# constant current capacitor charge

Thread Starter

#### andrewmm

Joined Feb 25, 2011
243
Im sorry for this,

its 45 years since I was at university, and the dv/dt thing has well left me.

I'm also mainly a Ghz high speed digital design engineer , large FPGAs and CPUs , so this analog stuff I'm not to hot at...

So I have a 1.2 F capacitor "bank" , that Im going to have to charge up from 0 to 20 volts in 3 ms ,
what constant charge current do I need ?

I'm desperate enough, I might even have to simulate this ....

Many thanks

#### OBW0549

Joined Mar 2, 2015
3,306
I = C * dV / dT (units are amps, farads, volts and seconds).

C = 1.2F
dV = (20 - 0) = 20V
dT = 0.003 sec

Therefore I = 1.2 * 20 / 0.003 = 8000 amps.

Piece o' cake.

#### dl324

Joined Mar 30, 2015
10,525
So I have a 1.2 F capacitor "bank" , that Im going to have to charge up from 0 to 20 volts in 3 ms ,
what constant charge current do I need ?
Are you sure you want to charge with a constant current? What is the voltage you're using?

#### crutschow

Joined Mar 14, 2008
24,730
So I have a 1.2 F capacitor "bank" , that Im going to have to charge up from 0 to 20 volts in 3 ms ,
what constant charge current do I need ?
The basic capacitor equation is C = Q/V where Q is the charge and V is the capacitor voltage.
Since the amount of charge transferred is Q = I * t where I is the current and t is the time, we can substitute to get C = It/V.
Transposing gives the solution of I = CV/t or (1.2F*20)/3ms = 8000A.

(Opps, OBW0549 beat me to it).

#### ronsimpson

Joined Oct 7, 2019
614
So I have a 1.2 F capacitor "bank" , that Im going to have to charge up from 0 to 20 volts in 3 ms ,
Most capacitor will not like 8000A.
Maybe is should be 3s not 3ms. (8A)
The point is to watch the max. current of the capacitors.

Thread Starter

#### andrewmm

Joined Feb 25, 2011
243
Thanks for the equation,
Yep I need to work on the numbers, 8KA is a lot ,,,
but I needed the equation, so I gave some round about numbers, to get me started,

Thank you all for the very fast and detailed response,
Im sorry my analog brain could not work it out !!

#### AnalogKid

Joined Aug 1, 2013
8,378
E C = I T

jjw
Thread Starter

#### andrewmm

Joined Feb 25, 2011
243
Thank you all,
with your instructions,
I have convinced the team to go for
10F , 20 Volts , but 5 seconds charge time,
which is only 20 amps, and the repeat is only once a minute, so I have time to cool down the charger.

#### MrChips

Joined Oct 2, 2009
20,904
Thank you all,
with your instructions,
I have convinced the team to go for
10F , 20 Volts , but 5 seconds charge time,
which is only 20 amps, and the repeat is only once a minute, so I have time to cool down the charger.
40A

#### Papabravo

Joined Feb 24, 2006
13,533
Somebody needs to slow down and double check his computations. You're fooling around with things that could have serious safety implications. Do you have any concept of how much energy is in a 10F capacitor. Have you ever seen a large capacitor explode -- with shrapnel?

Thread Starter

#### andrewmm

Joined Feb 25, 2011
243
You want to see what its powering,

The main problem I had,
was I'm a digital engineer, in a world of physicists ,

and I could not convince them that we might have a problem,..
They are real experts on what they do, but some times .....

Thats why I needed your expert input,
it got them to re think things,
I'm "just" doing the digital controller,
At least they admit that designing a 52 Gb/s link is not so easy,
which I have done a few of,

even if I cant do basic maths any more,,

Thank you

#### AnalogKid

Joined Aug 1, 2013
8,378
Agree with Papa. 2000 W-s is nothing to play with. Be sure to check the cap datasheet for max current, max dV/dt, di/dt, etc. If that were a photographic strobe, it would be the energy for a 1 million watt flash.

ak

#### MrChips

Joined Oct 2, 2009
20,904
That is huge amount of charge. Rather than having to guess what you are doing it would help if you told us.
There are not too many things in AAC that are that TOP SECRET.

#### Papabravo

Joined Feb 24, 2006
13,533
You want to see what its powering,

The main problem I had,
was I'm a digital engineer, in a world of physicists ,

and I could not convince them that we might have a problem,..
They are real experts on what they do, but some times .....

Thats why I needed your expert input,
it got them to re think things,
I'm "just" doing the digital controller,
At least they admit that designing a 52 Gb/s link is not so easy,
which I have done a few of,

even if I cant do basic maths any more,,

Thank you
My words of caution were not meant to denigrate your talents and abilities. Once upon a time I was just the "firmware" guy. As I got more involved in embedded systems design and development, I made it my business to understand the physics of whatever I was working on for my own peace of mind and well being. My come to Jesus moment was a 100 KV power supply controller for automotive paint spray booths. As you might suspect 100 KV @50 uA will knock you into the middle of next week if you are not careful. I want you to live long and prosper.

Thread Starter

#### andrewmm

Joined Feb 25, 2011
243
Agree with Papa. 2000 W-s is nothing to play with. Be sure to check the cap datasheet for max current, max dV/dt, di/dt, etc. If that were a photographic strobe, it would be the energy for a 1 million watt flash.

ak
Thank you

Now I have the experts on board where I'm at , Im much happier,

I'm keeping it secret, not only for the client, but to protect the people that made the initial decisions,

Thank you guys,

#### shortbus

Joined Sep 30, 2009
7,604
@andrewmm, are you from Australia by any chance?