September 2000 Volume 22 Number 3
or roughly translated: single thread tear repairby Llisa Merz-Le and Carolyn Tallent
The Tear Repair Seminar and Workshop took place September 7th to the 9th at the Art Institute of Chicago. It was organized by Rob Proctor, sponsored by the Paintings Specialty Group, and hosted by the Conservation Department at the Art Institute. Approximately 64 conservators drawn equally from private practice and institutions came together from all over the United States and Canada for the seminar. Fifteen participants took part in the hands-on workshop which followed. Both were conducted by Professor Winfried Heiber who teaches at Hochschule fur Bildende Kunst and his assistant Petra Demuth, a graduate of the program.
A number of conservators have been working with thread bonding as a technique for tear mending for some time with varying degrees of success. Perhaps the two questions that most worry those who want to join their numbers are: is the join strong enough to hold without distortion; and is it possible to do a high quality treatment in a reasonable amount of time. So, probably everyone at the seminar began the day feeling an understandable degree of curiosity mixed with a healthy dose of skepticism. By the end of the first day the curiosity was even more intense, and the skepticism had evaporated.
Professor Heiber began the seminar by expressing his own comittment to the preservation of a painting's authenticity by preserving "all the small things." His pursuit to find a better way to mend tears grew from years of seeing the undeniable loss that occurs to a painting that is lined or even removed from the stretcher for structural work and then reattached. We have all seen such losses, where the design layer along a tacking fold crumbles or becomes severely cracked in spite of all precautions. We have seen linings that seem to take the life away from the painting even though they have corrected severe structural problems. Heiber captured the sentiment we all have felt at one time or another when he stated, "A painting that is lined becomes an extremely different painting; it loses its venerability".
This introduction was followed by a series of slides that showed tears before and after reweaving treatments. The tears ranged from short and linear to long and complex, and the mends were near invisible from both the front and back. None of the paintings shown were lined.
Prof. Heiber then described the role of fabric structure. As part of his research he tested 30 linen fabrics to understand what influence the warp and weft yarns play in tear development. He found that because of tensions developed during weaving, warp and weft threads are different lengths. If you cut a perfect square and take out one warp and one weft yarn, you will find that the warp is longer than the weft. This means that when stress is applied to the fabric, the shorter weft threads exceed their capacity to give sooner and break first. Heiber found that more often than not tears occurred parallel to the warp. We have all seen how often tears have long intact floating threads with broken cross threads, and this is the explanation.
Since 1985 Prof. Heiber and his colleagues have tested many adhesives in their approach to the tear closure and repair processes. Heiber gave a very comprehensive description of the pros and cons of all the past procedures and used diagrams and tests to support his theories. The result of years of work and tests culminated in the development of a reversible two part adhesive system. The consolidant mixture Heiber developed and highly recommends is a 1 to 1 mixture of 10% wheat starch paste and 20% sturgeon glue. The glue provides the strength and the starch paste acts as a filler. If you have to thin the adhesive, you should dilute the starch to 5% before making the mixture, so that the glue concentration is not reduced. He also uses a 3 to 5% sturgeon glue solution for thread preparation, as described below. The adhesive is kept at working temperature in a glass vial on a small hot plate. A small hole in the vial cap allows access while reducing evaporation. While the components can be stored if kept in syringes, with all air expelled, the adhesive mixture should be made fresh daily.
The participants involved with the workshop were asked to test the adhesive strength by cutting a single linen yarn and mending it with a 1/2 millimeter lap join using the adhesive. Tensile strength tests were executed on everyone's mended yarns, and the ultimate tensile strength of the repaired areas ranged between 150 grams to 400 grams.
A listing of the instruments used for the tear repair procedure can be found in the technical exchange column, and they will only be mentioned briefly in the following paragraphs. For the seminar and workshop Prof. Heiber worked with a stereomicroscope with video output. Everyone at the seminar had the opportunity to see the tear mending procedure in fine detail.
Heiber described his work area as "creating your world" and he systematically organized his tools around the stereomicroscope. The entire repair procedure was executed at about 12-25 times magnification. The first step in repairing the canvas tear was to consolidate ground and paint layers, especially those flakes with minimum canvas support. This was done with sturgeon glue, but could be done using any system that would not interfere with the sturgeon glue and wheat starch paste used to infuse and repair the canvas yarns.
Prof. Heiber began by placing the canvas face down on a fairly hard surface. He recommended starting in an area that will mend fairly easily and working out from that point. He emphasized that studying the tear, determining the proper orientation of the threads, and planning the mending strategy was very important. He said that the reconstruction of a weave structure ideally recreates: the weave pattern; the thread tension; the independent movement of the threads; and the plane of the canvas. It should also be durable. However, it is sometimes necessary to remove a thread for a short space, or otherwise alter the weave structure when mending if not enough material is available.
Once he chose an area to begin mending, Prof. Heiber selected two dental probes with corkscrew-like ends to manipulate the tear. The first step was gently to move all the thread ends up, or towards the back of the painting, to facilitate orientation. To anchor his chosen area he placed a finishing nail on either side with the nail head 1/8 inch away from the tear edge. These were held in place by weights in the form of iron bars, beveled so as not to obstruct the work.Using the probes, he organized the threads, examined them for length and break location, and rewove them where possible.
The broken threads were almost always frayed. Before the actual mending occurred, each yarn was preconditioned with water and/or the dilute sturgeon glue and drawn out with probes or tweezers to elongate them. To dispense adhesive Heiber used an insect needle inserted into a pin vise backward, with the head pointing out. He used this tool to carry the very small amount of glue needed to infuse the broken weft yarn ends, and recommended against a brush, because it either deposits too much glue or becomes dried and clogged. A very small brush can be used for the water.
It is extremely important to use the smallest amount of adhesive and only to infuse those yarns being mended. If too much adhesive is administered, the yarns become bulky and the mend becomes messy and distorted. One of the many positive aspects of the adhesive mixture is that it is reversible with re-wetting and excess glue can be removed.
The glue was deposited, and very fine tweezers and dental probes were used to lightly manipulate the yarn into a point. The sturgeon glue saturated the yarn, relaxing the twist in the fiber allowing it to become longer. If the glue-infused yarn dries during this process, it was simply moistened. The heated needle was also used to relax and lengthen the thread ends.
The heated needle is a welding needle with a custom made diagonal tip. It is kept at 35 to 40° C by the control unit.
The preconditioned weft yarns were now ready to be attached. The warp yarn had been pulled away from the area being treated and held with cross locking tweezers. The weft yarns that lay under the warp yarn in the weave pattern were attached first.
Heiber used the insect needle to administer a small quantity of the wheat starch paste and sturgeon glue mixture to the two yarn ends to be joined. The yarns overlapped just slightly, and he used a dental tool to manipulate the ends together.After checking the yarns for proper alignment and cohesion, he used the heated welding needle, with little to no pressure, to heat set the two yarns together. Heiber used the heated needle to caress the yarns together starting at the top and then to either side. He warned that too much pressure would cause the yarns to fan out and becoming bulky.
After the lower yarns were attached, he placed the single warp yarn in its proper location and began joining the top yarns.Heiber said that it is important just to adhere the tips of the yarns and not to adhere the warp and weft yarns together, to assure flexibility and movability.
Prof. Heiber used controlled amounts of adhesive, the best tool for each step and amazingly steady hands to accomplish this task. The result was a virtually invisible mend, even under magnification. It is also important to note that Prof. Heiber emphasized he had used this technique for many years in private practice. One can do single thread tear repair and not go broke; it is both excellent conservation and cost effective.