Marx Generators

Incredibly high voltage via voltage-multiplication.

 27 Aug 2003 

A Marx Generator is an extremely dangerous High-Voltage pulsed power voltage multiplier device. It CAN kill! Make sure that no one can come in contact with this - particularly curious onlookers and yourself. Separating the potential victims from any possible contact with the high voltage is really the only foolproof way of protection. Always keep a safe distance from your Marx Generator!
 
Note that a Marx generator deals not simply with high voltage, but multiplies it even further! In addition, this energy is all stored in high voltage capacitors, making them significantly more dangerous. All capacitors should be discharged carefully with a probe before servicing. In addition, sparks generate UV, Ozone, loud sounds and are of course electrically dangerous. Eye wear, ear protection and insulated probes are recommended. Finally, Marx generator discharges generate significant radio frequency interference (RFI), and may mess up TV displays or even pace-makers. Do not operate a Marx generator near anyone with a pace maker! The information here is provided for education only.

Introduction

Do you like the idea of tesla coils and other high-voltage sparking stuff, but don't have the time, money or patience to build something that elaborate? If cheap loud and long sparks are your aim, a Marx generator delivers them with ease and simplicity! Big versions of them are used in museums to demonstrate electricity and lightning(along with Cockcroft Walton Multipliers), but nothing beats the low-cost, ease, and amazing results capable of a Marx generator.

Basically, a Marx Generator is a simple but clever way to generate lots of voltage. A Marx generator is a type of voltage multiplier and the idea is surprisingly straightforward - charge up a stack of high voltage capacitors in parallel such as with a several kV flyback or other HV source, and discharge them in series. When you use high voltage capacitors and spark gap, the resulting sparks are quite impressive. Unlike Cockcroft-Walton multipliers, the Marx Generator needs no expensive diodes, and only uses one capacitor per stage. This reduces their cost enormously, at the expense of the mechanical complexity of the spark gaps.

Marx generators also make excellent lightening simulators, but are also very hostile to surrounding equipment due to the EMI and RFI generated during discharges which can run in the several thousands or more amps! Always remember to  discharge each capacitor carefully before handling it. These capacitors can store a serious amount of energy, and you wouldn't want them to discharge all this stored energy into your fingers. 

This project was inspired by Alan's High Voltage lab page and credit goes to him for getting me started in the world of Marx Generators!

How it works

The Marx generator consists of an array of resistors, capacitors and spark gaps arranged as above

The above is a general schematic of how a Marx Generator is laid out. Note that you can add as many 'stages' as you desire!

The capacitors (in green), are charged up in parallel via the 1M (one megohm - value not so critical, but 1M or above should work) resistors, so they each become charged to the input voltage from the high voltage DC power supply unit (there is also an added 10M resistor for current limiting and protection, the exact value is not critical). When the first (leftmost) spark gap breaks down, the voltage across the next stage increases (since the spark gap effectively connects the capacitors in series), causing it to break down, and so on for all the other gaps in a cascade fashion.

When all gaps have broken down, the low impedance of the ionised air in the sparks effectively connects all the charged capacitors in series, multiplying the input voltage by the number of capacitors (or stages). The ionised air path has sufficiently low resistance that the charge resistors now look like an open circuit. Thus the Marx generator is basically a smart way of charging capacitors in parallel, and discharging them in series, creating extremely high voltages.

The 10 MegaOhm resistor also has a ballasting effect to limit short-circuit current, and is needed to avoid a continuous arc from the HV power supply forming across the first gap after it fires (this prevents further firing). The value will depend on the type of power supply used. High values will reduce the maximum spark repetition rate, so you may be able to get more 'bangs per second' by reducing the resistor value and vice versa.

Although it is possible to make a Marx generator with just an array of resistors, capacitors and spark gaps, it can be hard to make it fire reliably, as it will depend on the breakdown voltage of the spark gaps, and there can be a fine line between not firing and firing before all the capacitors are fully charged. One solution is to initiate the breakdown of the first gap mechanically, e.g. by waving a suitably insulated screwdriver between the electrodes, or pushing the first gap contacts together with an insulating rod.
 

Components.

You only need...
1. High Voltage Capacitors
2. High Resistance Resistors
3. Solder, Wires, etc...
4. High Voltage Power supply unit

The most important thing is to make sure that the components are suitable for the input voltage used. This includes the resistors - normal 1/8 or 1/4 watt carbon or metal film resistors are not suitable for high voltage uses like this, since they are physically small and therefore will arc over! Larger (1 or 2 watt) carbon resistors, or high-voltage metal glaze resistors should be used not because of their greater power rating, but because they are physically longer and prevent arcs from arcing across the surface of the resistor. Alternatively, several standard resistors can be used in series, in such a way that each resistor sees no more than its voltage rating (typically about 500V). For example, ten 100K 1/4 watt resistors could be used in place of each 1M HV resistor. The value of the 1M resistors is not too critical - higher values will increase recharge time, but lower values will increase losses as some of the spark discharge leaks back through the resistor chain, especially with smaller value capacitors.

Capacitors should ideally be ceramic types, as these are best suited to the fast pulses in a Marx generator - polypropylene pulse-rated caps would probably also work well, but are more expensive, and harder to find with high voltage ratings. Use the highest voltage types you can find. Larger values will give fatter sparks, but take longer to recharge... You can often run these in excess of their rated voltage, but obviously this runs an increased failure risk.

An input voltage of about 4-8KV is recommended. Above this, many problems will detract from the 'quick and dirty' approach, requiring much greater care in construction to reduce corona losses from sharp edges, multiple series resistors would be needed, and the capacitors will be harder to find. The high resistance of the charging network mean that corona losses can significantly degrade performance above about 5 KV unless care is taken to avoid any sharp edges - e.g. by making wire ends and joints smooth.. soldering blobs of solder at sharp parts ... potting the whole setup in insulation or oil.

A simple design requires a low-current source of high voltage DC in the range 4-8KV, such as low-powered flyback generator. The number of stages you use will determine the output voltage, and hence spark length.

Marx Generator 1

 Monday, 27 Oct 2003 

This is my first Marx Generator. Marx Generator 1, model 1.

It's simply a bunch of wires, resistors, and capacitors soldered up together. Very simple. These are 1nF (labeled on the cap) / 810pF (labeled at the shop where i got them from? Quite close values anyway) 15kV ceramic caps. They have a nice insulation and generally should be good caps. It's a pity I only got 7 of them. All resistors are 1W 1 MegaOhm resistors. At these lower voltages, arc-over should not be a problem... I hope.

Here are the caps, resistor and power supply. The resistor is a 1W 1M resistor. The cap on the left is the 15kV 810pF cap (the one I am using) and the cap on the right is a 30kV 510pF cap. I am using the 50W Liyoda electronic halogen transformer (read more about flyback driver circuits at my flyback page here) to drive the flyback which drives the Marx generator. I am using the small flyback (the one closer to the resistor). It's from a black and white TV and with my wimpy set up, it outputs around 5kV at around 10mA DC (estimated. I need a HV probe..). Perhaps even less.... I also need a better power supply! My caps are definitely overrated (at 15kV!) , but that is not a bad thing :).

Here is a close-up of the construction effort. Notice the excess solder usage and the loop-surface-blob gaps. Although not important for correct operation, they are a nice professional touch, making everything more reliable and repeatable, and reduces sharp points for corona loss. It also looks nicer than bent bits of wire too. This all-smooth-surface construction has proven to be very useful :-)

Gap spacing and charging current are important. The gaps should all trigger at once. In practice the first gap fires and the over-voltage avalanches up the stack, making them not too critical. Experimentation should be made! Therefore, I made all the gaps above the first wider, and controlled the trigger voltage from the first gap by using a plastic rod to trigger. If your charging current is too high, the first gap may light continuously. This was quite a problem for me, but I fixed this by adding more 'ballast' resistance. For now, I'm using a piece of plastic to trigger the first gap...

My current setup (model 1) has 7 stages with an estimated 4kV+- per stage, equating to about 30kV output. Which each cap storing 810pF, and total capacitor charged at 30kV, the whole discharge is 115.7pF at 30kV... 1/2 x 0.0000000001157 x 30000^2 = 0.05207Joules per bang.. not much actually, but produces large healthy 3cm arcs as can be seen in the photo below! First light! It works :D. It's loud, nasty, and really awesome. The blurry orange object on the right is a piece of plastic i used to trigger the spark gap. Notice how bright the arcs are :D

Unfortunately, my 50W power supply unit died soon after this photo was taken :(. The flyback, however, was unharmed. That's all for now, until I get a new power supply. I am considering adding more stages to it. The electronic transformer probably died due to some EMI which went back into the circuit, destroying the electronics. My loss is your gain in knowledge!

Marx Generator 1 Updated

 15 Mar 2004 

Due to the sudden death of my flyback driver PSU, I had to discontinue my high voltage flyback experiments, and this lead me unable to charge up the marx generators capacitors in a suitable way... until now...

I have recently acquired a SONY brand Switch Mode Power Supply unit. 12V at 34A! This is MORE than enough power for driving my flyback! (For more information on the flyback charging circuit, check out my flyback driver page). I used a small flyback from a black and white TV... it's a 'modern' type flyback and is already rectified (outputting DC). Everything has remained the same as Marx Generator 1, a new higher voltage charging supply is used! I also made sure to employ a large amount of filter capacitance to help smooth any switching noises.

So I plugged it in...

Much better! The voltage is MUCH too high! This is about the maximum I can open the spark gaps.. more than that the resistors start arcing over, the capacitors start arcing over and .... well just about everything arcs over! The spark gaps are around 1.3cm... which suggests around 15kV charge in the capacitors... 15 x 7stages yields 105kV! That's about 0.56J per bang. Very loud and bright. It fires itself about twice a second. In the photo, the main gap is 7cm, but the arc can jump a 10cm gap or more! A lot of corona is evident (From the hissing sound and the large amounts of ozone produced), and a lot of energy is wasted there. I think I need higher voltage capacitors... Immersing it in oil would reduce much of the insulation problems, but I have yet to tried it.

This is a success so far!

Back to main page
(c) Gao Guangyan 2011
Contact: loneoceans [at] gmail [dot] com