Marx generators (MGs, invented by E. Marx) are a nifty way of increasing voltage AND current. Wait a minute, how can that be? Increase voltage AND current? Nope, there still isn't a free lunch in physics. The price paid in the case of the Marx generator is that it isn't a continuous power supply, it produces pulses of power. The pauses are necessary to allow the system to recharge. The principal on which it works is simple: Capacitors are charged in parallel and then discharged in series.

It's something like using rechargeable "AA" batteries. When recharging, the batteries are placed in parallel in the charging unit. This means it only requires 1.5V (actually somewhat higher) to charge the batteries, no matter how many are put in the charger.

When the batteries are used however, they may be placed in a series configuration, such as in most flashlights for an example. Batteries in series add their voltages, and so two "AA" cells of 1.5V each will add up to 3V when placed in series. So the batteries in the flashlight were charged in parallel, but discharged in series. The main difference between the capacitors in a MG and the "AA" batteries in the flashlight is that capacitors can accept a charge and discharge almost instantly, where as the batteries are charged and discharged over a period of time. For a schematic and more details, click HERE.

Pictured below is a quick and dirty Marx generator I whipped up to get some pictures of. Unfortunately, my nice MG failed due to abuse and was dissected and scavenged for parts long before I had a digital camera. The one below is charged with a 7kV/23mA OBIT, crudely rectified by the diodes you can see hanging off the transformer terminals. Unaided, the transformer will only jump a gap around .22 of an inch. With the MG, it jumps .73" inch easily, and occasionally longer.

   

The capacitors are 70pF/40kV doorknob style. Pity I only had three of them. The spark gaps are wire with a blob of solder on the end for a terminal. All of the components are mounted on sections of terminal strip, screwed onto a piece of plastic cutting board.

As I didn't have any appropriate resistors when I was making the MG pictured above, I went the inductor route. The inductors shown are 1/2 " cpvc (only because it's what I had laying around) with 22 ga. magnet wire wrapped on 'em, aprox 4" long. The center tap is a length of wire wrapped around a couple of stripped turns and twisted tight. I wrapped some larger inductors, and so far have found little difference. Here is a inductor picture.

Things to look out for when creating a MG: Arc over of components is a serious issue, and it can be difficult to add insulation or increase spacing later. Immersion in oil helps a lot, and tubing can be used on leads to increase insulation. If using resistors, physical length is a consideration. Too short, and they will arc over. This was the problem of this even quicker and dirtier MG pictured below that I made.

This extremely quick and dirty unit was born when a friend of mine happened to notice a schematic of a MG lying around my house. He was intrigued, and we decided to make one. In less than an hour, we were laughing at the arcs. The capacitors are 10kV/.001uF mylar, and are usually available at All Electronics. I forget what the resistors were, and as you can see in the photo, they have been covered in liquid electrical tape. This was because they were arcing over, as mentioned above. Finally, we ended up dunking the unit in oil. It was powered by a negative ion generator, also usually available at All Electronics. This little unit was meaner than it looks, and continued to function even after many of the resistors had burnt out. Presumably the carbon tracks left behind were sufficient to allow the system to charge. When charging, the oil swirled and moved around the capacitors, and items nearby were statically charged.

I am currently collecting components for a demonstration quality MG to replace my old one. The proposed design will use homemade rolled poly caps and inductors. I am hoping to use a negative ion generator powered by a hand crank 110vac generator, but I may have to resort to another power source. As with all my projects, a primary objective is to build the unit from inexpensive common items to encourage students to build their own.