exequor]

you have some nice projects on your website sepulker; I like the night vision project in particular.

 

[faust9]

You also made a mistake with the formula 1/2CV^2..
unless i am mistaken your new Joule calculation is .5(0.0018)(450)^2 = 182.25 J

He has two caps (In parallel I presume) not one.

 

[sepulker]

Right, I was assuming two.

I have another question:
I want to increase the maximum voltage I can put into the bank. Assuming all capacitors are the same model... is it healthier to:

1. Stack capacitors in series
2. Use a voltage doubler like the Cockroft-Walton

It would be interesting to see the pros and cons

 

[willib]

Right, I was assuming two.

I have another question:
I want to increase the maximum voltage I can put into the bank. Assuming all capacitors are the same model... is it healthier to:

1. Stack capacitors in series
2. Use a voltage doubler like the Cockroft-Walton

It would be interesting to see the pros and cons

in my opinion , for raw power , stack them , besides the current output of the Cockroft-Walton and its variations is low, on the order of 100mA , although they do talk about one that looks like it could produce more current , it has sort of a complex triggering scheme because it doesn't use P channel FETS because at the time it was written they were more expensive than they are now.. http://murray.newcastle.edu.au/users/students/1999/c9221792/doubler.htm your doubler circuit
how much voltage are you looking to make?
Power fets are only rated for around one hundred volts or so , so stacking them would be your only choice..

 

[willib]

did you know that there are high voltage capacitors in microwave ovens , on the order of 2100V ..
@ 1.1uF that comes to 2.4255 Joules..

 

[sepulker]

On Ebay, I found Nippon Chemi-Con capacitors rated at 400V 3600UF. The guy has 60 available. So let's see... each one can hold 288J each. If I buy 10 and split the stack, that's 9000UF at 800V (i think that's right?) at around 2.9kJ. I can stack it more for a better approach to a spark gap... actually...

What would be better for switching the current? A spark gap? Thyristor SCR, Thyratron? (Maybe I don't have to stack them)

If I assume my load is a coil and it has a impedance low enough to allow a fast pulse, which should be enough energy to crush a soda can, right?

This would be perfect for my demonstration.