To:



TEST METHOD FOR THE

DETERMINATION OF THE

APPARENT SPECIFIC GRAVITY OF SOILS

[pic]

GEOTECHNICAL TEST METHOD

GTM-14

Revision #1

GEOTECHNICAL ENGINEERING BUREAU

APRIL 2007

GEOTECHNICAL TEST METHOD:

TEST METHOD FOR THE DETERMINATION OF THE

APPARENT SPECIFIC GRAVITY OF SOILS

GTM-14

Revision #1

STATE OF NEW YORK

DEPARTMENT OF TRANSPORTATION

GEOTECHNICAL ENGINEERING BUREAU

APRIL 2007

TABLE OF CONTENTS

1. SCOPE 3

2. DEFINITION 3

3. APPARATUS AND SUPPLIES 3

4. SAMPLE PREPARATION 4

5. TEST PROCEDURE 4

6. PERIODIC MAINTENANCE 6

APPENDIX 8

A. Specific Gravity Conversion Factor - K Factor (Table 1) A-1

B. Apparent Specific Gravity Determination (Worksheet) B-1

C. Procedure for Calibration of the Volumetric Flask for use

in the Determination of the Apparent Specific Gravity of Soils C-1

1. SCOPE

1. This manual describes the procedure of the New York State Department of Transportation Geotechnical Engineering Bureau to determine the apparent specific gravity of a soil. Values presented in this manual without a decimal point or a tolerance is approximate.

2. DEFINITION

1. Specific Gravity: The ratio of the mass in air of a given volume of a material at a stated temperature to the mass in air of the same volume of gas-free water at a stated temperature.

3. APPARATUS AND SUPPLIES

1. Oven: A thermostatically controlled oven capable of maintaining temperatures of 230±9° F (110±5° C) for drying the hydrometer analysis samples.

2. Balance: An AASHTO M-231, Class C balance for weighing the material to be used in the analysis.

3. Thermometer: A thermometer conforming to ASTM 14.01.

4. Sieve: A ¾ in. (19.0 mm), ¼ in. (6.3 mm), No. 10 (2.00 mm) or No. 40 (0.425 mm) sieve conforming to the requirements of AASHTO Designation M-92 (excluding Column 7 of Table 1, p. 128) and a pan.

5. Flask and Rubber Stoppers: Calibrated flasks (refer to Appendix C) of 0.5 qt. (500 mL) and 1.1 qt. (1000 mL) capacities in which to de-air the soil samples and water. A rubber stopper is required for each flask containing de-aired water or water and soil to seal the flask in order to minimize the absorption of air by the water or soil sample.

6. Pulverizing Apparatus: A mortar and rubber covered pestle, or similar device as specified in AASHTO T-87-86, suitable for separating the aggregated soil particles without reducing the size of the individual grains.

7. Desiccator: A desiccator in which to allow the soil samples to cool without absorbing moisture after the samples have been dried in an oven.

8. Funnel: A funnel to aid in transferring the soil sample from the tare to the flask

9. Potable Water: A supply of potable water to de-air for filling the calibrated flasks.

10. Water Dispenser: A dispenser for de-aired water to wash down the sides of the flask during the de-airing process.

11. De-airing System: One or both of the following de-airing systems:

3.11.1 Method A: Vacuum: For de-airing water and most soils. It is recommended that two vacuum systems be available to de-air two flasks simultaneously so as to both save time and to minimize the error introduced by the de-aired water or soil absorbing the air in the flask.

3.11.1.a Vacuum Pump: A vacuum pump capable of sustaining a vacuum of at least 27.0 in. (685 mm) Hg at standard temperature and pressure.

3.11.1.b Filtering Flask: A heavy walled filtering flask with the side opening attached by plastic or rubber tubing to the vacuum pump and the top opening attached by plastic or rubber tubing to a stopper to be inserted in the calibrated flask. The filtering flask acts as a trap to prevent soil particles from passing into the vacuum pump.

3.11.1.c Safety Cabinet: A rigid enclosure with an access door, made of Lucite, Plexiglass, or other suitable material, in which to place the filtering flask and calibrated flask during de-airing. The purpose of the safety enclosure is for protection in the event of an implosion of the flask when subjected to the vacuum.

3.11.1.d Vacuum Gauge: A standard vacuum gauge, for measuring the pump vacuum, capable of measuring the range of 0 to 30 in. (0 to 760 mm) of Hg.

3.11.2 Method B: Boiling: For de-airing soils that effervesce violently when subjected to a vacuum.

3.11.2.a Hot Plate: A hot plate on which to heat the soil sample and water to boiling. When heating, the calibrated flask containing the soil sample and water should be placed in a pan containing a ½ to 1 in. (12.7 to 25 mm) thick layer of sand to evenly distribute the heat from the hot plate.

3.11.2.b Glass Funnel: A small glass funnel to be used as a vent cap on the flask to avoid loss of material when boiling the soil sample.

4. SAMPLE PREPARATION

1. Dry the soil sample in an oven at 230±9° F (110±5° C) for at least 12 hours or until the sample achieves a constant weight, whichever occurs first.

2. Place the sample in the desiccator and allow cooling to room temperature.

3. Carefully break up the soil agglomerations using the pulverizing apparatus without reducing the size of the individual grains.

4. Separate the sample on the desired sieve size (¾ in. (19.0 mm), ¼ in. (6.3 mm), No. 10 (2.00 mm) or No. 40 (0.425 mm). For material finer than ¼ in. (6.3 mm), 100 g of material are normally used (200 g for minus ¾ in. (19.0 mm) samples). However, if sample material is limited, as little as 50 g (100 g for minus ¾ in. (19.0 mm) samples) may be used.

5. TEST PROCEDURE

1. Add sufficient potable water to a 0.5 qt. (500 mL) calibrated flask to saturate the soil sample to be tested.

Note 1: For a minus ¾ in. (19 mm) sieve sample, use a 1.1 qt. (1000 mL) flask.

2. Thoroughly dry the outside of the flask and the inside of the neck and weigh the flask and water. Record the weight to the nearest 0.01 g on Form SM 294b Apparent Specific Gravity Determination, Line 7 (Appendix B).

Form SM 294b uses the International System of Units (g, °C) for recording the weights (Weight of flask & water, Weight of flask, soil & water, etc.) and temperature.

3. Place a funnel in the flask and add approximately 100 g of the soil sample to the flask.

Note 2: For a minus ¾ in. (19 mm) sieve sample, add 200 g of the soil sample to the flask.

4. Remove the funnel, clean off the outside of the flask and the inside of the neck and weigh the flask with the water and soil sample. Record the weight to the nearest 0.01 g on Line 6.

5. Wash down the inside neck of the flask using the water dispenser to wash the fines into the body of the flask.

6. De-air the sample by either of the following methods:

Note 3: Some samples may effervesce violently when subjected to a vacuum. These samples should therefore be de-aired by boiling.

1. Method A: Vacuum: Place the flask inside the safety cabinet and apply a vacuum to the flask to de-air the soil sample. Occasionally swirl the flask gently to facilitate the release of air. For safety considerations, release the vacuum before swirling the flask. The door of the safety cabinet should be closed while the flask is subject to the vacuum.

Note 4: If any portion of the sample is splashed above the calibration line, it must be washed down immediately with de-aired water. If any of the sample is pulled by the vacuum from the flask, the test is invalidated and must be rerun.

2. Method B: Boiling: Insert a small glass funnel in the neck of the flask to prevent the loss of any of the soil sample. Heat the sample to boiling using the hotplate and the pan containing the ½ to 1 in. (12.7 to 25 mm) thick layer of sand. Boil the sample for one half hour, then cool to room temperature.

Note 5: If any portion of the sample is splashed above the calibration line, it must be washed down immediately with de-aired water.

7. Fill a 1.1 qt. (1000 mL) flask to approximately the middle of the flask bulb with portable water at about 86° F (30 ° C). Apply the vacuum to the flask to de-air the water. Occasionally swirl the flask to facilitate the release of the air. The door of the safety cabinet should be closed while the flask is subject to the vacuum.

8. When the water or soil and water sample has been de-aired, remove the flask from the vacuum. De-airing is adequate when no bubbles can be seen after gently swirling the flask.

Note 6: Flasks containing a de-aired soil sample or de-aired water must be kept stoppered to minimize the absorption of air.

9. Using de-aired water, fill the flask containing the de-aired soil sample to the point that the bottom of the meniscus is at the calibration line. Carefully add the de-aired water by pouring down the side of the flask to avoid mixing air into the sample.

10. Dry the outside of the flask and the inside of the neck without disturbing the meniscus. Weigh the flask with the soil and water to the nearest 0.01 g and record the weight on Line 12.

11. Place the thermometer in the flask, read the temperature to the nearest 0.1° C and record the value on Line 9.

12. Determine the Calibrated Flask Reading from the appropriate Table of Weights for Calibrated Flask and Water vs. Temperature for the flask used (refer to Appendix C) and record the value on Line 10.

13. Round the temperature to the nearest 0.5° C and determine the K Factor from Table 1 (Appendix A). Record the value of the K factor on Line 15.

14. Add the weight of dry soil (Line 8) and the weight of the flask and water from the appropriate table (Line 10) and enter the result on Line 11.

15. Find the weight of displaced water by subtracting the weight of the flask, water and soil (Line 12) from the calibrated weight of the flask and water plus soil (Line 11). Enter the result on Line 13.

16. Calculate the specific gravity by dividing the weight of dry soil (Line 8) by the weight of displaced water (Line 13) and enter the value on Line 14.

17. Correct the specific gravity to 68° F (20° C) by multiplying the specific gravity (Line 14) by the K factor (Line 15) and enter the result on Line 16.

6. PERIODIC MAINTENANCE

1. Following each test:

1. Wash and air-fry the flask used in the test. Use care to avoid scratching the inside of the flask with the wire cleaning brush.

2. Inspect the flasks for cracks or defects. Discard damaged or defective flasks.

3. Return the flasks to the storage cabinet to reduce the possibility of damage.

2. Daily:

1. Drain the accumulated water from the vacuum pump. Open the stopcock and drain the fluid until only oil flows out.

2. Check the oil level and fill as needed. The oil level should be slightly below the fill line shown on the window to allow for expansion of the soil when the pump is warm.

3. Weekly:

1. Wipe the oil and grime from the surface of the vacuum pump.

2. Check the vacuum of the pump. Attach a vacuum gauge to the suction line and verify the vacuum to be at least 27.0 in. (685 mm) of Hg.

4. Periodically:

1. Check the condition of the plastic or rubber vacuum tubes and replace when necessary.

2. Clean the filtering flask as needed.

3. Check the accuracy of the balance and recalibrate if necessary.

4. Check the dry weight of the calibrated flask. If the dry weight varies by more than 0.01 g from the weight used in the calibration, check the calibration of the flask.

APPENDIX

|Specific Gravity Conversion - K Factor |

|Temp (° C) |K |Temp (° C) |K |

|16 |1.0007 |23.5 |0.9992 |

|16.5 |1.00065 |24 |0.9991 |

|17 |1.0006 |24.5 |0.9990 |

|17.5 |1.0005 |25 |0.9989 |

|18 |1.0004 |25.5 |0.99875 |

|18.5 |1.0003 |26 |0.9986 |

|19 |1.0002 |26.5 |0.99845 |

|19.5 |1.0001 |27 |0.9983 |

|20 |1.0000 |27.5 |0.99815 |

|20.5 |0.9999 |28 |0.9980 |

|21 |0.9998 |28.5 |0.99785 |

|21.5 |0.9997 |29 |0.9977 |

|22 |0.9996 |29.5 |0.99755 |

|22.5 |0.9995 |30 |0.9974 |

|23 |0.9993 |30.5 |0.99725 |

Table 1

[pic]

Procedure for the Calibration of the Volumetric Flask for use in the

Determination of the Apparent Specific Gravity of Soils

TABLE OF CONTENTS

1. SCOPE C-2

2. CALIBRATION PROCEDURE C-2

3. COMPUTATIONS FOR FLASK CALIBRATION CURVE C-3

4. FLASK CALIBRATION WORKSHEET (Table 1) C-4

5. FLASK VOLUMETRIC TEMPERATURE CORRECTION CURVE (Figure 1) C-5

6. GRAPH FOR COMPUTED CALIBRATION CURVE (Figure 2) C-6

7. TABLE OF WEIGHTS FOR CALIBRATION FLASK

AND WATER vs. TEMPERATURE (°C) (Table 2) C-7

Procedure for the Calibration of the Volumetric Flask for use in the

Determination of the Apparent Specific Gravity of Soils

1. SCOPE

1. This appendix describes the procedure of the New York State Department of Transportation Geotechnical Engineering Bureau to calibrate the volumetric flask used in the determination of the apparent specific gravity of a soil. Values presented in the appendix without a decimal point or a tolerance is approximate.

2. CALIBRATION PROCEDURE

1. Inspect the flask to be calibrated to ensure that it is free of cracks and imperfections in the glass. Assign the flask a unique alphabetic or numeric identification and permanently mark the designated characters on the flask.

2. Wash the flask with soap and water. Rinse and drain the flask and place in a 230° F (110° C) oven to dry overnight.

3. Remove the flask from the oven and allow cooling to room temperature. Make certain there is no moisture on the inside or outside of the flask. Weigh the dry flask and record the weight to the nearest 0.01 g on the Flask Calibration Worksheet (Table 1), Line 3 and Line 11, and on the graph for the computed Calibration Curve (Figure 2).

4. Fill a 1.1 qt. (1000 mL) flask to approximately the middle of the bulb with potable water at a temperature such that the temperature of the water after de-airing is between 64° and 72° F (18° and 22° C). The de-airing process causes the water temperature to decrease approximately 9° F (5° C) in approximately 15 minutes of operation.

5. Fill the flask to be calibrated with de-aired water so that the bottom of the meniscus is at the calibration line. Dry the outside of the flask and the inside of the neck. Use care when drying the inside of the neck to avoid touching the meniscus. Weigh the flask and water and record the weight to the nearest 0.01 g on the Flask Calibration Worksheet, Line 2.

6. Place the thermometer in the flask, read the temperature to the nearest 0.1° C and record the value on Line 1.

7. Subtract the weight of the dry flask (Line 3) from the weight of the flask and water (Line 2) and enter the result on Line 4.

8. Use the Flask Volumetric Temperature Correction Curve (Figure 1) to determine the value of (1+3tE)G1 for the temperature recorded on Line 1. Enter the appropriate value on Line 5.

Note: The flask is calibrated by the manufacturer at 68° F (20° C). Weights at temperatures other than 68° F (20° C) must be corrected to agree with the weight corresponding to the volume of water in the flask at 68° F (20° C). The weight of water in grams equals the volume of water in cubic centimeters at 39° F (4° C). Above 39° F (4° C), the volume of the flask and water will expand in accordance with (1+3tE)G1.

9. Divide the weight of water (Line 4) by the value of (1+3tE)G1 (Line 5) to obtain the volume of the flask. Enter the value on Line 6.

10. Repeat Section 2.4 through 2.9 until five values of the volume of the flask (Line 6) are within 0.1 cc.

3. COMPUTATION FOR FLASK CALIBRATION CURVE

1. Average the five values of the volume of the flask obtained from Section 2.10 and enter the result on Line 9.

2. Convert the average volume of the flask to the average weight of water in the flask at the given temperatures of 54°, 61°, 68°, 75°, 82° and 90° F (12°, 16°, 20°, 24°, 28° and 32° C) (Line 7) by multiplying the value of (1+3tE)G1 at the given temperature (Line 8) times with average volume of the flask (Line 9). Enter the result in the appropriate column on Line 10.

3. Add the weight of the water at the given temperature (Line 10) to the weight of the dry flask (Line 11) and enter the sum in the appropriate column on Line 12.

4. Plot the values of the weight of the flask and water (Line 12) versus the corresponding values of temperature (Line 7) on the graph for the computed Calibration Curve (Figure 2).

5. The values for the weight of the flask and water can be obtained directly from the Calibration Curve, or the Calibration Curve can be used to generate a table of values for the weight of the flask and water for temperatures from 61° to 82° F (16° to 28° C) in 1/10° increments (Table 2).

[pic]

Table 1

[pic]

Figure 1

[pic]

Computed Calibration Curve

Figure 2

TABLE OF WEIGHTS FOR CALIBRATED FLASK AND WATER VS. TEMPERATURE (°C)

(CALIBRATED FLASK READING)

Comp. by:________________

FLASK DESIGNATION ____________ Check by:________________

|Temp (°C) |Wgt. Of Flask and Water|Temp (°C) |Wgt. Of Flask and Water|Temp (°C) |Wgt. Of Flask and Water|

|16.0 | |20.0 | |24.0 | |

|.1 | |.1 | |.1 | |

|.2 | |.2 | |.2 | |

|.3 | |.3 | |.3 | |

|.4 | |.4 | |.4 | |

|.5 | |.5 | |.5 | |

|.6 | |.6 | |.6 | |

|.7 | |.7 | |.7 | |

|.8 | |.8 | |.8 | |

|.9 | |.9 | |.9 | |

|17.0 | |21.0 | |25.0 | |

|.1 | |.1 | |.1 | |

|.2 | |.2 | |.2 | |

|.3 | |.3 | |.3 | |

|.4 | |.4 | |.4 | |

|.5 | |.5 | |.5 | |

|.6 | |.6 | |.6 | |

|.7 | |.7 | |.7 | |

|.8 | |.8 | |.8 | |

|.9 | |.9 | |.9 | |

|18.0 | |22.0 | |26.0 | |

|.1 | |.1 | |.1 | |

|.2 | |.2 | |.2 | |

|.3 | |.3 | |.3 | |

|.4 | |.4 | |.4 | |

|.5 | |.5 | |.5 | |

|.6 | |.6 | |.6 | |

|.7 | |.7 | |.7 | |

|.8 | |.8 | |.8 | |

|.9 | |.9 | |.9 | |

|19.0 | |23.0 | |27.0 | |

|.1 | |.1 | |.1 | |

|.2 | |.2 | |.2 | |

|.3 | |.3 | |.3 | |

|.4 | |.4 | |.4 | |

|.5 | |.5 | |.5 | |

|.6 | |.6 | |.6 | |

|.7 | |.7 | |.7 | |

|.8 | |.8 | |.8 | |

|.9 | |.9 | |.9 | |

Table 2

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download