This is our plasma globe charging set-up. It is very crude. The vacuum pumps are compressors removed from an old air conditioner and an old refrigerator. The air conditioner vacuum pump(middle) can pull near full vacuum. The other unit(right) can only pull about -20" Hg, but a higher volume. It is used initially to reduce the load on the high vacuum unit. On the upper left you see our valve manifold. This is connected to a reserve tank(not shown) which we pump down to full vaccuum and use as a source of vaccum during the charging process.
Our methodology for charging is crude and cheap. To do the job accurately and scientifically would require thousands of dollars of equipment. This is a basement hobby for us so our "rig" meets our needs.
Here is the valve manifold with a bottle of Krypton attached. Below the bottle of Krypton is a valve which interfaces from the bottle connector to 1/4" NPT and then connects to the brass "charging nipple" which is has a shut-off valve on inlet and outlet to isolate it. On the left you see a mercury manometer used to measure the vaccuum accurately. In the background is a round vacuum pressure gauge used to get confirm that the reserve tank is a full vacuum. We have used Neon, Krypton, and Xenon in our Plasma Globes. They all can be obtained in disposable 10 Liter bottles as shown. These bottles have the same connection on top as a typical propane torch tank.
This picture and the one below show the valve manifold in detail. The red valve(#1) in the back isolates the reserve tank from the system. The next valve(black handle)(#2) isolates the vacuum pumps. The next valve(#3) isolates the mercury manometer. The next valve(#4) isolates the "charging nipple" from the vaccuum system. The valves(#5) on the inlet of the "charging nipple" were previously mentioned. The last valve(#6) isolates the plasma globe. In the foreground is the tubing that connects to the globe. The size of the charging nipple is the key to controlling the volume of gas that enters the globe. One starts the process with valves #1,#2,#3,#4,& #6 open and #5(#5 is two valves) closed. Start the vaccuum pumps and pump down the system to full vacuum. Now close #6, #3, and #2, and #1 and shut down the vacuum pumps. Then open #5 and this will pressurized the charging nipple(and some of the piping) with gas at the pressure in the bottle. Close #5 to isolate the gas tank. Open valve #6 to allow the gas to flow into the globe. To verify the gas pressure in the globe, close valve#1(valve#2 should remain closed, valve#3 open), and open valve#4. This will connect the manometer to the globe and verify pressure. The process then repeats closing #3 & #6, opening #5 to re-charge the nipple, then opening #6 followed by #3 to verify pressure.
The number of times the process must be repeated is dictated by the volume of the brass "charging nipple". The gas bottle will charge the "charging nipple" to the pressure in the bottle. The length(and diameter) of the nipple determines its volume. Thus the gas will expand to fill the globe and the pressure will drop based on PV=nRT. In our case the volume of our 12" dia globes is approximately .523 cu Ft and we target -22" Hg(3.9 PSIA) for globe gas pressure. We must repeat the charging process many times to charge the globe to our desired pressure. We have a few different size nipples and adapters to speed up the process. Since we try for 95% Neon, 2.5% Krypton, and 2.5% Xenon then the "charging nipple" shown is used to add the Krypton and Xenon. After charging and testing the Globe should be permanently sealed.
Warning: Opening valve #5 and #3(with #4 open) at the same time will subject the manometer to the high pressure in the gas bottle and blow out the mercury. Opening valve #5 and #6 at the same time will charge the Globe to the high pressure and likely cause the globe to explode. Also, be aware of the danger of the globe imploding at full vacuum. All of these situations expose the operator to danger and proper safety precautions should be taken.