Discharging a high voltage capacitor? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Discharging a high voltage capacitor?

1 2 next

Move Thread

LAN_403

Plasmaarc452

Thu Aug 12 2010, 05:09AM

Registered Member #1394 Joined: Sun Mar 16 2008, 06:18PM
Location:
Posts: 111

So I am designing a triple stage (1kj or less) coilgun for a school competition and this time I am set on doing it right and safely. My previous experiments with large capacitors showed that discharging with a screw driver is not safe, its scary, and very ghetto This is why I would like to implement a way to discharge the capacitors without touching them. It HAS to be digitally controlled no mechanical switches. I plan on being able to click a button on my computer screen and have the capacitors start discharging.

Right now I am in need of a little input on how to go about this. My best idea/plan is to use an scr to switch the three banks into a 5 to 10 watt resistor. The resistor would then dissipate the energy as heat discharging the banks in a minute or so.

Will that setup even work? I am trying to learn as much as possible here so if anyone out there has done this before could you please post some info anything will help.

Is there a better way to discharge the energy?

radiotech

Thu Aug 12 2010, 06:09AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

What is the voltage of the capacitors you wish to discharge? There are two ways to commutate SCR's, one is with a gate pulse, the other way is to apply a high enough voltage across the anode and cathode. So your SCR's must be rated at a margin above their off-state breakover voltage. Also, a SCR can be discommutated by applying a reverse current across the anode and cathode when it has fired. Typically this is done by discharging a capacitor through the A-K circuit. Not a happy thought in a maze of resistors and capacitors
being sequentally switched.

For life and limb I wouldn't trust a SCR crowbar.

Plasmaarc452

Thu Aug 12 2010, 06:34AM

Registered Member #1394 Joined: Sun Mar 16 2008, 06:18PM
Location:
Posts: 111

The capacitor bank will be at 500v in banks of 3 capacitors (9 caps in total making 1kj).

I am confused as to what you are saying about using an scr like this. WIll it work if I just want to apply a gate pulse and let the energy drain? Once it is in the process of draining it doesn't have to be stopped it can go until the capacitors are discharged.

This is purely a safety feature that just has to work, so if the scr is not a good option do you have any other suggestions on how I could go about doing it?

GhostNull

Thu Aug 12 2010, 07:23AM

Registered Member #2648 Joined: Sun Jan 24 2010, 12:45PM
Location: Australia
Posts: 291

A good method if discharging Large capacitor banks quickly is to use incandescent light bulbs. Your bank is 500v, so if your light bulbs are 110v then you will need 5 in series.

radiotech

Thu Aug 12 2010, 07:26AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Once you fire the SCR, it will discharge the capacitors through the resistors you select. As the voltage lowers, the current lowers until it falls below the minimun needed to maintain commutation. The capacitor will have a residual charge. This presumes the scr did in fact fire. (How would the digital safety system know that ?)

If you use 4500ufd total in 3 banks of 1500 ufd, and connect a bleeder resisistor of 10,000 ohms, or 3- 30,000 ohm (one for each bank) it will discharge the caps to a safe level in about 2 minutes.
Left on permanently, It will subtract 0.05 amps from the banks or about 8 watts from each bank, at the time of firing. (about 2.5 % of the energy) Bleeder resistors have to be permanently connected, not switched, if safety is involved.

Plasmaarc452

Thu Aug 12 2010, 01:57PM

Registered Member #1394 Joined: Sun Mar 16 2008, 06:18PM
Location:
Posts: 111

Yea I came across bleeder resistors in a video yesterday and thought about using them. That seems like the best idea, I am not concerned about losing 2.5% of the energy to implement such a safety feature.

The entire coilgun is digital I didn't really make that clear above. Everything from the power supply, charger, scr firing, optical detection system, and voltage measuring will be done via a large master board and many smaller slave boards. Each of these boards is fully accessible via usb allowing a laptop computer to essentially control the gun using my software. As far as I know no one has attempted to build something that is controlled like it will be.

Antonio

Thu Aug 12 2010, 11:07PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

Shield everything very carefully, or at each "bang" all your digital circuitry will go nuts...

Plasmaarc452

Fri Aug 13 2010, 02:50AM

Registered Member #1394 Joined: Sun Mar 16 2008, 06:18PM
Location:
Posts: 111

I had planned to use protections against back emf on the coils to protect the scr's. The circuitry connected to anything involving high voltage would be optically isolated to prevent any weird problems coming back.

Any suggestions on that note either? Anything else for protection I could put in?

radiotech

Fri Aug 13 2010, 04:09AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

As far as I know no one has attempted to build something that is controlled like it will be

You have described a SCADA system.

SCADA systems are telemeter/control used for everything in industry
but the controlled end has to have failsafe routines to drop back to local control if the supervisory link is corrupted.

Plasmaarc452

Fri Aug 13 2010, 01:57PM

Registered Member #1394 Joined: Sun Mar 16 2008, 06:18PM
Location:
Posts: 111

Thats what makes it so interesting, there aren't set plans from somebody else on how to do it! Making the coilgun is the easy part but adding the digital circuitry safely is another matter.

I am making special points about how it is controlled when my project is complete. I didn't know its actual name SCADA I always said industrial control technology but I will include that now in my final report on it.

For anyone out there that knows digital circuitry are there ANY MORE protections I could put in! Please the last thing I want is to fry my board over something that could have been prevented.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob