The following article by Dr. George A. Borun was the subject of a presentation at the 1992 SCAVM Violin Making Symposium and originally appeared in the June, 1992, issue of the SCAVM Bulletin.
George Borun is a chemical engineer living in La Jolla, California, and is now retired as founder of the firm which he headed. He is a regular contributor to our Bulletin and was president of the Association in 1990, 1991, and 1998.
This article was prepared in 1992 and some of the sources and prices may have changed.
Silicone heater blankets were developed by NASA research to keep the insulated walls of satellites warm so that the enclosed instruments could function. They are now used in many applications like heat forming of flat sheet plastics and heat tracing of pipes. In the Spring issue #25 of the American Lutherie, Mike Keller describes how he uses the heating blankets for heating and bending guitar sides. He also mentions an earlier article in Fine Woodworking for bending wood with the blanket. The potential for use of these devices in violin making became obvious to me. I investigated alternate sources of the blankets, their comparative cost, and with some experimentation and adaptation, developed specific applications for violin making. I shall demonstrate today the process of bending violin ribs, the bout, lining, purfling, and how to remove the most stubborn glued violin neck, quite easily. The techniques are applicable to viola, cello, and bass making, as well as viola de gamba, dulcimers and other stringed instruments. I shall also show you how you can use the silicone blanket for heating bows prior to bending or re-bending. I shall also suggest how the process could be furher adapted to other luthier operations.
The silicone heater blankets are available from two sources, with a number of standard sizes in stock at each company:
The WATLOW ELECTRIC MANUFACTURING COMPANY, 12001 Lackland Road, St. Louis, MO 63146, (314) 878-4600.
Watlow markets through distributors. It has a large selection of stock size heaters, up to 35" long. The prices are available only from the distributors.
BENCHMARK THERMAL, 13185 Nevada City Ave., Grass Valley, CA 95945, (530) 477-5011, FAX (530) 477-6507
Benchmark has a smaller stock size selection, but their prices are considerably cheaper. They will make any size you wish, but that adds to the cost and delivery time. The stock maximum length is 20", but this is more than long enough for violins. In bending bass ribs, I use two 10" wide blankets for the longer pieces, and only one for the center bout. The best units are 4 Watts per square inch, rather than the 2 Watt model. This type will get up to 500 degrees F in a couple of minutes. Product catalogs are available which list the various sizes. Both of these companies also sell temperature controllers. They are, however, expensive. You can use a light dimmer on the smaller units to reduce, (but not control) the heat. There are other devices like rheostats that can be used provided they are rated high enough. You could use a simple snap disc thermostatic switch from W.W. Grainger, # 4E120, for $13.39, but it cycles on at 210 degrees, and off at 250 degrees F. This may be good for some low temperature applications. The snap disc unit will have to be attached to the metal liner at some mid point in order to detect the average temperature. The snap disc unit is rated at 10 Amps. My experimentation was done without the thermostatic controller. I used the blanket directly powered through a 14/3 grounded power cord plugged into a ground fault outlet. I also have two "Powerstat" variable transformers rated at 7.5 and 10.0 Amps, so either is good to use to lower, (but not regulate) the temperature.
For violins and violas, I use: BF-2479, 2" x 18", 144 Watts, 1.25A, 115 VAC. About $19.00. For cellos, I use BF-2481, 5" x 20", 400 Watts, 3.47A, 115 VAC. About $39.00. For basses, I use BF-2480, 10" x 12", 480 Watts, 4.17A, 115 VAC, about $46.00, used in the bout area, and combined with the BF-2307, 10" x 18", 720 Watts, 6.76A, 115 VAC, to bend the bottom of the corpus. About $65.00. All of the numbers are for the Benchmark units. Other stock sizes are available. The blankets come with 12" leads, and are 0.055" thick. The heating grid is traced through the entire blanket, and gets very hot almost instantly. You MUST use a ground fault outlet in your power outlet. This is cheap insurance to prevent electrical shock. It costs about $15.00 at any hardware store, or buy from W.W.Grainger stores. # 3A391, 20 Amp, $14.60. SAFETY FIRST!
You will need your standard inside mold, or make a special mold of plywood for each model. The advantage is that you can switch from one style and size to another using the same blanket. I used either my existing corpus mold, or retrofitted an old unused inside mold which I sanded down in size to accommodate the 0.02" metal, plus 0.055" heating blanket, plus the thickness of the ribs. The bending surface has a thin sheet metal cover which distributes the heat evenly. My demonstration equipment uses both aluminum and brass on the violin mold, and stainless steel on the bass mold. The thickness should be 0.02" or thinner. Stainless steel is harder to bend, but will not stain, and lasts forever. I should mention that you can glue the silicone blanket down to the metal heat sink on the mold, or the mold itself, using the cement available from both suppliers. It is the RTV 732 from Dow Corning, and is 100% silicone rubber, good up to 450 degrees F. Be aware that, once the RTV cures, there is no way that you will be able to remove it short of cutting it off. We use it in manufacturing to glue in gaskets and bind parts of dissimilar materials. It really holds! It may be practical to glue the heating tape to the stainless steel and use that as the "mobile" heating assembly to different mold forms.
The mold is prepared by covering the curved surface with the metal strip. The blanket then goes against the metal surface. The ribs are then bent over the blanket, and clamped down using a form fitting block. You could change the sequence and bend by placing the wood over the metal mold with the heating blanket next. There does not seem to be much difference in the sequence, as the results come out the same. You can use the same mold as your corpus mold, but you will note that the bent ribs will be slightly greater in curvature on the upper and lower bout, and slightly less on the "C" bout. It will still be better than the results of the old bending iron. To be exact, you can compensate for the thickness of the metal surface and the wood thickness, and if the blanket is inside, add the thickness of the blanket. My suggestion is to have the blanket on the inside, so the clamping frame does not stretch the heating element and separate it. The wood is prepared as follows: Cut the ribs as you would normally do, sanded to the proper thickness. From 1.5 mm for violins to 2.5-3.0 for basses. The blanket for the violin is 2" wide, so you have room to operate. Soak the wood for 15-20 minutes in hot water, or in hot water with 1% by volume, of household ammonia added (only ammonia, no soaps or other additives). The ammonia will soften the ridged cellulose cells allowing them bend without fracturing. Use the ammoniated water in a well ventilated area. When the wood is heated, the water evaporates, and the ammonia comes off as a gas. The cellulose becomes rigid again in its new shape, all stress free. Place the wet ribs over the heating blanketed mold and turn the blanket on. The heat will start immediately and the ribs will start to steam and bend. DO NOT FORCE THE BENDING PROCESS. The ribs will bend quite well by themselves when they are hot enough, with only a little pressure to hold it against the blanket. Since the wood gets very hot, use a form mold to clamp the bent wood against the ribs. You will not be able to use your hands. On outside curves, you can weight the loose unclamped end with a heavy spring clamp, and the weight will gradually pull the wood in contact with the blanket. This actually goes very fast. I also use a wood block to guide the bending process on the upper and lower bouts. There is a chance that the wood will get very hot and start to bake if you take too long to do the bending. You can forestall this if you simply wet the wood if it gets too dry. In any case, you should be done in 10 minutes at most. When you have the ribs bent, turn the power off on the blanket and let the balance of the water evaporate out by the residual heat for 15-20 minutes and let the mold cool down. If you do not use the ribs right away and they open up, you can re-heat them up for a couple of minutes by just placing it back in the mold and turning the heat. Most of my corpus molds are made from multi-layered plywood, which is of course glued. If you heat the rib mold for a prolonged time, the glue softens and melts. This could cause the mold to de-laminate. It may be best to use solid wood for high production operations where parts are bent one after the other. There are many ways to clamp the ribs to the mold during the bending process. You will need a curved block clamp in the "C" bout that fits well. Also one in the sharp curve of the top and bottom bout. Most of these curved clamps you now use in your conventional gluing process will work. I like the rod, eye bolt with wing nut, and wood block, holding clamp that is easy to apply, especially on the larger instruments.
The beauty of the silicone blanket is that you can use it for many other processes in violin making. Among these are:
1) Removal of old fingerboards Apply the heating blanket to the fingerboard for 5-8 minutes. Insert a spatula at the rear joint, and the board should pop right off.
2) Heating the plates prior to applying heated hide glue. This will keep the glue liquid longer.
3) Bending the plates before gluing. Cellos, and particularly basses, have the plates bent at the upper bout. I make an angled ramp that duplicates the desired bend, plus 5 degrees overbend. This compensates for snapback when unclamped. Clamp the main plate edges down to the board with the bend line at the edge of the ramp. Now lay the heating blanket on the bending line. Turn on the power and heat the area well. Gradually push the plate down to meet the ramped incline. Clamp down the edges. Remove the blanket and allow to cool.
4) Bending and rebending bows. A major prpblem in bow making is the proess of bending the bow after you make it or rebending after the curve becomes fatigued from use. The 2" wide blanket when "clothes pin clamped" around the bow will apply even heat in the whole area to be bent. When hot enough, the bow can be set in a set curved form to cool while in the proper bend curvature. This eliminates the alcohol lamp and scorched wood with its uneven heat.
5) Bent plate forming. WIliam Fulton gave a presentation at the 1990 SCAVM symposium, and has published the technique of bending flat thin plates to form the curved surface of a violin plate. The plates are then carved in the conventional maner He used a tank of boiling water and a special clamp and wedges to spread the wood apart Instead of the tank of water, you could use a silicone blanket to apply heat over the whole area and on both sides at once. The wet ammoniated wood could also be sandwiched between two blankets to first soften before clamping in the edge form. Heat can be applied continually to each side while the wedges are inserted to spread the wood apart, and kept there until the bend is complete. It will also help dry the plates out much faster than air drying. The plate bending process could now be applied to the larger instruments like the cello, or even a bass, without having a huge boiling tank.
I have not applied the blanket to a varnished instrument, as I believe it would cook and lift the varnish. I suppose that if you used a heat controller, you could apply the heat very slowly to loosen a glue joint. I would not have the courage to try this on a valuable instrument.
The use of the silicone blanket in stringed instruments is limited only by the imagination. It is inexpensive to buy and use, and can be adapted to existing forms and clamps now in use by the maker. It should also be a great boon to the larger violin making companies, that produce a large number of instruments, as it is a process conducive to high production of bent parts. A significant advantage of the blankets is that they can be used for many different models and sizes of instruments by simply changing the underlying mold. The procedures outlined in this paper and talk will be easy to adapt to the processes now used by violin makers and repairers. I predict that it will be in standard usage in a very few years.
All Bulletin articles are copyrighted ©1998 by the Southern California Association of Violin Makers. Contact Bulletin editor John Gilson, at the address given on our home page, for permission to reproduce Bulletin material.
Return to home page of the Southern California Association of Violin Makers .