Capacitor Charger -- Schematic Check / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Capacitor Charger -- Schematic Check

Move Thread

LAN_403

Killa-X

Sat Jan 01 2011, 02:24AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

I want to make a simple capacitor charger. One that is cheap, and simple. Plug it into the wall, flip the master switch, and it will charge. When you discharge the bank through the SCR, voltage usually remains in the capacitors. I have a 1Kohm 20W resistor that is triggered by a switch. Currently, I use a 600V 1A mosfet in series with a 1Kohm 20W resistor to drain my capacitors. It's a 980J bank, and the mosfet and resistor can discharge a full 450V load easily.

Why no transformer: To make things cheaper, I want to use a voltage quadrupler. In the past, I used one on a variac to charge my capacitors to 450V to reform them. I noticed when i found the correct variac voltage, it was able to charge the capacitors very fast. So now I want to simply put direct mains to it, Quadruple it, and have a OP-AMP monitor the charge. Once it hits the preset 450V, it will turn off the relay.

Ineed to varify this schematic is correct. I need to know if the relay will be on while its below 450V, and turn off while it's above or equal to 450V. The reference voltage will be 5V, and the divider will be 450V == 5V.

Any questions, let me know.

One other thing. I can get 75ft of speaker wire cheap. However, it's 24awg. If I have a 9V battery, regulate it to 5V, and have a 1000uF capacitor, will i still get close to 5V by time it reaches the SCR? I wasn't sure if the capacitor would help maintain that 5V, more than it would without a capacitor...

[Edit: Image size]

dmg

Sat Jan 01 2011, 03:03AM

Registered Member #2628 Joined: Fri Jan 15 2010, 12:23AM
Location:
Posts: 627

Well, ive been looking over the concept I showed you earlier.. one thing that still troubles me is the relay chattering when the op-amp it reaches its peak and triggers it to turn off (but then turn on again from a voltage drop maybe?). I hope that the RC at the negative line will be enough to retard this effect on the referance, and that a voltage drop will not be that significant to give false starts.. but the more I look at it the more im worried about the relay flickering rapidly.

Probably a one-shot circuit for when it triggers off will prevent it from trying to close again. (at least to reset the one shot after a certain ammount of time, where it will ignore the false triggering before turning off again.)
I could be wrong though.

Killa-X

Sat Jan 01 2011, 04:11AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

GBD - My variable charger, that was 0-450V stop, was like that. It had a relay that cut AC power to 2 transformers, hooked backwards for HV. This method did work, but it drew a LOT of amps due to saturated cores.

The point is, The relay did chatter. it would click like crazy because the SECOND that drops below 450V, it clicked on, then off, then on, off, etc etc to maintain that power. I remember we went over this once, and gave up and used a TRIAC as its silent.

Electroholic

Sat Jan 01 2011, 10:19AM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

How accurate does the end voltage have to be? might want to add some hysteresis there so your relay doesn't chatter.

Also im not sure if a 741 can provide enough current for your relay.

Add a back emf diode on the relay coil to protect the 741.

and maybe some isolation between the CW and your CapBank. Discharging the CW every shot is probably not good for the CW diodes.

Driving the SCR with a voltage source is probably not a good idea either.

Myke

Sat Jan 01 2011, 11:33AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

Electroholic wrote ...

How accurate does the end voltage have to be? might want to add some hysteresis there so your relay doesn't chatter.

Also im not sure if a 741 can provide enough current for your relay.

Add a back emf diode on the relay coil to protect the 741.

and maybe some isolation between the CW and your CapBank. Discharging the CW every shot is probably not good for the CW diodes.

Driving the SCR with a voltage source is probably not a good idea either.

Agreed to all points mentioned.

The 741 can't run off a single supply in most cases and a 5V single supply is definitely not enough.
You would want to use a comparator (which can and should have some sort of feedback for hysteresis) such as the LM393 or the LM339. They are designed for circuits like this. They aren't good at amplifying like op-amps but are better at switching and can usually handle a wider input voltage range. The comparators can only sink current so you would need to use a pull up resistor on the output (10k or so) and connect that to the base of a transistor. Since it's the switching device (if you chose to use a relay), you would need the protection diode from the emitter to collector.

For the multiplier, you would probably want to series some light bulbs to limit the current. When the capacitors are discharged, the multiplier sees a dead short and isn't very happy.

Don't forget that the multiplier capacitors need to have a voltage rating of twice the peak line voltage plus a little head room (except for the first cap which will only see the peak of the line voltage).

You should always have the bleeder resistor connected to the capacitors. Even then, you shouldn't trust the bleeder with your life.

Killa-X

Sat Jan 01 2011, 06:04PM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

I see. I'm use to the LM393, GBD designed the circuit on there with the LM741. Another thing I could do, is use a Triac instead. In the past, I made a dead simple variable voltage stop, and it used a Triac. This eliminated the relay chattering, and it maintained a SOLID voltage.

I did some basic tests using 400V 470uF caps. The peak current to my meter was 12A, but the analog read a max of 2A. The voltage on each capacitor, with 120V input, was 360V.

Currently, I have always relied on a 1K 25W resistor. Drains the bank in about 20-25 seconds. Using a volt meter, I can make sure it drains.

As far as protecting the CW diodes, The only way I can do this is to cut the circuit off by a switch I take it? Or would a simple resistor be able to help with this issue...

I've been told a relying fully on a comparator can be a little risky...I could use a crowbar, but id have to invert the signal so when the crowbar is off, it outputs high, triggering a relay. When it hits 450V, the crowbar puts out high, but inverted to low, turning off the relay, and stopping the charge....

Could a resistor in series from the CW and bank prevent issues like discharging..?

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