Neon signs are without a doubt one of the most ubiquitous advertising and art forms of the modern times. Very few can say they’ve never seen one–be it in a shop window, a beacon of light on a dark and lonely highway, or in a movie. Yet in spite of how commonplace they are, and that they’ve been around for over 100 years now, very few know how they work and even fewer know how they are made. While it is not really possible to do justice to the teaching of this subject in a printed form–one must ultimately experience the process first hand, it is my hope to be able to shed some light on the subject, answer a few questions you may have, and maybe spark some interest in the reader to pursue their own neon adventure or at least have a greater appreciation for the melding of art and science that takes place in something as mundane as a red neon “OPEN” sign in the store window.
Neon and Argon are among the noble gases in the periodic table of the elements and are the most commonly used in what we generally refer to as a “neon sign.” Other noble gases such as Helium, Krypton, and Xenon are also used for specific effects but these uses are less common. Each of these gases, when ionized with electricity will give off a characteristic glow of light as the electrons of the atoms are excited and then fall back to their ground state, emitting photons of light in the process. Neon gives the familiar reddish/orange glow that we see in the common open sign. Argon gives a blue color. Often a tiny bit of mercury is added to blue pumped units to make them brighter and more brilliant and mix of Ne and Ar is used to improve cold weather vaporization of mercury in outdoor applications. Other colors are created with phosphor coatings on the inside of the glass tube. Shades of white, orange, yellows, gold, green, purple, etc. All the result of phosphor coatings being excited by the base light from the ionized gas inside. If you have ever accidentally broken a common household fluorescent lamp and had the powdery mess to clean up, you understand what this internal coating is like. In fact, a common fluorescent lamp is nothing more than an Argon filled, coated tube but instead of a high voltage being used to provide the electron excitation a lower voltage is used in conjunction with a heater element at each end. Fluorescent lamps are therefore referred to as “hot cathode” lights. Neon is “cold cathode” lighting. (often “Cold Cathode” is used to refer to large diameter units, but they work the same) Cold Cathode requires a high voltage transformer, on the order of 2000 to 15000 volts depending upon the diameter and length of tube. (more on this later) Due to the lack of a heater filament, however, neon and cold cathode lighting have a VERY long service life. Neon is one of the longest life manmade lighting sources that has ever been devised. It is not unheard of to find 50+ year old neon units that still function….and some have been found running after 75 years of operation!
Typically, most neon signs are simple two dimensional designs–this is by no means a requirement but for sign work it is most common. Myself and many others have used neon in freehand, three dimensional artwork as well, but again, most of what you see out there was bent on a flat table to match a pattern for the letters or designs to be depicted in the completed sign. These patterns are usually drawn in reverse, ie: what you see on the pattern is the backside of the finished work. The reason for this is that it allows the tubebender to follow the pattern and make any crossovers, doublebacks, or other bends to get from one letter or part to another and have them elevated in a different plane from that of the front facing part. This way, when the work is finished, the background parts can be blocked out easily so as not to detract from the finished piece.
When one designs a pattern and begins to work on the bending of the tubing, it is important to always bear in mind that the tube begins as a straight piece, usually 4 ft in length…..you must plan the bends: Rarely can you simply start at one end and work forward….usually a letter or feature will require some intermediate starting point and sequence of bends that is not intuitive from just looking at a completed letter. See the sequence shown for bending the letter “R” for example…and note that the pattern is backwards.
During the bending process, the other important goal is to try to maintain the diameter and wall thickness of the tube as much as possible throughout the piece. The primary reason is to avoid thicker and thinner parts or constrictions that will result in internal stresses in the glass as it cools. Such internal stresses can lead to a weakness and eventual breakage in that spot. A blow hose and swivel is therefore used to enable one to help maintain the dimensions as a tube is heated and bent, thereby avoiding the “kinked garden hose” effect.
A quick word about torches: There are a lot of different types that can be used to work glass, for neon work in the US, the crossfire and ribbon burner are most common. In scientific glassblowing and in other countries where Pyrex (borosillicate) glass is used it is common to see gas/oxygen burners such as the Carlisle CC or similar bench burners used. The choice of fires is often a matter of type of glass being worked, training, and personal preference. (I could probably write a book on this aspect alone, but luckily for you I’ll leave it be for now.) In addition to a small hand torch for welds and “tipping off” purposes my primary fires are a 5-point crossfire (ten tips, five on a side pointing at each other) and a 24″ ribbon burner. The ribbon burner has a slide that allows me to vary the length of the flame from a few inches long to the full 24″ length of the head. The ribbon is intended for large arcs (making the letter “O” for example) and the crossfire is used for most other bends and welds. These torches use propane and a regenerative air blower to run them.
Once the glass has been cut and bent to the desired shape with electrodes fitted and sections joined, etc, etc it is time to pump, bombard, and fill the unit. This process involves connecting the unit to a manifold with gauges, a high vacuum pump, and metering valves for the rare gases. When I first began my own neon journey, the homemade glass manifold with glass stopcocks was the order of the day and it was somewhat of a right of passage for a tubebender to fabricate his or her own manifold. These manifolds worked very well but could be a little temperamental and hard to clean. (cleanliness is necessary to achieve high vacuum and a long life tube!) About the time I started, the Townsend metal manifold was the “hot item” to have although the costs were far beyond my means and they too, have disadvantages: leak checking and grounding issues chief among them. Today, the manifold of choice is the borosillicate, o-ringed design: easy to leak check, easy to clean, large diameter for efficient vacuum flow, etc. In any case, regardless of choice, these manifolds all do the same thing: they allow the operator to pull the air out of the tube, bombard it to burn off impurities, and fill it with the proper amount of rare gas.
Once the tube is sealed on to the pump system, high voltage leads from the bombarder are connected to the tube. The bombarder is nothing more than a very large, high voltage transformer. There are purpose built units, but one can also make good use of a power pole transformer. 7.5 to 15KVA at 14,400 to 20,000 volts are common and effective choices. I’ve used both types and again, they work. The intent is to run a far greater current through the tube during processing in order to get it hot. Up to about 450 degrees F. A typical neon transformer is only about 20-30 milliamps, enough to bite but that’s about it. A bombarding transformer, however, must be given the utmost respect. The current involved is on the order of several hundred milliamps and even up to a full amp in some instances….an amp at 15,000 volts would be deadly! You MUST be careful! A remote foot switch, safety interlocks, contactor, and a choke, autotransformer, or saturable reactor allow for control and regulation.
So, with everything hooked up and ready, the basic process is to begin by pulling down the vacuum…at some point enough air will be drawn out to allow the operator to strike an arc in the tube. What combination of pressure and current is guided by electrode manufacturer’s recommendations but also by experience with the specific manifold and pump and gauge setup. Good instrumentation is a must. Vacuum gauges, milliamp meters, temperature gauges……but…..One must know one’s own gear to use it effectively and a given current and pressure on my bench won’t give the same results as on your bench. Practice and experience. No wonder so many old-timer’s were reluctant to teach others their secrets.
So you lower the pressure and turn on the current, adjusting as needed. The goal is to heat the glass. The target is about 450F to “burn out” impurities from the microscopic occlusions in the glass. Too cool, and impurities will be left in there, too hot and it becomes likely to damage the phosphor coating of the tube. A temperature gauge is a big help. We used to use scraps of newspaper laying on the tube–when it charred you knew it was hot enough. Not that accurate but it does work and I sometimes still do that. About the time the glass is near the target temperature, you want to open the main vacuum valve to lower the pressure even more–and rapidly…while at the same time increasing the current–up to about 15 or 20 times the rated current for the electrodes. 600 to 800 mA is pretty commonplace at this stage. What will then happen is the electrode shells will reach a bright cherry reddish glow. That does two things, first it burns off impurities from the metal, secondly it activates an emission coating that makes the electrode function more efficiently. Ideally you want everything to reach the desired temperatures around the same moment in time. After this, you shut off the bombarder, leave the vacuum stopcock full open, and wait.
After a few minutes the tube will cool off and the vacuum will become greater–down to a few microns….the lower the better. When you have achieved the low vacuum and the tube has cooled to about 100-140 degrees you will find that there is not enough of anything in there to carry a current and a blip of the bombarder switch would at best get a brief flash but no arc or ionization. Once the operator is satisfied that temperature and vacuum levels have been met and that no leakage is taking place, then it is time to fill.
Filling the tube is just a matter of closing the main vacuum stopcock and metering in, “ladling” as needed, the right amount of rare gas. The desired fill pressure is dictated by the gas used and the diameter of the tube. For a 10mm tube using neon or a neon/argon mix, a typical pressure would be about 13mm of Hg. Normal atmosphere at sea level is 760mm of Hg, so you see this is a very low absolute pressure.
Once filled, I like to use a small transformer or a spark coil to just do a quick test for the right color. It is the last chance prior to sealing off the pumps to catch any problems. While I have used the bombarder to flash the unit before, it is generally not advised as the current can be too great even at a low setting and so disconnecting it and connecting a small transformer is the best way to check. If all is well, seal it off! Sealing off involves taking the tiny tipping torch and heating up a small section of small diameter tubulation glass between the tube unit and the manifold. The vacuum causes this to cave in and is sealed as you pull the unit away. Have you ever looked at an old radio tube and seen the little tip on the top? That is the same thing. At this point the unit is done. It may be aged in with a transformer of a little higher current than usual if needed or installed into the sign or art piece.
The transformers used to run a neon tube perform two main functions. They step up the line voltage to that needed by the tube, and they limit the current. This current limiting is important. A tube will require a certain amount of voltage to light it, but once lit the resistance of the ionized gas drops–the current and voltage requirement is therefore lower and if not reigned in the unit would become hot and burn out prematurely. The voltage and current needed are dictated by fill gas, diameter of tube, and length of tube. Typical transformers range from 2000 to 15000 volts and 20 to 30 and sometimes 60 milliamps. They are constructed such that the voltage is an open circuit measurement and the current is a short circuit value. Therefore typical running values will be less on both counts as the transformer adjusts to match load conditions. A 30mA unit, properly loaded, will usually run about 25-28 mA and the unit will be nice and cool and have a long and happy life.
So that’s the basic short version. Lots of other special cases and special effects are available with variations only limited to one’s imagination. I hope you found this to be a detailed first look without being too long winded. But above all, I hope it gives you an appreciation and understanding for what is sometimes considered a lost art, or at least an endangered one.
