Electronic Media Review, Volume Six: 2019-2020
A 1950s neon sign was conserved and partially restored at the McCord Museum. The neon sign came from the storefront of the Notman & Son photography studio, whose founder William Notman was a major 19th century Scottish-Canadian photographer working in Montreal. The goal of the treatment was to stabilize the heavily damaged sign and restore it to working order so that it could function for an exhibition. First, the metal base of the sign was conserved through cleaning, paint stabilization, and inpainting. Next, the broken and mostly missing neon tubing were recreated. Neon lights come in many different colors depending on the noble gas inside the glass tubes and whether any fluorescent coatings are used. Our restoration was initially hindered by a lack of documentary evidence regarding the original color and form of the neon tubes. Fortunately, the museum had a historic photograph of the sign that was used for reference. A local neon artist examined and tested original fragments of the glass tubing to determine their color. With this information, the neon artist could recreate the tubes and install a new transformer, allowing the sign to work again for exhibition.
The neon sign “Notman & Son” was used on the storefront of a photography studio in Montreal and dates from the mid-20th century. The founder of the studio, William Notman (1826–1891), was a significant 19th century Canadian photographer. Notman was born in Scotland in 1826 and, after joining the family textile business and committing financial fraud, fled to the New World in 1856. He quickly established his own photography practice in Montreal and became successful. His studio made portraits of Montrealers, important national and international figures, as well as landscapes and infrastructure projects. He promoted photography as an art form and pioneered innovative techniques such as painted photographs, composite photographs, and studio tricks to create outdoor “winter” scenes in his studio (Samson and Sauvage 2016).
After Notman’s death in 1891, the studio was passed down to his sons. It was sold to The Associated Screen News production company in the 1930s and then to a pair of photographers, Dudkoff and Montclair, who ran it from 1955 until its closure in 1993. The neon sign dates from this last era of the studio, used on a downtown Montreal storefront from the 1950s to 1970s. Meanwhile, a consortium recognized the importance of the Notman archive, which bought and transferred all of his original negatives and prints to the McCord Museum to preserve them (Samson and Sauvage 2016). More material, including the neon sign, was donated when the last proprietors of the studio retired. Today, the McCord Museum has nearly half a million items in the Notman Photographic Archives, including prints, negatives, record and account books, and studio equipment, forming an invaluable historic resource—essentially a century-long photographic record of Montreal.
The neon sign was included in the major retrospective exhibition Notman: A Visionary Photographer at the McCord Museum in 2016 to 2017. We decided not only to perform a conservation treatment on the sign but also to restore it to working order—meaning having it light up again, though inside the museum and not outdoors. This was rather unusual for our museum, as most treatments stay within the realm of conservation and rarely restoration, even for utilitarian or electronic objects. However, we wanted a bold, impressive exhibition, and a working neon sign would certainly be impactful. Restoration would respect the original purpose of the object and make it more understandable to visitors. Another reason was its social value: the sign was a place marker for the Notman studio, an important Montreal institution that has great importance to our museum collection, which made an involved treatment worthwhile.
There was also precedent for such a project, as there was existing interest in historic signage in Montreal from the Montreal Sign Project, led by Dr. Matt Soar, Professor of Communication Studies at Concordia University. Neon and other commercial signage can become landmarks and play important roles in social and urban histories. The project salvages old signs and displays them inside university campus buildings, with the goal of understanding local history and material culture (Montréal Signs Project 2019). Dr. Soar put us in contact with a neon artist who had experience making historic recreations for this project.
The neon sign definitely needed treatment. It had been neglected in a basement for decades before donation to the McCord in 2012. There was virtually no historic or technical documentation about the sign and no previous preservation work done.
The object is a channel letter type neon sign (fig. 1). Metal walled letter channels contain the neon tubing, which spells out “Notman & Son” in script-style lettering on a large ovoid steel base. The exterior of the letter channels and the base are painted a glossy black; inside, the letter channels are white to reflect the neon light. The paint was not analyzed but is likely an industrial paint or lacquer meant to waterproof the metal from outdoor water exposure. The glass neon tubes sit exposed inside the open letter channels and plug into sockets along the base, connected with interior wires in a single continuous circuit to a transformer inside. There are also sheets of mica insulation inside the base behind the sockets. The sign dimensions are roughly 200 cm wide × 45 cm high × 30 cm deep.
The sign was in poor condition. While overall the metal structure was stable, there were areas of localized corrosion and the paint film was heavily damaged, making it look neglected. There appeared to be flood and possibly fire damage. The sign was extremely dirty, covered with dust, debris, mud, and soot. The paint film had major losses and was peeling away in large, thick flakes, and multiple paint layers were apparent. Some areas of paint were blistering due to corrosion of the metal underneath. Other areas were brittle and blackened, as if burned. The metal was bent and misshapen along the bottom section of the “S” in “Son.” Most of the neon tubing was missing—only two short broken fragments remained (fig. 2), which appeared to be sections of the “N” and “m” in “Notman.” The wiring and transformer inside the base were corroded and non-functional.
The treatment was broken down into two parts. The treatment of the painted metal base was performed first to conserve and stabilize as much original material as possible. This was completed by conservators in the lab following standard conservation protocols. The restoration work was done next—the recreation of the missing glass tubes and replacement of defunct electrical components so that the sign could work again, and safely. Since this work required specialized equipment, it was completed by a professional neon artist in the artist’s studio.
The goal of the treatment was to stabilize and improve the appearance of the sign, favoring preservation and conservation of original materials and undertaking restoration or replacement only where necessary. We had no information on the manufacture of the sign, but it was considered a commercial neon sign and not a piece of contemporary neon art (for which the artist might be available for consultation). Since the original neon tubing was mostly missing and the electronics were damaged and unsafe to operate, recreation and replacement was required to have the sign work again safely. The neon tubes are exposed on this sign; thus, replacement with another technology such as LEDs was not appropriate. We justified the project ethically because restorations such as the repair of broken tubes and replacement of spent electrodes and transformers (which have a limited life-span) are normal parts of commercial neon sign maintenance (Auer 1991). Although Cagasan (1991) recommends repainting old sign casings for a more even appearance and to protect the metal from corrosion, we opted to preserve the remaining original paint and inpaint losses, as our sign would be kept inside and under museum conditions. Overall, our restoration would respect the original materials, function, and appearance of the sign. Neon technology has not changed much over time; thus, the new restoration components would essentially look and function the same as the original ones. The replacement tubes and electronics would also be removable and replaceable into the future; therefore, the restoration portion of the project still fits within a conservation ethics framework.
We cleaned the sign overall, then consolidated and inpainted the paint film to make the sign look as complete as possible. First, as an overall cleaning measure, the sign was vacuumed to remove dust and debris with various brush nozzle attachments. The painted outer surface of the sign was then cleaned. Tests showed that Orvus WA paste in deionized water followed by a rinse of 1:1 deionized water:ethanol was the most effective wet cleaning method. This was carried out overall using cotton swabs. The surface was immediately buffed dry with microfiber cloths to minimize moisture contact with the metal elements and prevent any streaks. Corrosion on areas of exposed metal was mechanically removed with stiff hair brushes or steel wool lubricated with mineral spirits for heavily corroded areas. Dented and misshapen sections of metal were mechanically reshaped with various hand tools, such as pliers wrapped with Microfoam. Minor paint losses occurred during these processes but was accepted, as this paint was damaged (i.e., burned or blistered) and not salvageable.
The larger thick, peeling flakes of paint were consolidated by applying warm BEVA 371, heated in a double boiler until readily flowing, under the flakes. After the BEVA dried, the paint flakes were softened and pressed down with a heated spatula to re-adhere them to the metal. Excess BEVA was cleared from the surface with organic solvents. Next, a barrier layer of 10% Paraloid B-48N w/v in toluene was applied to areas of bare metal with sable hair paintbrushes. This resin was selected for its good adherence and coating properties for bare metal (CAMEO 2018), and the concentration and solvent gave good working properties for even application without pooling or brushstrokes on horizontal and vertical surfaces. The resin served as an isolation layer for inpainting with water-based acrylic paints for our comfort and safety and also worked to further consolidate the edges of paint losses, especially thin areas where BEVA did not have good penetration. All work involving solvents was done while wearing personal protection equipment (nitrile gloves, lab coats, and sometimes respirators with organic solvent filters) and with fume extraction.
For inpainting, we applied acrylic paint over the isolation layer to fill in paint losses for basic visual reintegration. This would make the intended design appear complete and could be accomplished in a limited time frame. The exterior black paint film was inpainted with Liquitex Acrylic Gloss Medium with Kama Aqua Dispersion Pigments, which had good working properties and gave an even surface with no visible brushstrokes. The opaque glossy black paint matched the original paint film closely; thus, the inpainting was very successful. The inside of the letter channels were more challenging because there were several layers of white and off-white paint present, possibly from past repainting campaigns. Here, losses were inpainted with a color that matched the largest area of adjacent original paint, usually a white or titanium buff. Liquitex acrylic heavy body paints were used directly for this as they gave good color and semi-matte surface finish matches. While the interior of the letter channels still looked uneven after inpainting due to the multiple paint layers, these areas would be partially obscured by the neon tubes when installed and would not be visible when the sign is in operation and the glowing light obscures the details.
This treatment was completed in five weeks by the author and three colleagues—two conservators and one conservation student intern. One treatment protocol was established and carried out by all. The sign was large enough that two or three people could work on it at the same time without interfering with each other. Overall, the results were good: the stability of the sign and its appearance were greatly improved, bringing the condition from poor to good. Figures 3a, 3b, 4a, and 4b show before and after treatment details. With the base stabilized, the restoration part could begin.
About Neon Lighting
The modern neon light was invented in the early 1900s by George Claude, who modified a basic gas discharge lamp into something more practical and reliable. Colorful, bright, efficient, and long-lasting neon became popular for advertising and outdoor signage, uses we still see today, though with less popularity than its mid-20th century heyday. Neon lights—also referred to as neon lamps or luminous tubes—are sealed glass vacuum tubes filled with low-pressure noble gases (neon, argon, etc.) with electrodes at the ends. When a high-voltage electric current is applied, the gas atoms ionize and the charged plasma stream of charged ions connects the current between the electrodes; at the same time, positively and negatively charged particles collide with each other and become excited, emitting light (photons) as they return to ground state energy.
Evidence for Restoration
In theory, new glass tubes could be fabricated and plugged into the neon sign and it would work just as it originally did, since the technology remains the same. We had a problem preventing us from proceeding, however: we needed evidence regarding the original appearance of the sign to ensure that we were making a faithful reproduction. This meant the form of the missing glass tubes and the color of light that they produced, since many colors are possible. We did not have any schematics or documentation for the object, only the sign itself.
Regarding the form of the tubes: we only had two short original fragments, about 12 in. and 6 in. long, not enough to recreate the entire sign. These fragments did tell us the diameter of the tubes, at 0.5 in. The metal base has letter channels that outline the general shape the glass tubes must make, with sockets where individual tubes must start and stop and glass support stands. However, because the letters are in cursive script, the path between the sockets is not always obvious—the glass tubes must double back in several areas to make it to the next letter or socket (e.g., at the top of the “N” and “T”). The form is actually quite complex and three-dimensional.
Luckily, we did have evidence as to what the sign originally looked like to support our restoration. We had a historical photograph of the intact, working sign—not from the creator of the sign or the donor of the object but from elsewhere in our museum collection. The photo was taken circa 1979 by Edith Mather (1925–2016), an amateur Montreal street photographer whose archive was recently acquired by the McCord. The photograph (fig. 5) shows the sign on the studio storefront. Curators took special notice because of the connection to the Notman studio. We were able to find the location of the image, 1176 Sherbrooke Street West, Montreal, by cross-referencing archival business directory listings for the studio and going to the addresses, which still exist, to see if they matched the photograph. The neon sign is captured in the photo and gives just enough detail to make out the path of the glass tubes, giving the neon artist a plan to work from. Mather’s photograph also demonstrates that the Notman studio and sign were notable enough to document. While the photograph was useful, it was a silver salt black-and-white print; thus, we still did not know the color of the light.
Neon lights come in many possible colors, depending on the noble gas inside the glass tubes and the glass itself (fig. 6). Each gas produces a characteristic color. For example, neon gives a red color, and argon gives blue-violet. These are the two most commonly used gases for commercial neon signage because they produce bright light and are efficient to operate (Cagasan 2012). If the glass is clear, the flickering gas plasma stream and color is seen. The glass may be tinted or painted to boost or alter the color or, to produce an even greater range of colors, fluorescent coatings called phosphors are used. Phosphors are metallic compounds, typically rare earth or transition metals, applied in organic binders such as nitrocellulose to the inside of the glass tubes. The phosphors are excited by the ionized gas and produce their own color through fluorescence. A drop of mercury is added to tubes containing argon gas, which vaporizes and produces a large amount of UV radiation that excites the phosphor to produce a brighter light. Phosphors can produce light that is colored green, yellow, blue, purple, and white (as in overhead fluorescent lighting). Phosphors can also be used with neon gas for a color mixing effect: for example, the yellow phosphor with red neon gas gives an orange color.
Our glass fragments were broken; thus, the gas had obviously escaped but there was a phosphor coating present on the inside of the tubes. Unfortunately, the coatings appear opaque white when not in operation; therefore, we could not tell what color it would produce. Instrumental analysis was not available to test the phosphor coating.
At this point in the process, we got help from Montreal neon artist Gérald Collard, of Neon Family studio. Like many neon artists, he is passionate about the history and artistry of neon (which he calls “poetry in the night”) and was happy to work with us to make a reproduction. In fact, Collard had already made reproduction neon tubes for another iconic Montreal sign, the Farine Five Roses sign, a city landmark.
Collard examined the glass tube fragments. Since neon or argon plus mercury were the most common gases used for commercial mid-20th century signage, these were the two options for the gas. He did not see any characteristic signs of mercury—no liquid droplets, cloudiness on the glass, or black residue around the electrodes. He thus concluded that the tubes would have contained neon gas instead (Collard, pers. comm.). This was a subjective determination, but we trusted his deductions and experience.
Collard then suggested that we electrify the phosphor coating to see it fluoresce and determine its color. One glass fragment still had an electrode at one end, and the other end was broken. With our permission, he trimmed the broken end and fused a new electrode onto it. This intervention meant removing a couple of centimeters of original glass but was considered appropriate since it would give us the information we needed to proceed. The tube was now complete, sealed with electrodes at the ends so that it could conduct a current. The tube was hooked up to the manifold and voltage was applied (fig. 7), and the excited phosphor glowed green! However, this was without any neon gas inside the tube; the green phosphor coating plus red neon gas would mix and ultimately give an orange-red color. The museum had also posted a question about the sign on social media to see if any member of the public remembered seeing it. Responses did say red or orange; thus, we felt confident in our determination.
Recreation of the Neon Tubing
Now, knowing the form and color of the neon tubes (the combination of gas and phosphor), new ones could finally be recreated. The new tubes had to fit perfectly into the metal base and required specialized equipment. The now stabilized sign base was transported to the neon studio for this step.
The following procedure describes the neon artist’s process for our and any neon sign. First, the design is traced out on a template for the glass tubing to be checked against. Lengths of glass come already made from manufacturers in various diameters in 4 ft. lengths, either clear or with phosphor coatings. In our case, 0.5 in. diameter glass with a green phosphor coating was used. Historically, this green phosphor was zinc silicate, though various compounds are used today (Elenbaas 1971).
The neon artist’s main job is to shape the glass by hand. The tubes are heated in flame burners to bend and fuse them to the required length and shape. The length of a single neon tube is limited by the power of the vacuum to evacuate air and the pressure of the introduced gas, but large designs can be achieved by using several separate tubes and connecting them in a circuit. This also allows for multiple gases and colors to be used in one design. Opaque black enamel cross-over paint can be applied to block light from sections of tubing—for example, the spaces between letters—to create visual separation in a continuous length of tubing.
When the shaping of the tube is complete, cold cathode electrodes are fused to the ends. The tubes are hooked up to a manifold and a very high current is applied to vaporize any impurities (water, dirt, etc.) inside the glass, a process called bombardment. Then the tube is brought under a vacuum to remove the impurities and air and is filled with a small amount of the selected gas (neon) at low pressure. Once the tubes are sealed off, they are ready to install on the sign by plugging the electrodes into the sockets. The tubes are secured with copper wire ties to glass or plastic posts on the sign base. The electrodes are connected with copper wiring to a specialty neon transformer that delivers a constant high voltage with low current, between 1,000 and 15,000V at 18–30 mA depending on the length and volume of the tubes. The damaged original wiring and transformer were replaced with new equipment in our sign. Besides replacing a non-functional unit, modern neon transformers are advantageous because they are solid-state technology, lighter, cooler, more efficient, and safer to operate than older versions. Finally, the sign must pass an electrical inspection. It may be run for a period of time to reduce initial spluttering and stabilize the quality of light produced.
Collard made an interesting comment about our sign and neon transformer safety. The sign’s metal letter channels are very narrow and the glass tubes are close to the sides. This raises the risk of the current jumping from the tubes to the metal, causing electrical shock or even electrocution or fire (this may explain the burned paint). Modern neon transformers have a safety feature called a Transformer Overload Protection Switch (TOPS) that automatically shuts off when any fluctuation is detected in the current, preventing such shocks from occurring (Crowe 1991). We initially had trouble getting the sign to turn on, likely because of this feature, but once the tubes were properly stabilized with copper wire ties, it worked.
The glass tubes and sign base were transported back to the museum separately to avoid breakage. The base was first mounted on the wall of the exhibition space; then, the tubes were installed and secured with copper wire ties. The sign did work and light up for the 6 month-long exhibition. The orange-red light looked very dramatic in the gallery space (fig. 8), which was painted black for this exhibition. However, the neon light raised an issue for an exhibition space: we did not want excess light shining on other objects and risking fading. Fortunately, this was taken into account during the design process of the exhibition. The sign was mounted on its own wall at the beginning of the exhibition, away from any light-sensitive artifacts to ensure that they were kept safe. This also ensured that the sign had a dedicated electrical outlet, avoiding the need for long power cords or complicated installation.
Since we knew that the sign would be isolated, we did not measure the lux output or temperature of the sign while it was in use. During operation the neon tubes are not hot, but the sockets and transformer do get warm, with transformer specifications listing operating temperatures of 30°C to 40°C. We did not notice any heat damage around the transformer before treatment or after exhibition (fig. 9) or any effect on the consolidation or inpainting treatment. This is 20°C below the glass transition temperature of BEVA, Paraloid B-48, and acrylics, at Tg 50°C to 60oC (CAMEO 2016, 2018); thus, the elevated temperature was not problematic.
Overall, this project was a success. We conserved and stabilized as much original material as possible while also restoring the functionality of the sign with new additions that respect the object’s design and technology. The sign lit up and worked well in the exhibition gallery space thanks to a sympathetic design. After exhibition, the sign returned to storage but remains functional for future display, study, and use. The new tubes and electronics still have decades of service life remaining. In addition to seeing a functional object work again, this project was rewarding in getting us out of the conservation lab and collaborating with new people, along with making connections with other objects in our collection.
Finally, a postscript: our Notman exhibition travelled to another city last year. The museum determined that the original neon sign, even after stabilization, was too fragile to travel. Therefore, a replica was made for the travelling exhibition – not just another recreation of the glass tubing but rather a replica of the entire sign, metal base included. This replica was made by the neon sign artist and metal fabricator Scott Adamson of Gaslight Electric Sign Co. as an exact copy of the original (now conserved and restored) sign. The replica actually looks much better than the original since all of the materials are new and undamaged. The replica was made as an exhibition piece but hopefully it will be acquired to add to the interesting history of the object and the Notman collection at the McCord Museum.
Thank you to the McCord Museum, Anne Mackay, Sara Serban, Elisa Contreras Cigales, Matt Soar, Gérald Collard, and Scott Adamson for their help on this project. Additional thanks go to the Canadian Association for Conservation of Cultural Property for a CAC Professional Development Bursary to attend the 2019 AIC Annual Meeting and present this paper.
Auer, M. J. 1991. The preservation of historic signs. National Parks Service: Preservation Briefs 25.
Cagasan, J. J. E. 2012. A methodology for preserving Las Vegas neon electric displays. M.S. diss., Columbia University, New York.
CAMEO (Conservation and Art Materials Encyclopedia Online). 2016. BEVA ®371. Museum of Fine Arts, Boston. http://cameo.mfa.org/wiki/BEVA®_371 (accessed 07/02/2019).
CAMEO (Conservation and Art Materials Encyclopedia Online). 2018. Paraloid B-48N. Museum of Fine Arts, Boston. http://cameo.mfa.org/wiki/Paraloid_B-48N (accessed 07/02/2019).
Collard, G. 2016. Personal communication. Neon Family Studio, Montreal, Canada.
Crowe, M. F. 1991. Neon signs: their origin, use, and maintenance. APT Bulletin: The Journal of Preservation Technology 23 (2): 30–37.
Elenbaas, W., ed. 1971. Fluorescent Lamps. Philips Technical Library. London: Macmillan.
Montréal Signs Project. 2019. http://www.montrealsignsproject.ca/ (accessed 07/02/2019).
Samson, H. and S. Sauvage, eds. 2016. Notman: A Visionary Photographer. Paris and Montreal: éditions Hazan/McCord Museum.
Strattman, W., ed. 2003. Neon Techniques: Handbook of Neon Sign and Cold-Cathode Lighting. Cincinnati: ST Media Group International Inc.
Ribbat, C. 2013. Flickering Light: A History of Neon. London: Reaktion Books Ltd.
SOURCES OF MATERIALS
BEVA 371 Solution
330 Morgan Ave.
Brooklyn, NY 11211
Kama Aqua Dispersion Pigments
7442 St Hubert St.
Montreal, QC H2R 2N3 Canada
Fisher Scientific Company
112 Colonnade Rd.
Ottawa, Ontario K2E 7L6, Canada
Orvus WA Paste
330 Morgan Ave.
Brooklyn, NY 11211
Paraloid B-48 Resin
330 Morgan Ave.
Brooklyn, NY 11211
McCord Museum, Montreal, Quebec, Canada