Hello everyone,

I am pretty well versed in low voltage circuits, but I don't really grasp how high voltage / capacitator banks work.....

What I am looking for:

I need a capacitator bank that I can charge with 24V / 5A. The bank should be able to discharge 1200W in a millisecond. The idea is that this will power a Xenon Flash lamp, as a component of a ruby solid state laser.

Charging could be a higher voltage as well... But I am stumped when it comes to what type of capacitators I would need. Do I need charging voltage capacitators (24V?) of output voltage? Do I put them in series, or parallel? How do I know how fast a capacitator can discharge?

Any help would be greatly appreciated!

 

The time it takes a capacitor to discharge is given by the solution to a differential equation. You pick an initial voltage and another lower voltage and that will allow you to compute the time it takes for that to happen.

ETA: BTW It is possible to compute the energy stored in a capacitor, measured in Joules, but it is not possible to compute the power delivered to a load, measured in watts, unless you know the impedance of that load.

 

Hello everyone,

I am pretty well versed in low voltage circuits, but I don't really grasp how high voltage / capacitator banks work.....

What I am looking for:

I need a capacitator bank that I can charge with 24V / 5A. The bank should be able to discharge 1200W in a millisecond. The idea is that this will power a Xenon Flash lamp, as a component of a ruby solid state laser.

Charging could be a higher voltage as well... But I am stumped when it comes to what type of capacitators I would need. Do I need charging voltage capacitators (24V?) of output voltage? Do I put them in series, or parallel? How do I know how fast a capacitator can discharge?

Any help would be greatly appreciated!

A capacitor charges/discharges exponentially, so look up a thing called "time constant".

 

Here is an approximate solution for a particular combination of capacitance and load. The discharge power dissipation may look linear over the short time period, but it is not. It is also very noticeably not constant and in order to get at least 1200 watts average you need to start higher than 1200 watts.



I could get closer to 1200 watts, but you get the idea.

 

With a load of 0.45 ohms you need to look at the ESR of the capacitors. (internal resistance) Some caps have such high resistance that that you cannot get that much current out.

 

With a load of 0.45 ohms you need to look at the ESR of the capacitors. (internal resistance) Some caps have such high resistance that that you cannot get that much current out.

You also need to check the ESR of the capacitor bank because those ESR's will be in series and parallel.

 

For a flash tube, the charged-up capacitor voltage usually is hundreds of volts, like 300-400 V, not 24 V. 24 V is not high enough to sustain conduction after the lamp is triggered.

Not to rain on your parade or anything, but . . .

1200 W over 1 ms equates to 1.2 MW-s. Emphasis on mega. That's a lot. That is going to take a huge pile of non-standard capacitors, plus ultra-low inductance wiring. Look into photographic supply stores, see what a 1 kilowatt-second flash rig looks like, and multiply that by 1000.

Here is a link to a 2400 W-s supply. It is 1/500th of what you need, and $2700.

https://www.bhphotovideo.com/c/product/50036-REG/Speedotron_11260_2403CX_Power_Supply_120V.html

ak

 

24V will not operate a xenon flashtube. It takes about 300V plus a multi KV pulse to initiate the discharge. Do some research before doing this.

 

1200 W over 1 ms equates to 1.2 MW-s

No, it is 1.2 W sec also known as Joules.

 

No, it is 1.2 W sec also known as Joules.

With an initial value of:

\( E_{C}\;=\;\cfrac{1}{2}CV^2\;=\;\cfrac{1}{2}\times100\times10^{-3}\times(24)^2\;=\;28.8\text{ J} \)

At least we know that our hypothetical capacitor has sufficient energy to deliver 1.2 Watt-sec and then some.

 

Thanks for all the explanation guys / gals!

So, if I understand correctly, I need a capacitator that can deliver 28.8J for it to be able to deliver 1200W for a milisecond pulse, with a max ESR of 450 mO. That should not be too much of a problem.

As for voltage, am I correct in assuming that I can charge a kV capacitator with a 24V power supply? Charging would just take longer right? Or am I mistaken in that? (I checked out camera bulbs and they deliver 300V from a 1.5v battery).

Still some things I am wondering about:
I researched trigger voltages and found that the rule of thumb is 300V + 100V per inch of tube. The tubes I am looking at are roughly 11 inches, so then I would need 1400V? How does the wattage of the bulb come into play then?

For 28J at 1-2KV, I need about 56uF capitance. As far as I can find, there are no single capacitators that do this. Do I then need to put for example 10 5.6 uF capacitators in Parralel?

Thanks in advance.

 

NO. The total of the load impedance and the ESR cannot exceed 450 mΩ

 

So, if I understand correctly, I need a capacitator that can deliver 28.8J for it to be able to deliver 1200W for a milisecond pulse,

No, it requires 1.2J theoretically. But the voltage drops as you discharge, so maybe 10 times that would be practical. Don’t know where you got 28.8J.

As for voltage, am I correct in assuming that I can charge a kV capacitor with a 24V power supply?

Yes, but you need an oscillator and transformer to boost the voltage.

The trigger does not need much current or energy at all, it is usually generated by pulsing a coil.

You can easily find circuits that do all of this.