Falling Weight Deflectometer Calibration Center and Operational Improvements: Redevelopment of The Calibration Protocol and Equipment

APPENDIX B. SUGGESTIONS FOR SUCCESSFUL ANNUAL CALIBRATIONS

The following suggestions are not a part of the formal protocol, but they have been found to be helpful in the success of the procedure.

1. The deflection sensor calibration procedure will work better if the deflections are 16 mil (400 μm) or more. It is not required that a special test pad be constructed, but if a regular concrete floor is used, locate the ball-joint base on a spot where the deflections are large enough.
   
   
2. Before doing any calibrations, verify that the FWD computer and the calibration computer are registering the correct date and time. Reconcile any issues before proceeding.
   
   
3. Locate the calibration data acquisition system as close as possible to the FWD computer so that the two system operators will be able to converse easily.
   
   
4. The signal conditioner and load cell should be connected and warmed up for at least 30 min before calibration. This will reduce the variability of the data during calibration. The signal conditioner and accelerometer only need to be warmed up an additional 5 min if the signal conditioner is already warm. If deflection sensor calibration is done first, the load cell can be warmed up while relative calibration is being performed.
   
   
5. Prior to calibration, the FWD should be warmed up using the standard operating procedure for the particular brand of FWD. This will reduce the variability of the data during calibration. Around 25 drops are needed for this purpose. More may be necessary if the FWD is colder than room temperature.
   
   
6. At the beginning of each programmed series of drops, add two seating drops (no data recorded) for which the data are not recorded. Seating drops are not required, but they will reduce the variability of the data during calibration. Seating drops are particularly needed after the deflection sensors have been moved in the calibration stand.
   
   
7. For load cell reference calibration, position the FWD so that the load plate is near the center of the calibration test pad. It is very important that the FWD is level when the calibration is performed. By doing load cell calibration on a different area of the test pad than the deflection sensor calibration, the life of the test pad is increased. It is acceptable to perform load cell calibration on an area of concrete floor that is away from the test pad provided that the FWD is level.
   
   
8. Position the reference load cell beneath the FWD load plate, ensuring that the three guides are properly aligned around the plate. Do not loosen or remove the alignment fingers on the reference load cell. Zero the signal conditioner with the load plate high so that there is no external load on the reference load cell.

   Note: For accurate results, it is important that the reference load cell be zeroed with the FWD load plate in the raised position. Also, the signal conditioner excitation and gain must be set carefully to the levels at which the reference load cell was calibrated, as indicated in the WinFWDCal prerequisite information during setup.

9. Ensure that the FWD load plate is seated squarely on the reference load cell and on the concrete. To check this, perform a couple of seating drops and then try to slip a single sheet of copy paper under the load plate anywhere around the perimeter. It should not be possible to get the paper under the load plate or the three supports under the load cell any more than approximately 0.25 inches (6 mm). This will reduce the variability of the data during calibration.
   
   
10. For deflection sensor calibration, the mounting bolts for the ball-joint base and the bolts that clamp the stand to the base must be tight. Vibration will cause excessive variability of the readings, leading to an inability to pass the data acceptance criteria. The bolts holding the ball-joint in its socket should be tight enough to hold the stand upright but still allow the ball-joint to rotate freely.
    
    
11. The sensor stand connector pin (see figure 125) should be tightly attached to the stand. Apply Loctite^® to the threads to assure that the pin will not turn. This will reduce the variability of the data during calibration.
    
    
12. Verify that the deflection sensors are held firmly on the shelves of the sensor stands. This will reduce the variability of the data during calibration.
    
    
13. Attach the reference accelerometer box on the center shelf in the stand using two thumb screws. Press down on the box while tightening the screws to ensure that the box is seated firmly on the shelf. Try to keep the accelerometer aligned vertically in Earth’s gravity field at all times to avoid hysteresis in the accelerometer readings.
    
    
14. Place the geophones in the single-column calibration stand as shown in table 8. If only seven sensors are being calibrated, then the top two positions (A and B) should be empty in trial 1, and the bottom two positions (I and J) should be empty in trial 2. An on-screen figure in WinFWDCal, similar to table 8, will show how the sensors should be placed in the stand.

    Note: The goal is to have the sensors centered on the accelerometer and to invert them uniformly in the second trial. If additional calibration trials are needed, they should be performed in sets of two using the positioning shown in the table.

15. Place the KUAB seismometers in the double-column calibration stand as shown in table 9. If nine sensors are being calibrated, put them in the top two positions. For trial 2, rotate the stand 180 degrees. Do not invert the sensors in the stand. The wide axis of the stand should be perpendicular to the wave coming from the FWD load plate. An on-screen figure in WinFWDCal, similar to table 9 will show how the sensors should be placed in the stand.

    Note: For trial 1, stand behind the stand facing the FWD load plate with sensors D1, D3, etc., in the left column. For trial 2, rotate the stand vertically on the ball swivel so the column with sensors D1, D3, etc., is on the right.

16. The accelerometer calibration is slightly temperature sensitive. Temperature is monitored continuously by WinFWDCal; however, the measurement point is at the KUSB DAQ. It is important to keep the DAQ and the accelerometer box out of direct sunlight to avoid measurement errors or false warnings.
    
    
17. Use a gentle downward pressure on the handles of the calibration stand while the reference and relative calibration data are being collected. At least half of the bubble on the level should be inside the black circle on the sight glass while data are being collected. Holding the stand consistently for each drop will reduce the standard error in the calibration trial.
    
    
18. For load or deflection reference calibration, if either of the following conditions occurs, the calibration testing should be repeated after identifying the source of the following problems and correcting them:

• Excessive noise at load levels of 9,000 lbf (40 kN) or more. The noise, due to electrical noise or mechanical vibrations, is of concern only if it results in an erroneous baseline value or an erroneous peak reading. The time history graphs provided by WinFWDCal should be reviewed to determine if the noise is a concern before rejecting the data for a drop.
  
  
• Standard deviations at any load levels that differ by more than a factor of three between the reference system data set and the FWD dataset.
  
  
• Standard deviations less than 25 lbf (0.11 kN) for the load cell and 0.16 mil (4 m) for the deflection sensors are acceptable regardless of the ratio. WinFWDCal will check the ratios and post an error message, if applicable.

19. When not in use, the load cell, accelerometer, signal conditioner, and other calibration equipment should be stored in a protected location. The accelerometer should be stored in a +1 g gravity field to eliminate hysteresis. To accomplish this, attach the accelerometer box to the calibration platform using the two thumb screws. Use the bubble level to adjust the platform on the storage shelf.
    
    
20. In the event that the accelerometer is found to not be level after a period of storage, it will take at least 24 h at room temperature to eliminate most of the internal hysteresis. Full recovery can take 3–6 days, and the time period is longer at cold temperatures. This is true whether or not the accelerometer is powered by the signal conditioner.