TDR方法(ASTM D6780)在碾壓土壤含水量與乾密度量測適用性之評估

Abstract

土壤填方之含水量與密度控制對輾壓土壤工程影響甚鉅，ASTM規範近幾年推出非放射性土壤含水量與密度量測之時域反射法（Time Domain Reflectometry, TDR）ASTM D6780。然而此規範之現地評估與可能潛在問題尚未有文獻探討，本研究之主要目的即是針對規範之試驗方式進行測試，了解其量測準確度並加以改善。 ASTM D6780現地含水量與乾密度量測法中提出兩種試驗方法：單步驟法與雙步驟法，本研究以模型試驗與現地試驗進行量測分析，並搭配著砂錐法與烘乾法檢驗比較結果。根據本研究成果顯示兩方法在含水量之量測準確度可達1%~2%，然而乾密度量測誤差超過預期。誤差分析結果顯示，在單步驟法上具有理論上之缺陷；雙步驟法主要誤差來源為現地感測器貫入擾動。針對貫入擾動之改善上，本研究提出短期與長期之兩種改善策略：短期上，在TDR其含水量量測可達需求之結果下，提出TDR法與砂錐法聯用可提升乾密度準確度至2%；而長期上以感測器之重新設計為主，共有三種方案：減少貫入鋼釘數、縮小貫入鋼釘尺寸以及改採貫入式感測器，根據測試結果，由施作性、耐久性以及穩定性進行評估，以改採4根式直徑6mm感測鋼釘者最佳，目前可將乾密度誤差提升至3%。最後並更具本研究經驗提出後續改進之研究方向。

The quality control of gravimetric water content and dry density are important to compacted soils. ASTM recently included the non-nuclear Time Domain Reflectometry(TDR) method (ASTM D6780) for field measurements of water content and dry density. But its performance has not been extensively evaluated in the field. The main objective of this study is to evaluate and improve the accuracy of ASTM D6780. ASTM D6780 includes two methods, the one step method and the two step method. Both methods were experimentally evaluated in a physical model (in laboratory) and in the field, and compared with dry sand cone method and oven dry water content. Results show that water content accuracy of both methods reaches 1%~ 2%, but in the dry density measurement is much less accurate than expected. The poor performance of ASTM D6780 one step method was attributed to the flaw in its theoretical derivation. The unsatisfactory performance of ASTM D6780 two step method is found mainly due to penetration disturbance in the field measurements using the MRP rods, resulting inconsistency in the calibration relation between TDR measurement in the mold and insitu TDR measurement using MRP. Two counter measures were proposed. In the short term, a procedure combining TDR and sand cone measurements was proposed to verified to have accuracy dry density measurement better than 2%. In the long term, several modifications to the ASTM D6780 two step method were proposed, including reducing MRP rods to 2, reducing diameter of MRP rods, and replacing MRP with a TDR cone penetrometer probe. All modifications improve the measurement accuracy of dry density to about 3%, with 2-rod MRP being less stable. The reduced 6mm-diameter 4-rod MRP probe was currently suggested because it is not only accurate, but also durable and easy to operate. Future improvements were also suggested based on the experiences of this study for future studies.