The ATLAS Scintillating Hadron Tile Calorimeter Project

Tags

The ATLAS Scintillating Hadron Tile Calorimeter Project

The ATLAS Scintillating Hadron Tile Calorimeter Project:

Submodule Production

Dr. Steven Errede, David Petersen

Department of Physics, University of Illinois at Urbana-Champaign

1110 West Green Street, Urbana IL 61801-3080, USA

The ATLAS detector is one of two new particle detectors being produced for the Large Hadron Collider (LHC) in construction at CERN, Switzerland. The University of Illinois is one of 144 institutions collaborating on the ATLAS project. Our main responsibilities are to produce one third of the submodules needed for the ATLAS Scintillating Hadron Tile Calorimeter and test around 3,000 of the photomultiplier tubes needed for the calorimeter. The purpose of this paper is to continue Jason Lee’s paper in which he rigorously described the procedures the University of Illinois is using to produce these submodules as well as the procedures utilized to ensure quality control. His paper, along with many other documents concerning the TileCal at the University of Illinois can be found at http://web.hep.uiuc.edu/atlas/docs.html or can be directly accessed from this link http://web.hep.uiuc.edu/atlas/docs/JasonATLAS.doc. It is suggested that one read his paper before continuing reading this one. This paper will also examine and organize many of the quality control measurements and put them into graphic documentation.

I. RUSTPROOF PAINT – PHASE ONE

Mr. Lee’s paper ends with the anticipation of rustproof paint arriving so that the submodules may be painted and shipped to Argonne National Laboratory, where they will be assembled into a full module.

The black paint arrived from the Czech Republic. In order to paint each submodule an elaborate procedure must be followed. After the final welding is complete, the submodule must be removed from the welding jig and placed on a table. While on the table, both ends of the submodule must be duct taped so that the paint does not stick to the ends. After removing the duct tape that covers the bolt holes, the end caps must be bolted on and the slots checked one more time. This includes checking all of the slots’ width using the four machined metal slot checkers. Also the height of each slot must be checked using a different piece of machined metal.

Before the submodule is to be painted, the paint also must be prepared for use. Whenever working with the paint, a gas mask should be worn. The writing on the paint cans is all in Czech, so the contents of the paint remain unknown. However, if one is to breathe in the paint fumes for periods longer than 45 minutes, he will notice that his lips and nose have begun to go numb. About 1 to 1.5 liters of deionized water must be added to each can of paint before it can be added to the tub of paint which is connected to the painting tank. Sometimes no water is required to be added to a can of paint. How much water is necessary depends on the viscosity of the paint in the tub. We test the viscosity using the viscosity cup and it should take 14 to 18 seconds for the paint to run through this cup. While still in the can a thick white film forms over the paint. After adding the water, the paint must be stirred thoroughly. This is completed with an oar and a motorized propeller. The paint is very thick and chunky at first, but loosens up after about 5 minutes of stirring. The can of paint can then be added to the tub of paint. Before adding the paint, a strainer must be placed over the tub of paint so that none of the big chunks of material in the paint enters the tub of paint. Usually two cans of paint are added to the tub of paint every 5 or 6 submodules.

The tub of paint itself must then be stirred, as seen to the left, again using the oar and the motorized propeller. Once this is done, the lid is placed back on the tub of paint, but not tightened. Then the filter that is used must be cleaned out with water before proceeding with the painting. The paint is now ready to be used. Now the submodule must placed in the painting tank.

The tub of paint is connected to a pump and filter that are connected to the paint tank via a series of valves and hoses. The submodule must be placed in the paint tank. After this done, a plastic cover is placed over the paint tank that is connected to nitrogen. The nitrogen is then pumped into the paint at a rate of 2 psi. The nitrogen is necessary because this particular paint coagulates very quickly when exposed to the air and dries too quickly and forms clumps in the paint tank if nitrogen is not used.

After the paint tank has been filling with nitrogen for a couple minutes, the painting begins. The filter must be attached and the pump turned on. After tank is filled with paint, the pump is to be shut off. After a couple minutes of letting the submodule sit in the full paint tank, one must skim off the white layer that has formed on the top of the paint in the paint tank. This can be done after turning off the nitrogen and lifting up the plastic lid over the paint tank. After skimming the paint off, replacing the plastic lid, and turning the nitrogen back on the paint can be drained from the paint tank. Reversing the flow of the paint, the tank must be drained at rate of 2.5 to 3 cm a minute. After the tank is drained, the filter must be disconnected and cleaned.


Before the paint is allowed to dry on the submodule, the source holes must be vacuumed out to remove any paint inside them. Again using the vacuum, excess paint must be removed from all four sides of the submodule. Only after going over all four sides with a roller and paintbrush may the submodule be raised to ceiling and allowed to dry for at least 30 minutes.

The paint takes at least 3 days to dry so when placing a freshly painted submodule on wooden blocks, pieces of metal bent to 90 degrees are placed on the wooden blocks so that the submodule rests on thinnest part of the metal. This is done to prevent the wooden blocks from sticking to the submodule. Only, after at least three days have passed may the final quality verification be done on the submodule.

II. FINAL QUALITY VERIFICATION

After the paint has dried, the final quality verification can be performed. The slots must be checked again using the process previously described. The ends of the submodule should also be cleaned of paint using a wire wheel, which can be seen being completed to the left. After this is done, the submodule’s top and bottom must be filed because the paint does not dry smoothly. After filing, the submodule is ready for the final coat of paint.

III. RUSTPROOF PAINT – PHASE TWO AND WRAPPING



In order to obtain a nice, even coat of paint on the submodule it must be spray painted. Using the same paint, one must fill up a spray paint container and connect this to a source of high air pressure. Again, a gas mask must be worn when working with the paint. After an even coat is applied to all four sides of the submodule it placed on the wooden blocks to dry for another day. After 16 submodules have been final painted, the submodules can be placed on wooden pallets and wrapped in plastic. They are then secured to the pallets using metal banding.

IV. QUALITY CONTROL – AFTER TACK WELD

Mr. Lee’s paper discusses only the beginning of the quality control measurements that are involved in the production of each submodule. There are three types of measurements that must be performed on each submodule, they are measuring the height and the perpendicularity of the submodules, and measuring the bolt holes on the wide end of the submodule.

Mr. Lee describes the measurement of the height of the submodule at fourteen different points using the Mitutoyo digital depth gauge after tack welding is complete and the load plate has been removed from the submodule. These points are labeled 1 – 14 on the quality control sheet (Appendix A). These heights are measured from the spacers on the top layer of the submodule to the stacking table, on which the submodule sits on. The heights are measured in millimeters.

While still on the stacking table, the perpendicularity of the submodule can be measured also using a cylindrical square and a Best-Test height gauge with dial indicator. The perpendicularity is measured at twelve different points around the submodule, these points are labeled A – L on the quality control sheet (Appendix B). Placing the cylindrical square flat against the side of the submodule at these various points, one moves the dial from the highest point on the cylinder of cylindrical square that is nearest the submodule out to the end of the cylinder, noting the change in the reading on dial. If the dial moves clockwise, then the perpendicularity is positive, if the dial moves counter-clockwise, then the perpendicularity is negative. Also, the number that the dial moves must be doubled in order to account for the short cylinder on the cylindrical square. The dial is the only measuring device used that is not calibrated in millimeters, but instead it is calibrated in inches.

After the perpendicularity is measured, the submodule is then placed in the welding jig and rolled into the welding room. Before final welding, the bolt holes on the wide end of the submodule must measured using a Mitutoyo digital vernier caliper. First precision bolts with a head diameter of 22.23 mm are placed into the bolt holes. The bolt hole separations between the weld bars must be measured first. Using the caliper and starting at the top two bolt holes (the top of the submodule is defined as the last layer put on, or the only side with spacers on top) the distance separating the bolt holes is measured and recorded on the quality control sheet (Appendix C). Then the middle bolt holes separation is measured followed by the bottom bolt holes separation.

Next the weld bars must be checked to make sure that they are centered in the up (top) and down (bottom) direction of the submodule. First the left weld bar is measured and then the right weld bar. This is done by leaving the precision bolts in and clamping a precision metal bar that is exactly 19.05 mm to the top of submodule, making sure to let some of the bar hang over the weld bar being measured. Starting with the top bolt hole and working down, the distance between the bolt hole and the bar is measured using the caliper. After all three distances on the left weld bar are measured and recorded, then the right weld bar can be measured. If these bolt hole separations are all within the allowed specifications, then the submodule is ready for final welding.

V. QUALITY CONTROL – AFTER FINAL WELD


After the submodule has been final welded and cooled for about an hour, the submodule may be removed from the welding jig and placed on a granite table for final measuring. Before placing the submodule on the granite table, the table and submodule must be cleaned with alcohol so the measurements taken will be accurate. The submodule must be cleaned in order to remove any grit or dirt that remains from the final welding and grinding that is done after the final weld. All the measurements taken after the tack weld must be taken again. However, one more set of measurements must be done while the submodule sits on the granite table.

The height of the submodule must now also be measured at six more points. These points are in the middle of the submodule and the heights are measured by using a Mitutoyo digital depth micrometer that is placed down through the source holes. To measure the first point, labeled point 15 on the quality control sheet, one must put the depth micrometer down through the first source hole that is closest to the wide end. The micrometer must rest on the spacer closest to the wide end. After this point is recorded, the heights of points 16-20 must be found by placing the depth micrometer down through both source holes of the top spacer. Then taking the average of the two measurements, one may obtain the height for the corresponding point on the quality control sheet. After the height for the twentieth point is measured, the submodule is ready to be prepped for painting.

VI. QUALITY CONTROL ANALYSIS - HEIGHT

The design height for each submodule is 291.7 mm. The submodule is allowed to be 0.3 mm over the design height, or 1.5 mm under the design height. The first type of quality control analysis done is to plot the average height for each individual submodule. The heights after the tack and final weld are both included on the same graph. Also included on this graph are the maximum and minimum heights for each individual submodule, again the heights measured both after tack weld and final welds are included on this graph. When analyzing the measurements, human error and any dirt or grit that may remain on the table or the submodule after cleaning may affect the accuracy of the measurement. When considering the measurements that result from the use of the Mitutoyo depth gauge or depth micrometer, it must be noted that one should allow +/-0.05 mm for error in each measurement resulting from the occasional piece of grit getting in the way.