Eight Plane IPND Mechanical Testing

摘要

A mechanical test of an 8 plane IPND mechanical prototype, which was constructed using extrusions from the testing/tryout of the 16 cell prototype extrusion die in Argonne National Laboratory, was conducted. There were 4 vertical and 4 horizontal planes in this 8 plane IPND prototype. Each vertical plane had four 16 cell extrusions, while each horizontal plane had six 16 cell extrusions. Each plane was glued together using the formulation of Devcon adhesive, Devcon 60. The vertical extrusions used in the vertical planes shares the same dimensions as the horizontal extrusions in the horizontal planes with the average web thickness of 2.1 mm and the average wall thickness of 3.1 mm. This mechanical prototype was constructed with end-seals on the both ends of the vertical extrusions. The gaps were filled with epoxy between extrusions and end-seals. The overall dimension of IPND is 154.8 by 103.1 by 21.7 inches with the weight of approximately 1200 kg, as shown in a figure. Two similar mechanical tests of 3 layer and 11 layer prototypes have been done in order to evaluate the strength of the adhesive joint between extrusions in the NOvA detector. The test showed that the IPND prototype was able to sustain under the loading of weight of itself and scintillator. Two FEA models were built to verify the measurement data from the test. The prediction from FEA slice model seems correlated reasonably well to the test result, even under a 'rough' estimated condition for the wall thickness (from an untuned die) and an unknown property of 'garage type' extrusion. A full size of FEA 3-D model also agrees very well with the test data from strain gage readings. It is worthy to point out that the stress distribution of the structure is predominantly determined by the internal pressure, while the buckling stability relies more on the loading weight from the extrusions themselves and scintillate. Results of conducted internal pressure tests, including 3- cell, 11-cell and the IPND prototypes, have been correlated to the FEA analysis very well. The authors believe they have quite good understanding of response of the NOvA structures subjected to the internal pressure, while the understanding of buckling stability is far behind. Therefore, more effect should be laid to improve the buckling considering that the FEA analysis usually is not able accurately modeling the stability as good as the stress analysis. The IPND structure was mostly built using 'scrape' piece extrusions (whatever available in shop). Therefore, a future test should be more focus on by using a actual real extrusions, for example like Nova -27 (if a final choice is made) and extrusion from a tuned die (very important). The authors should/will repeat 11 layers test with an actual thicker piece for the vertical to verify the adhesive joint and similar large scale prototype with a symmetry case, either 9 or 11 layers with the dial indicator on the both side.