PowerLabs HomeBuilt Plasma Globe!

 Here it is at last… The definitive Plasma Globe!


Plasma Globe The objective of this project is simple: To produce the best plasma globe possible given the resources available. This means using the largest glass globe obtainable, and equipping it with the best vacuum system / electrode assembly / power supply I can obtain and/or make. This globe is meant to be used as a decorative piece for eventual exhibition, and should therefore be made presentable (nice base, no hanging wires, etc), safe (all H.V. parts enclosed, and shatter-proof glass). It should also last long and be able to operate for extended periods of time, which can only be achieved through a very well sealed vacuum system and a well designed power supply.


Glass jarHere is a picture of the flask. It is made by Schott, a German glass manufacturer, and is made from Duran� borosilicate glass (Special Thanks to U. Zimmerman for pointing that out to me), which is highly resistant to chemical attack and drastic changes in temperature. The flask is the round bottom, long neck type and has a capacity of 6000mL. It cost me $100 with the 5cm dia cork. The flask outer diameter is approximately 30cm (12″), and its walls are some 2mm thick (its HEAVY!)

 I am considering removing the writing on it… Initial tests with cheaper glassware and Hydrofluoric acid (HF) proved unsatisfactory (a frosted spot remains where the label used to be). Anyone have any better ideas of what to use?


Building Central electrode.Here is a picture of the center electrode being assembled. The electrode itself is a 26cm tall glass column with a 4cm diameter glass ball in the end. The tube is than filled with 25grams of aluminum filings, a high voltage insulated wire is inserted through the end and the tube is sealed with epoxy. Overall cost here (aluminum filings, glass for blowing the tube, and high voltage wire) was approximately 20 dollars.

 Assembled electrode inside flask: A hole was drilled into the cork that was approximately 80% the size of the tube, making for a very tight fit. The tube was than smeared with epoxy and inserted through the hole. Also notice the H.V. wire sticking out of the bottom.

Vacuum System:

Vaccum system partsHere you see the parts used for the vacuum plumbing. From left to right: Pressure rated (to 200PSI) nitrile tubing, for connecting the pump and gas tank to the lines, copper tubing for connecting the valves to the flask, tube clamps, tube adaptor, valve, extension, “T” junction, vacuum gauge, gauge adaptor, and so on… In the bottom there is another valve, the cork for the flask, and some liquid Teflon thread sealant (a must for any vacuum work), rated at 8000PSI.

Assembled vaccum systemAnd here is the assembled vacuum plumbing. There are two sections to it: The bottom section which includes a valve and a tube adaptor, and is meant to be attached to the gas tank and the top section which incorporates a vacuum gauge to the valve / tube adaptor assembly and is meant to be connected to the vacuum pump. The advantages of having the system split in two are obvious: The flask can be pumped down and backfilled with any gas simultaneously, and the pressure can be varied and monitored at all times. Hence all 5 factors (voltage, frequency, gas, pressure and power) can be adjusted in real time while the system operates, and when optimal conditions have been reached it can be sealed and should remain that way. This plumbing system cost me $50

Vaccuum pumpAfter months of searching around for an oil-sealed rotary vacuum pump, and finding that the cheapest model available cost $500 (no second hand units were found), I finally got a $15 deal on a small refrigeration unit pump (given its size it probably comes from an air conditioner or a very small refrigerator). It is *FAR* from ideal, but hey, it works! :). As a plus, this unit is unusually quiet (it can not be heard even when working at full power) and rather fast (pumps 6 litres down to its lowest pressure in under 3 minutes).
As a minus, it can barely reach 690mmHg (27″), and it gets rather hot in trying to do so! The pump weights 4,8kG and measures 20X10X15 (lXWXH).

The meter reads 690mm HgThe flask is baked and pumped down to the pump’s limit. Several other (higher) pressures were tried, but with air this proved to have the best results. I believe even lower (>700mmHg) pressures would be desirable with air, but have no way of testing yet. The vacuum system in the flask is flawless and so far has held its vacuum for 2 days without any changes in the vacuometer.

Results, Pictures and Videos!

 Below are 4 still frames of a 30 second video I took of the second test of this globe. As new and better test videos come, this and the pictures will be substituted (what you are seeing now is already version 5.0 of this page). For now the specs are: Air at 690mmHg, 50kV high frequency power source (de-powered Tesla Coil). The video shows arcs tracking my hand along the glass with the lights on, and than off. Click any of the 4 pictures below to download the 1.18mB .mpg video. Note: Tesla Coils are far from ideal power sources for plasma globes: The arcs are very hot and tend to move around and rise very quickly, making a rather frantic display. I personally prefer the slower moving arcs from a lower frequency power source.

Still frames from plasma globe prototype video