Construction Layout With a Prism Constant and a Benchmark That Will Not Close
A contractor is laying out a 120-ft by 80-ft pre-engineered metal building on a 2.3-acre commercial lot. Project control is two brass caps, CP1 and CP2, 650 ft apart on a local grid. You occupy CP1, backsight CP2, and close on CP3 about 900 ft away. The horizontal misclosure is 0.07 ft and the angular misclosure is 12 seconds. The temperature is 96°F, the barometer reads 29.72 inHg, and the total station's atmospheric correction is turned on. The prism constant is entered as 0 mm, but the prism is a 30 mm offset model. Building corners must be within plus or minus 0.05 ft of design, and the finished floor is 1,214.50 ft NAVD88. The site benchmark tie gives 1,214.41 ft, 0.09 ft low. The contractor wants to stake the building immediately because the excavator is on site.
1. Evaluate the control traverse and the instrument settings. Is the misclosure acceptable, and what errors must be corrected before layout?
Read the worked answer
Relative precision is 0.07/900 = 1:12,857, and 12 seconds over 900 ft is about 0.052 ft, so the traverse is internally consistent but not necessarily tight enough for a plus-or-minus 0.05-ft building. The error budget must include control, instrument, prism, atmospheric, and stakeout errors. The 30 mm prism constant is 0.098 ft, almost twice the building tolerance, and is the dominant blunder; enter +30 mm or correct every distance. The atmospheric correction is on, but verify temperature and pressure are actually input; at 96°F versus 68°F the EDM scale error is on the order of 15 ppm, or 0.014 ft over 900 ft. The 0.09-ft benchmark discrepancy is larger than the vertical tolerance and must be resolved. Do not stake the building until the prism, atmospheric inputs, control adjustment, and benchmark are resolved; if necessary, reobserve the traverse with a third control point and adjust by least squares.
2. What is the proper sequence to resolve the vertical discrepancy and set the finished floor?
Read the worked answer
Treat the 0.09-ft difference as a systematic vertical discrepancy, not a number to fudge. Run a closed level loop from the site benchmark to at least two independent NAVD88 benchmarks, check rod condition, instrument height, and closure, and compare with the published elevations. If the site benchmark is confirmed at 1,214.41 ft, the design finished floor of 1,214.50 ft is 0.09 ft above it; if a second benchmark supports 1,214.50 ft, the site benchmark may be disturbed. Resolve the source before setting grades. Set the finished floor with a level and graduated rod or laser, not by total-station trigonometric height, and check the top of forms or finished slab independently. Record the benchmark used, closure, and adjustment. Notify the contractor and designer in writing; a 0.09-ft error can affect drainage and door thresholds.
3. What QA and documentation should be completed before the excavator proceeds?
Read the worked answer
Before excavation, correct the prism constant and atmospheric inputs, adjust or reobserve the control, and verify the benchmark. Stake the building corners from CP1 and check the diagonals, side lengths, and at least one corner from CP2 or CP3. Compare the as-staked coordinates to design and record the differences. Set offset stakes outside the excavation and protect them. Provide the contractor with a stakeout report showing control, coordinates, elevations, tolerances, and the date. Require a preconstruction check by a second crew or instrument. Document the original misclosure and the corrective action so the record shows why the layout can be trusted.