Half an inch either way was sufficient. The hand wheel rotated along with the central pipe, the chucks, the tubing, and the covers for the ends of the tubing, and was rotated faster by hand to impart motion to the glass in one direction, and held counter to the rotation to reverse direction of' chuck travel. It was found that this method gave us very sensitive control over movement of the glass prior to sealing, even though done remotely through commands and hand signals to an assistant.
							On chucking the 18 inch tubes, about a 5 mm gap was left between the two cylinders to be sealed, and a 2 inch diameter pad of ceramic fiber cloth tubing, 2 mm thick cushioned the interior of the glass tubing from the chuck jaws.
				Figure 3
		The glass tubing varied, in outside diameter by as much as 1 cm, so centering the glass on the chucks was done by measuring up from the top surface of the central pipe to the outside edge of the glass tubing at four places 90 apart along the circumference. A compromise is affected between the sides of one tube and the low sides of the other by rotation of the tubing on the chucks before final tightening of chuck jaws. To mount the sections of glass tubing on the chucks we lifted and held the central 6-inch steel tube with the overhead crane, removed one "A" frame, slid the glass tubing over the open end, then replaced the "A" frame. The ends of the tubing were sealed to make the assembly airtight with galvanized metal covers held in place with heat resistant tape. Access to the interior for blowing was made through the shaft of the hand wheel and holes were drilled in the 6 inch steel pipe to admit pressure of blowing to inside of glass assembly. After sufficient preheating, the oxygen/hydrogen burners were wheeled into place, lit, and adjusted for maximum heat, At the point that the heated ends began to bead up and get noticeably heavier, the movable chuck was engaged, closing the gap between the ends of the glass tubing. Heating was continued until a smooth seal was accomplished. Although provision was made to blow into the center area, it was not necessary, as the speed of rotation (20 rpm) kept the glass from either spinning out or shrinking in. Sealing time was about 15 minutes per seal, Only in the later stages of sealing was it found necessary to paddle the seal with a graphite tool to maintain roundness, this due largely to uneven wall thickness in the glass cylinder.
									Figure 4
		Glassblowers were protected from excessive heat during sealing operations with aluminum coated full length shop coats, and welders face masks. The use or these safety devices allowed us to move at will as close as we wished to the heated zone.
								Page 1 | Page 2 | Page 3 | Page 4 | Page 5