Leveling screws are made of stainless steel. Their threads are 1/4inch diameter with 48 threads per inch. The tips are ground to a sharp point to minimize the meter moving on the baseplate while the meter is being leveled.
When the meter is at L and R for standard cleaning and adjusting, the leg tips are resharpened if needed.
The threads of the legs turn in special metal inserts that are glued into the bottom of the white box. To reduce the chance of the stainless steel threads galling, the inserts are made of a lead rich steel. This alloy is capable of corrosion. Always dry the carrying case and meter legs when they become damp.
Never store the meter in a damp case. A light lubricant on the threads will reduce the risk of corrosion.
CAUTION, do not leave the legs extended too far below the base of the white box. They will punch through the base pad of the carrying case, bump the bottom of the carrying case and jar the meter. Also, there is a greater risk of damage to the threads.
If the instrument has received a hard blow, the meter may shift inside the white box. This can cause the Nulling dial and arrestment shaft to be off center as they pass through their holes in the black lid. This is of little consequence other than appearance. However, the dial pointer is fixed to the black lid and may now rub against the dial. To correct this problem, the single hole in the pointer can be filed oblong and the pointer replaced on the black lid a little farther away from the dial.
The nulling dial is a flat aluminum disk with a hundred division marks etched on it and a knurled knob to assist in turning the dial. Except for early meters, the knob has two holes drilled in it to accept the prongs of the high speed crank. The high speed crank is mentioned in the Options section. It is convenient when the nulling dial must be turned through a great range of the meter.
To remove the nulling dial, loosen the two set screws within the knurled knob. Do not put excessive force on the nulling shaft. Hold the screw driver with one hand and the knob with the other hand. Below the dial are three parts that act as an adjustable friction brake for the dial. When replacing the dial, the farther down the dial is pressed against the spring washer, the greater the friction when rotating the dial. The ideal setting is to have just enough friction that there is no backlash of the dial. When the dial is reinstalled on the meter, be sure to phase the number on the dial with the last number on the counter.
The focus of the eyepiece (ocular) is adjustable. To move the eyepiece up or down, first loosen the knurled ring below the top of the eyepiece. Turning counterclockwise loosens. While looking through the eyepiece, move it up or down until a good focus is found. Then, retighten the knurled ring to hold the eyepiece in position.
The arrestment knob is knurled and made of aluminum. It is held on the shaft by two very small set screws. It is positioned very close to the arrestment stalk to minimize the chance of dirt entering the top of the stalk.
If something is dropped on the top of the knob, the knob may be driven down and drag on the stalk. To repair, use a 3/32inch Allen tool to loosen the knob set screws, raise the knob about a tenth of a millimeter (a few thousands of an inch) and retighten the set screws.
If the meter is subjected to salt spray or is stored damp, enough corrosion may develop between the knob and the stalk to cause dragging. Repair in the same way as above.
If the knob is very hard to turn and is not dragging on the stalk, the problem is more serious. The bearings or seals in the stalk may require servicing at L and R.
The black lid is made from 1/8 phenolic board. The color was selected to minimize reflected light which could hamper viewing through the microscope. Each lid is individually fitted to its white box and meter. It is not weather resistant. Two changes are in progress:
The lids will be interchangeable starting approximately at meter G1000 and D180. This will make easier retrofitting the meters with new electronics and LCD display. The lid will be more weather resistant.
To remove the black lid, turn the three leveling knobs counter clockwise and lift the level legs up and out of the meter box. Remove the nulling dial and the three brake parts beneath it. Remove the four lid screws, one located at each corner of the lid. Now lift the lid upward. It may be stuck to the top of the white box. Loosen with care.
Some of the glues used in the past to fix the window into the bottom side of the black lid were water soluble. When these meters are used in moist conditions, the windows loosen. To repair, remove the black lid, clean away the old glue and re-glue. If "Super Glue" is used, only 3 or 4 drops are needed, as surface tension will distribute the glue. Allow to dry completely before replacing the black lid. If not vented, the fumes will condense and make a milky coating on the glass.
If replacements are required and it is not convenient to obtain them from L and R, you may have them cut from thin glass. The easiest source of this glass is a microscope slide.
Starting approximately in 1980 with meters G553 and D46 the meters have been equipped with a fused safety device which is intended to help protect the electronic circuitry from reversed polarity of the power leads and/or power input overvoltages of greater than 22 volts. This is not to imply tolerance of reversed polarity, or overvoltage greater than about 16 volts.
The fuse is located near the power input connector. The black bakelite meter lid must be removed to replace the fuse. Use only a 3 amp ATO fuse for replacement. Spare fuses are provided with new instruments with this protection. They are located in a small box taped to the inside of the lid of the aluminum carrying case. Do not bypass or use a larger rated fuses.
A blown fuse will shut down all electronic circuitry in the instrument: heater circuit, electronic readout, and level and reading lights.
If you are out of fuses they may be obtained from L and R or most automotive parts stores.
Whenever older meters are returned for service, we encourage the retrofitting of the fuse, especially if the meter has the CPI option.
This switch is a standard single pole, single throw toggle switch. It can be replaced with any miniature SPST toggle switch.
In older meters that have not been converted, the reading light is on the operator's side of the white meter box. It is behind a round aluminum cover plate. The plate is held in place by three screws. The housing for the reading light is adjustable. It holds a No.330 12volt incandescent aircraft bulb. Spare bulbs are stored in a small container taped to the inside of the carrying case lid. The bulb can be rotated so its filament is optimal for the optics and the bulb holder can be moved to place the bulb in the best position for brightness and sharpness of the image. Two screws secure the bulb holder in place.
A high intensity red LED lamp became standard on all Model G meters without CPI since approximately G970. It is an option available on all land meters and is of value in hot climates. It operates from a lower voltage, requires a change in wiring and has a different lamp holder.
On the upper right side of the white box is the electrical fitting that passes 12volt power into the meter. There are two types. They look almost identical. The early type was used from 1967 through 1984 on meters from about G154 through about G742 and D1 through D105. These fittings are characterized by their cable orientation key being under the cable and by two power pins being inline with the cable. These fittings and their cables are obsolete and can not be obtained. Since changing the power fitting to the new type is more difficult than at first appearance, it is imperative that these obsolete fittings and their obsolete cables be replaced when the meter is at L and R for other servicing. Should the obsolete cable fail in field use, the meter would have to be returned to L and R or some primitive splice or solder connection made in the field.
The new Mark power fitting can be recognized by the orientation key being on the top and the power pins being 90° from the direction of the cable. Also, there is a halfround groove just inside the socket that accepts a halfround protrusion on the molded cable fitting at the meter end of the power cable. The new Mark fitting solved a serious drawback of the older fitting. The new fitting has a halfround protrusion on the cable fitting and a matching halfround groove in the metal receptacle. These hold the cable in place and a good electrical contact is assured. Cables with the old type Mark fitting sometimes were bumped loose when returning the meter to the carrying case. The meter would go off power and temperature, resulting in errors in the observations.
Very early Model G meters had an Amphenol 126216 power connector. Most of these have been replaced with old or new type Mark connectors.
Prior to 1984 (about G720 and D99) batteries and chargers were equipped with threepin Bendix JT02A83S connectors. These became very high priced and then obsolete. Since 1984, the less expensive and more common twopin Cannon KPT06B82S connector has been used. The matching cable connectors are JT06A83PSR and KPT06B82P, respectively. As meters are serviced at L and R, the 3pin connectors are being updated to 2pin connectors.
Three types of levels are available for the land meters. The standard and original levels are spirit bubble (liquid) levels. In 1979 electronic levels became an option. These are high precision air damped pendulums. Their position is measured with a capacitance bridge and registered on two galvanometers or an LCD (liquid crystal display). In 1989 inexpensive electronic levels were introduced. They are liquid levels with electrodes and conductive fluid.
These are the original levels used in almost all L and R land meters. In most Model G meters their sensitivity is 60 seconds per division and in most Model D meters their sensitivity is 30 seconds per division. They can be stored at temperatures down to 51°C(60°F). The lowest temperature for correct reading is 18°C(0°F). The levels in the early meters are illuminated from below by a No. 327 bulb when the reading light switch is turned on. This is a 28volt aircraft bulb operating at 12volts. Its lifetime should be very long. It looks almost identical to the reading bulb (No.330). The 330 should NOT be used to illuminate the level.
Since approximately meter G970, all meters without CPI have high intensity LED lamps for the levels and for optical reading of the meter beam position. Since 1996 all meters are equipped with LED level and reading lamps. These low heat lamps are optionally available on all meters returned for service. They are of great value to meters that will be used in warm climates. The lamps operate from lower voltage and thus the meter wiring is different than for the incandescent No.330 and 327 lamps. Also, the lamp housing are different.
The bubble levels have three distinct disadvantages:
The levels are adjusted with a hexagonal allen tool provided with each meter. The adjusting screw is at one end of each level and is reached through a small access hole in the black lid. The hole is covered with a small square black metal plate. Loosen the small screw holding the plate in place and rotate the cover plate to expose the access hole.
Some older Model G meters before approximately G350 do not have externally adjustable levels. It is necessary to remove the black lid to make level adjustments. Two special tools were provided with these meters to facilitate level adjusting. One is an aluminum tube to hold the eyepiece while the black lid is removed and the other is a special wrench for adjusting the level nuts. Many of these meters have been retrofitted with external access level adjusting and retrofitting should be considered when these older meters are at L and R for other servicing.
These levels are ideal except for their high cost. They overcome all of the above mentioned disadvantages of the spirit bubble levels. For specifications and a diagram, see the Options section.
Electronic level adjustment procedures are similar to those used for the spirit levels. The directions for spirit levels can be utilized, realizing that 4 galvanometer increments equal approximately 1 bubble division: 15 seconds of arc for a Model G and 7 or 8 for a Model D. If more or less sensitivity is desired, it may be adjusted at L and R.
The longer level adjusting tool is required to adjust the electronic levels, as they are farther below the meter lid than are the spirit levels.
Once adjusted, the pendulum levels are very stable and remain in good adjustment. However, their adjustment should be checked whenever the meter is shipped or encounters a drop or severe bump.
The easiest method of initially adjusting the electronic levels is to first adjust the spirit levels in accordance with this manual. Then, with the Allen wrench, reset the electronic level adjustment screws until both leveling galvanometers display 0 (center of their range). The locations of these adjustment screws are shown on the following illustration. Do not readjust the electronic levels on a daily basis as they are more stable than the spirit levels. The optical reading line may increase one small division when going to high latitudes or decrease one small division when going to equatorial latitudes.
When the gravity meter is moved over very large differences in latitude, displacement sensitivity changes can occur. If the sensitivity change is large, both long levels (electronic and spirit) must be readjusted. This will cause a small difference in the reading line position, therefore it should be adjusted also.
Using the electronic levels, level the gravity meter carefully.
Using the optical readout, set the microscope cross hair on the reading line specified for the gravity meter. If the reading line is not known, assume a reading line position midway between the beam stops. With the gravity meter nulled at the above optical reading line, adjust the zero of the electronic beam readout microammeter to make it read zero. Use the electronic beam readout in all further adjustments.
Keeping the crosslevel in the level position, tilt the longitudinal level 2 small scale divisions on the longitudinal level microammeter, first in one direction and then in the other. Note both the magnitudes and displacements of the corresponding beam microammeter. Repeat several times to be sure of the results.
If the beam microammeter moves to the right (upscale in the microscope) the same amount for each tilt, then the chosen reading line is correct.
If the reading line is not correct, proceed as follows. Make the previously described 2division long level tilt by lowering the side of the gravity meter which has a single leveling screw. If the beam microammeter then moves upscale more than it does with the opposite tilt, the assumed reading line is too high (too far to the right on the beam microammeter). Otherwise the reading line is too low.
Shift the reading line 2 microammeter divisions in the direction determined above and repeat the above step. Continue shifting the assumed reading line position until the correct position is found or passed. This procedure will determine the correct position for the reading line within 2 microammeter divisions or better. Continue adjusting until the reading line is determined as well as possible. After this has been done, reset the zero of the beam microammeter to make it read zero at the correct reading line.
If there is a need to replace the pendulum level in the field, care must be exercised. If the wire connected to the side of the level must be resoldered, over heating can damage the interior of the level. The heat should be applied to the tip of the wire lug. A heat sink should be located on the lug between the level housing and heat. A small pair of needlenose pliers should be adequate as a heat sink.
There is a small chance that a level might stick to one end of the range of travel. A very light tap of the finger on the meter lid will free the pendulum. When the meter is next at L and R for servicing, the sticking level will be cleaned as part of the standard servicing.
These levels are a compromise. They are low in cost and they overcome the first two disadvantages of the spirit bubble levels. Optimum damping for a level would be about 70% of critical. The pendulum levels have optimum damping. At thermostated temperature, the liquid electronic levels are damped to approximately 40% to 50% of critical.
They consist of a bubble inside a curved glass tube. The fluid is a conductor and there are electrodes in the tube to sense the change in resistivity as the bubble moves along the curved tube. Since they are located just above the sensor in a thermostated area, they are not affected by ambient air temperature or sunlight.
The CPI was mentioned in the Options section of this manual. It is the most popular and important option. It senses the position of the beam with greater accuracy and less fatigue than the microscope. It is a capacitance bridge with a capacitance plate on the beam and a fixed plate above the beam and another fixed plate below the beam.
The electronic card of the CPI fits into a pocket on the outside of the meters heater box. The card has had three main types during its evolution: discrete components card, Hybrid card and Ncard.
Discrete component cards were used from about G130 (1967) and the first Model D meters until approximately G375 and D10. Hybrid cards were used from then until G472 and D25. The Type N card was then introduced.
No Model G meters before G100 had a CPI as original equipment. Their type of card depends on when they were retrofitted with CPI.
When the Hybrid card is damaged by reverse polarity of the power supply, it is not repairable. The Ncard is usually easy to repair.
The following three diagrams show some of the details of the CPI.
There is a small rectangular black cover plate on the far side of the black lid. Beneath it are two variable resistors. One adjusts the gain of the galvo and the other sets the zero position of the galvo. Common gain settings for the Model G are 100 microGals per large galvo division for regional surveys and 50 microGal for more detailed surveys. For Model D, typical settings are 40 microGals for average work and 20 or even 10 microGals per large division for high precision work.
Whenever the galvo gain is adjusted the galvo zero position must be reset. Null the meter with the microscope. When the cross hair is on the reading line, turn the zero variable resistor until the galvo pointer is on the exact center of the dial.
Some meters have an extra switch on the black lid. It allows switching from low galvo sensitivity (gain) to high sensitivity. On most meters the galvo is set so one eyepiece division is equal to one or two galvo divisions. With this switch, the meter is usually set so one eyepiece division equals two galvo divisions on low and five divisions on high. On most meters with the switch, the gain may be set as high as 10 microGals per galvo division.
To set the sensitivity, place the switch in the high position. Set the sensitivity as outlined in the manual, to the desired level. Remember to reset the zero position after any change of galvo sensitivity. Then with the gain switch in the LOW position, adjust the HILOW gain pot (variable resistor) to the desired level. This gain pot is just a resistance in series with the galvo.
With the electrostatic nulling option, changing the galvanometer sensitivity will alter the time constant of the feedback loop and increase the amount of galvanometer jitter induced by the electrostatic nulling system.
Meters before about G467 and D27 have round galvanometers. These are obsolete and new replacements are not available. Occasionally a used one becomes available. There is a short pair of wires on these galvos and an electrical connector. Exchanging them is straightforward.
When the round galvanometer became obsolete, it was replaced with the rectangular galvanometer. These are less satisfactory. A single drop of rain can be drawn into the galvo by surface tension. Condensation on the working surfaces can cause the galvo to be sluggish and inaccurate. After the galvo dries it may work satisfactorily. However, there is a chance that subsequent corrosion may cause the galvo to fail at some time in the future. A very small line of clear silicone rubber between the clear plastic galvo cover and the black lid will give some protection from water but cause some bother whenever the black lid must be removed.
Replacing the rectangular galvanometer requires a soldering iron.
Development is in progress of a liquid crystal display (LCD) that will replace the galvo as well as perform other functions.
Normal beam damping is about 70% of critical damping. The variable damper allows normal beam damping to be increased by a factor of 25. Normal damping is accomplished by two air dampers located on the bottom of the beam. There is a small damper near the back of the beam and a main damper near the front or mass end of the beam. Each damper consists of a set of concentric inverted cups on the bottom of the beam and another set of concentric cups fixed to the bottom of the sensor box. The beam cups and fixed cups interfinger. To pull the cups apart, air must flow up and down several times to reach the inner cup of the damper. Meters with variable damping have extra cups in the main damper and a simple rotary valve to open some of the cups when full damping is not desired.
To adjust the damping, turn the meter on its side and remove the nylon screw on the underside of the white case. Some early models have an aluminum cover plate.
The screwdriver slot of the lower adjusting screw will then be visible. Note on the drawing that a coil spring keeps the lower screw in the downward position. With a screwdriver in the slot of the lower adjusting screw, push inward and engage the slot of the upper adjusting screw. For maximum damping the screw is turned clockwise about one eighth revolution (three quarters of a turn on some earlier meters). A very small amount of torque will be required to turn the adjusting screw and a positive stop can be felt. As the screwdriver is removed, the coil spring should force the lower adjusting screw away from the upper screw and back to its original position.
For minimum damping (that of a standard gravity meter), the above procedure is reversed. When maximum damping is engaged the damping will be approximately 50% that of the L and R underwater gravity meter and about 25 times that of a standard land meter.
So that faster readings may be obtained when maximum damping is engaged, an electrostatic beam positioner is provided. The beam positioner switches are the momentary switches located on the operator side of the top of the gravity meter. These switches are also called "cheater" switches. To move the beam (crosshair or galvo) upscale, the switch on the right is pressed, to move the beam downscale, the switch on the left is pressed. The beam is positioned at the reading line with a touch of the left or right switch. If the crosshair then drifts to the left, the nulling dial must be rotated clockwise to a higher reading. Again the beam is positioned at the reading line and the process repeated. The slower the crosshair moves away from the reading line the less the meter is away from the null (balanced) reading. The beam positioner will also operate when the meter is set for standard damping but will generally not be needed.
A solid state power supply provides the voltage (about 60 volts) for the beam positioner.
In using the beam positioner, it is important to keep in mind that the electrostatic force can only overcome approximately 75 milligals of spring tension. therefore, if the meter is set more than 75 milligals from the value of gravity, the beam positioner will not move the beam. It will then be necessary to use the spring tension adjustment (nulling dial) in conjunction with the beam positioner initially to get the meter in range of the reading. It may be necessary to tap the meter gently if the beam is slightly sticking on a stop. This may help avoid an excessive amount of dial turning.
For the gravity meter to operate properly it is important that the lower adjusting screw is down and away from the upper adjusting screw (See the cross sectional view).
Meters with variable damping have two switches on the black lid on the side toward the operator. They are for electrostatically pushing the beam to the null position. These switches can fail when subjected to excessive dust, mud or water. If this occurs in a location where replacement is not available, it is possible to remove the switch, and connect the wires in such a way that will allow the meter to function.
To remove the cheater switch temporarily, connect the brown wire to the green or violet wire, depending on which switch is removed. Do not connect the yellow wire to any other wire. Insulate the conductor of the yellow wire with tape.
The nulling counter is like an odometer on a car. Failure is very rare. If it does occur, it is best to send the meter to L and R. However, if this is difficult, it can be replaced in the field.
NOTE: This procedure can cause major damage to the instrument if the counter synchronization with the measuring screw is lost. That is the reason the two counters must be set to the same value and the stalk shaft not turned while the counter is not engaged.
Prior to mid 1966, a mercury thermostat controlled the heater. From G125 and D1 onward, a thermistor bridge has been used. Most of the early meters with mercury thermostats have been retrofitted with thermistor bridges and their mercury thermostats removed.
The heater control module is a brown or black cube about 2x2x2.5cm in size. It has two electrical connectors and red LED light. It is located under the Dacron insulation and rests atop the meters inner aluminum lid. Replacement is straightforward.
The temperature control system associated with the ThermistorTransistor Heating Circuit is solid state. A temperaturesensing thermistor bridge operates an amplifier and power transistor to control the current to the heating element.
The thermistor bridge contains two legs of thermistors and two legs of balancing resistors. One of the resistor legs normally consists of one 10k ohm resistor. This leg is known as the "fixedleg". The second resistor leg, with a resistance of 10k ±2k ohm, is known as the "selectedleg".
In our laboratory the operating temperature is set by choosing the resistor within the selectedleg. It is, therefore, necessary to change the selectedleg resistor(s) to change the temperature of the instrument. Temperature is an inverse function of resistance. A resistance change of 800 ohm results in an approximate inverse temperature change of one degree centigrade.
Because the gravimeter has an optimum operating temperature, we do NOT recommend changing the temperature, however, under extreme environmental conditions, operating at the optimum temperature may result in erratic readings and drifts. In cold conditions this would be caused by the heater not being able to provide enough heat. In hot environments this would be caused by the instrument heating above the optimum temperature. (This would be more noticeable in extremely hot climates than in extremely cold climates.) Temperature adjustment of ±5°C or more should be avoided if possible. There are field techniques described in another section of the manual that can reduce the effects of extreme ambient air temperatures. These other techniques should be exhausted before considering a change in the operating temperature.
To change the operating temperature less than two degrees, it is necessary only to change the selectedleg resistance. (See exception below ) Remove the black lid of the instrument and enough dacron insulation to expose the heat control module next to the thermometer well.
Gently remove the selectedleg resistor(s) mounted on a green and yellow twisted pair of wires (green and red in some older meters). Remove power to the instrument when cutting or soldering any wires.
To decrease the operating temperature, add (in series) resistance at the rate of about 1000 ohm /°C. Resistors must be precision film type (RN55D). Do not change by more than 2000 ohm resistance.
To change the operating temperature more than 4°C, or if the desired instrument temperature is not within 4°C of 50°C, it is necessary to change the fixedleg resistor before changing the selectedleg resistor. This fixedresistor (normally 10k or 11k ohm) is changed to maintain the balance and equal thermistor sensitivity.
To change the fixedleg resistor, remove power from the instrument and find the brown and purple twisted wires and components. Carefully cut away the outer tube of shrinkable tubing. There will be exposed one 10k ohm resistor (the fixedleg resistor). Remove the fixedleg resistor and replace it with one of appropriate value (see the table below). Resistors should only be RN55D metal film. Use shrinkable tubing to cover bare wire and to form the outer tube shield.
Resistor (ohm) |
Temperature (°C) |
|
|
|
|
|
|
|
|
|
|
At this stage, the selectedleg resistor(s) must be balanced. The resistance should be within about 3000 ohm of the fixedleg resistor. This is more or less a trial and error selection. Begin with a resistor equal to the fixedleg resistor. Add resistance for a lower temperature; use less resistance for higher temperature.
Because of resistor and thermistor tolerances, and because thermistor temperatureresistance curves are nonlinear, the exact resistance is difficult to predetermine for a specific temperature.
It is sometimes necessary to "tune" the resistance in the selectedleg with one or two smaller valued resistors to get the exact desired temperature.
A few older ThermistorTransistor Circuit bridges use two watt wirewound resistors. These resistors are perfectly satisfactory but are larger than the precision film resistors now used.
For further information or assistance contact our laboratory.
NOTE: Whenever drastic operating temperature changes are made of more than a couple of degrees, the internal pressure/vacuum must be released. Between the gear box, the arrestment stalk, and the eyepiece stalk, there is a seal cap retained by three 256 screws. Loosen all three screws and lift the seal cap. After the temperature change has been made and the temperature stabilizes replace the seal cap and tighten the screws. Caution, be sure there are no dacron fibers on the seal cap nor on the mating surface of the gravity meter sensor lid. If the seal cap is left open, the inert dry gas may be lost from the inside of the meter. This voids the warranty.
A small sealed cylinder containing a bimetallic thermostat is located on the side of the heater box. It is in series with the heater wire. If the meter is about to overheat, the bimetallic thermostat will open the circuit and stop the heating before thermal damage can occur.
The most likely cause of the overheating has been the inadvertent reversal of the 12volt DC power to the meter. This causes the heater control to command the heater to be on, regardless of the temperature sensed by the thermistor bridge. When the bimetallic reaches 69°C, it will open and the temperature will decrease. At some temperature the bimetallic coil will close, power will flow into the heater wires, the temperature will rise to 69°C and the cycle will repeat.
On meters with the AT03 fuse and reverse polarity protection, the AT03 will prevent the reversed 12volt polarity current from entering the meter.
The meter is sealed. This increases the accuracy of the meter. If the meter were not sealed, a change in atmospheric pressure would cause a change in the buoyancy of the beam mass and would give the false appearance of a change in gravity. In case one of the seals did leak, there is a buoyancy compensator on the meter's beam. It is a large fixed volume of light mass located in back of the beam hinge or pivot point. During construction of the meter, the compensator is positioned on the beam to remove at least 98% of the effect of atmospheric pressure changes, should a leak develop in the sensor enclosure.
If the sensor develops a leak, The compensator will allow an accuracy as good as other gravity meters in good working order. However, the accuracy will be less than optimal and less than desired for careful gravity observations.
On the average, seals begin to fail after eight to ten years. They become hard and brittle. L and R recommends "long term servicing" every eight to ten years. (See the section on servicing.)
It is easy to test for a seal failure if a large change in elevation is available. A change in elevation will cause a change in atmospheric pressure. Taking a series of gravity readings after a recent change in the meter's elevation will be revealing. As air leaks into or out of the meter, there will be a change in the readings. The rate of this change will depend on the size of the leak, the magnitude of the recent change in ambient air pressure and how accurately the buoyancy compensator was set during construction of the meter. Remember, at any one location, gravity is changing due to earth tides. This change is at most one microgal per minute. If observations are made every two to five minutes, corrected for earth tides and plotted against time, the resulting curve should be fairly straight and horizontal. If it is not horizontal, it is probable that a meter seal is leaking.
When the beam is clamped, it is pushed down against the adjustable bottom stops. These are set to limit the travel of the beam to 71/2 small optical divisions of downward travel from the null reading line. Clamping stretches the main spring, compared to its length when the beam is unclamped and at the reading line.
There is a mechanism in the meter that pushes the beam backwards as it is clamped and that "push" is exactly the amount necessary to shorten the main spring back to its length when it was unclamped and at the reading line.
Were it not for the dual movement of the clamping mechanism, the spring would be stored elongated. It would relax in this stretched position and when unclamped for the next reading, reverse creep would occur. It would take ten to twenty minutes for the meter to stabilize for a reading.
The meter is carefully adjusted during construction. If needed, it is also adjusted each time it receives "standard" servicing. Only a severe blow to the meter will cause the hysteresis compensation to move out of adjustment.
An observant meter reader can notice when the meter is working properly or when it takes too long to "creep" into final null position. Remember that gravity is changing due to earth tides by up to one microGal per minute. This should not be mistaken for mechanical hysteresis. A computed table of earth tide values will allow the observer to separate the earth tide changes in meter response from the hysteresis changes. For first time testers of meter hysteresis, it simplifies the observation if earth tides are not changing. Such a time can be selected by looking at a table of earth tides.
A more detailed and lengthy test for mechanical hysteresis is performed at L and R during "standard" servicing.
The purpose of this option is to improve the accuracy of field observations, especially in cold climates and to improve the accuracy of stationary observations when the land meter is used to record earth tides.
The double oven option was introduced in 1991. The meter's standard oven is placed inside an outer oven. The space between the two ovens is rigid urethane thermal insulation. Between the outer oven and the meter box is more urethane insulation. The inner oven is thermostated at its regular temperature (usually between 50°C and 56°C) and the outer oven is thermostated at about 38°C.
The configuration of early models is a cylindrical meter box made of light highstrength aircraft plastic, 30 cm in diameter and 30cm tall. The outer oven is cylindrical aluminum 22 cm in diameter. The black lid is designed to be water resistant.
Electronic levels and CPI are required for this option. The microscope viewer of beam position and the spirit levels are retained within the outer oven for use during factory servicing, but are not visible during normal use. A special baseplate is available with larger diameter, but with three feet in the same position as the standard baseplate.
Because of the careful selection of materials, the meter installed in the double oven option does not weight much more than the standard configuration.
There have been four major designs of the L and R charger/eliminator:
Early chargers were for NiCad batteries: These were provided as original equipment for meters up to G323 and D12 or about 1973. Almost all of these have been replaced by newer designs and the NiCad batteries replaced with gelatin stabilized leadacid batteries.
Next were the early gelcell battery charger. They are distinguishable by an amp meter, a red power indicator lamp and a single red LED lamp that indicated when the meter was receiving heating power. These chargers were built in 1973 and 1974 and many are still in service.
The more recent charger is similar in appearance to the above, except there are two LED's: one to indicate when the meter is receiving heating power and one to indicate when the battery is receiving more than a trickle charge. These were built between 1974 and 1989.
The present charger no longer has an amp meter. This was the most expensive component on earlier chargers and the part most frequently damaged by mechanical abuse. The amp meter has been replaced by an LED tenelement colored bar display.
When using any of the gelcell battery chargers, if the AC power source is intermittent, the meter will not go off heat as long as the charger has both a battery and the meter connected to it. When AC power is lost, a relay changes position. It disconnects the battery from the charging circuit and connects it to the gravity meter. When AC power returns, the relay returns the battery to its original charging position and the gravity meter to the regulated 12volt power.
The chargers are designed to operate from 115 or 230 volts ±10% from 47 to 420 Hertz AC power. The 115/230 switch is on the face of the one and two LED chargers. It is internal on the new tenelement LED charger. The charger may be damaged if it is set for 115 volts and connected to 230 volts. The charger will not function properly when set for 230 volts and connected to 115 volts.
The circuit breakers within the chargers can be tripped by a sharp blow as well as excess current. A blow may occur during transportation. If the charger is not working after shipment, first check the breakers. There are two breakers, one on the incoming AC power line and one on the outgoing charger line. Both are 1.5 amp rating. To reset the breakers:
If the receptacle for the bottom screws are loose, they may be tightened by pressing the aluminum together a small amount with pliers.
Chargers before 1987 had a separate battery charging cable with a round Amphenol 3pin fitting at the charger end of the cable. When disconnecting a battery from the charger, this cable always should be disconnected at the battery end first. If the other end of the cable is disconnected first, there are two exposed pins with unfused 12volt power across them. Severe damage could occur if these were shorted by brushing across a metal surface.
Chargers built after 1986 (with nondetachable charging cable) have a small modification which provides smoother power for those meters using the electrostatic nulling options when their gravity meter is plugged into the charger/eliminator. Earlier charges that have had this modification can be identified by opening and examining the PC board. The modification can be identified by two 0.75 ohm resistors in parallel located close to the relay, or by the small XREG label on the bottom of the charger housing.
The high speed crank can expedite gravity reading when there are large gravity differences between gravity stations. It is a small cylindrical gear box with a ratio of 9.5 to 1. There is a small crank at the top for turning while the housing is held by two fingers. At the bottom is a yoke with two pins that fit into matching holes in the top of the meter's nulling dial.
To use the crank simply place it on the top of the nulling dial, matching the two holes in the dial and the two short pins on the gear box.
We strongly urge the operator to use discretion in the rate of speed the dial is turned as it is possible that the life of the counter may be shortened if the dial is rotated too fast. Also, caution should be exercised when approaching the end of travel of the counter (0000.0 and 7000.0 for the Model G and 0000.0 and 2000.0 for the Model D).
The operator will prefer to use the device to bring the counter to an approximate gravity reading, then lift it off the dial and complete the final gravity reading in the usual manner.
The device is a precision product with high quality gears and bearings throughout. Care should be taken to protect it from dust and dirt as much as possible and to handle it carefully. Its usefulness can be decreased by dropping it and bending the shafts or housing.
In the early days of earthquake seismology, long period horizontal motions could be measured with the horizontal pendulum seismograph. As the axis of rotation became closer to vertical, the period became longer. Theoretically, if the axis is vertical, the period is infinite.
Dr. Romberg posed the question to his student, Lucien LaCoste, how to design a vertical seismograph with the characteristics as good as the existing horizontal pendulum seismograph.
In the illustrated suspension, there are two torques: gravitational and spring. If these two torques balance each other for any angle of the beam, the system will have infinite period. The smallest change in vertical acceleration (or gravity) will cause a large movement.
The torque due to gravity is:
Where W is the mass and d is the distance from the mass to the beam's hinge.
The torque due to the spring is the product of the pull of the spring and the springs lever arm, s.
The length of the spring is r and by the law of sines:
If the spring constant is k and the length of the spring without force is n, The spring force is illustrated by this graph.
The torque due to the spring is then:
The total torque is:
This equation would yield zero torque and would be satisfied for all angles of q if:
For n to equal zero, we must have a "zero length spring". That is , a spring whose forcelength graph passes through the origin or, at least, points toward the origin. The turns of a helical spring of zero unstressed length would bump into each other before the spring actually reached zero length. By making a helical spring whose turns press against each other when there is no force on the spring, a "zero length spring" can be made.
There are several ways to make a zero length spring. A simple zerolength spring is a flat spiral spring. The mechanical properties of a spiral spring are not as convenient as a helical spring.
To make a zerolength helical spring, the spring wire can be wound onto a mandrel. As the wire is wound, it can be twisted.
Another method is to hold the wire at an angle and with tension while winding it on a rotating mandrel.
Still another method is to "turn the spring inside out".
The actual spring used in the L and R meters are "negativelength". The spring wire is large enough and stiff enough that the spring would not act like an ideal spring if the spring were to be clamped at both ends. Thus, a very fine but strong wire is attached to the top end of the spring and another to the bottom of the spring. The top wire is clamped to the lever system and the bottom wire is clamped to the beam. The effective length of the spring is the combined length of the helical spring and the two fine wires. That combination is "zerolength". The helical spring by itself is "negativelength".
An important feature of the zero length spring suspension is its insensitivity to longitudinal and transverse vibrations (Harrison 1960, LaCoste 1967). Consider the spring to be made of identical masses with segments of weightless zero length spring between the masses. The top end of the spring is attached to A and the bottom to B. Since the spring segments are zero length springs, the forces each spring exerts on the adjacent masses are proportional to the spring length.
Therefore, if the masses are equally spaced vertically, the vertical component of force exerted on each mass will be zero regardless of its horizontal position or horizontal motion. Also, the vertical components of force are proportional to the vertical component of spring length. (The vertical component of the force vector remains the same.) Also, the vertical force on A and B will be independent of any horizontal motions.
LaCoste, L., A new type long period vertical seismograph, Physics, Vol. 5, pp 178180, July 1934
LaCoste, L., A simplification in the conditions for the zerolength spring seismograph, BullSeismological Soc. of Am., Vol 25, No. 2, April 1935
Harrison, J.C., The measurement of gravity at sea, Methods and Techniques of Modern Geophysics, Interscience Publishers, NY, 1960
LaCoste, L., Measurement of Gravity at Sea and in the Air, Reviews of Geophysics, pp 477526, Nov. 1967
Calibration of the L and R gravity meter is accomplished in two states:
First a relative calibration is obtained at numerous points over the entire range of the meter
Then the relative calibration data are adjusted to absolute values by field measurements over a known gravity difference
Considering the geometry of the micrometer screw, lever system and suspension, the interval factor curve should be a thirdorder curve. There will be minute deviations from the ideal thirdorder curve due to imperfections in the lever system, gear box, micrometer screw and the jewel and pivot where the micrometer screw transfers its movement to the lever system.
To obtain the relative calibration curve, a fixture is attached to the beam mass.
One part allows a calibration mass to be added and removed from the beam.
Another part is a long thin screw with a small adjustable nut that allows the meter to be nulled in the lab over the full travel of the micrometer screw.
First, let us consider the Model G meter with manual observation of the relative calibration data. If the calibration mass corresponds to a change of about 200 milligals and the meter can be read to about 0.01 milligal, then the accuracy of obtaining a relative factor is about one part in 20,000.
The relative calibration is obtained as follows; first the counter is set to 0100.00 then:
In 1994 / 95 the above process was computer automated. It allowed repeated runs over the range of the meter, greater accuracy, automated plotting and automated generation of the calibration table.
To adjust the relative factors to absolute factors, the meter must be read at two or more gravity stations of known gravity difference. A single factor is determined. When the relative factors are multiplied by this single conversion factor, the final calibration table can be constructed.
The greater the gravity difference between the calibration stations, the greater would be the accuracy of the conversion factor. Since the relative factors were determined by the manual lab procedure with an accuracy of about one part in 20,000 or 22,000, then it would be appropriate to have similar accuracy in the determination of the conversion factor.
The meter reading can be resolved easily to 0.01 milligal. Therefore a calibration range of 200 or 220 milligals would be appropriate for accuracy consistent with the relative determinations.
The high and low calibration stations should be close to each other to allow the observations to be repeated several times for greater accuracy and a further field check of the meters repeatability under field conditions.
There is little elevation relief and only small gravity changes in the vicinity of the L and R laboratory. The closest large change in gravity is 1,000 kilometers to the west, Cloudcroft in the state of New Mexico. Two stations were established accurately. Their gravity difference is approximately 240 milligals.
The field procedure consists of alternately reading the meter at one station then the other station until at least four readings are taken at the top station and four at the bottom station.
The observations at both stations are plotted against times on the same graph. Typical scales would be 0.05 counter unit per centimeter (or 0.10 units per inch) and one hour per two centimeters (or one inch). Two smooth curves are drawn through the upper and lower station data points. The second curve is constructed parallel to the first.
The separation between the two curves is the average counter reading difference D C
If the gravity difference between the two calibration stations is D G, then the absolute calibration factor over that range of the meter is:
The calibration table for the Model G will consist of 70 calibration factors, one for each 100 counter units. The 70 relative factors determined earlier are multiplied by the absolute calibration factor; Fabs. The calibration factor to use for counter readings between 0000.00 and 0100.00 is F50 x Fabs and the calibration factor for the readings between 0100.00 and 0200.00 is F150 x Fabs, etc. If F50, F150 and F250 are 1.02031, 1.02038 and 1.02045 respectively and
Fabs is 1.00321, then the first portion of the final table would be:
Reading |
Reading |
Reading |
|
|
|
|
|
|
|
|
|
|
See Converting the Counter Reading to Milligals in the first section of this manual.
The standard Model D has a counter range of 2000.00 and utilizes about 62% of the full range of the lever and micrometer system. This is a small enough portion of the thirdorder interval factor curve that, for most work, it can be approximated by a straight line. By minor adjustment within the meter, the slope of that line can be made almost horizontal. Thus a single calibration factor can be used for the full range of the meter.
Many of the meters between D1 and D109 have been assigned a single calibration factor. The meters were constructed and the factors determined at three points; one near 0100.00, one near mid range and one near 1900.00. These were determined at a field calibration range of 17 milligals located near the L and R laboratory. The meter was adjusted the approximate amount necessary to make the factor curve horizontal and returned to the calibration range. If the repeated tests at 0100.00, 1000.00 and 1900.00 indicated the calibration curve very close to horizontal, factors were also determined for counter readings near 0500.00 and 1500.00. If these also plotted on the horizontal line, a single factor was used for the meter.
On occasion, it was difficult to obtain a sufficiently straight line over the range of the counter or to adjust the meter so the interval factor line was horizontal. For those meters, a calibration table was prepared using a smooth curve through the five data points obtained at the Austin 17milligal calibration range.
Starting with meter D110, the laboratory calibration was used to determine relative calibration factors, then the meters were field calibrated over the 240 milligal range to determine the factor necessary to convert the relative factor to absolute factors. A calibration mass equivalent to about 100 milligals was used and five points were measured to determine the calibration curve. Since the meters did not have enough range to be read over the 240 milligal field range, a temporary counter was placed on the meters to allow them to read a little below 0000.00 and a little above 2000.00.
Though this technique was an improvement of the calibration using the local 17milligal field range, there were two disadvantages: the 100 milligal calibration mass covered too large a portion of the curve and could smooth some of the curvature and the special counter temporarily required for reading over the 240milligal range was an inconvenience.
Beginning in 1989 (meter D145) the laboratory calibrations use a 20 milligal calibration mass so the finer structures of the relative curve can be observed.
Also, a new field station was installed along the 240milligal range. It is approximately 160 milligals greater than the top calibration station at Cloudcroft. The standard counter can be used during field calibration over the 160 milligal range.
There are two important factors in the selection of the calibration mass
The three most likely periodic errors in the meters are once, twice and four times per revolution of the micrometer screw. Selecting a calibration mass of approximately 220 milligals for the Model G gives three revolutions of the micrometer screw between the OFF and ON of the mass. Selecting a calibration mass of approximately 20 milligals for the Model D gives approximately six revolutions between the OFF and ON of the mass.
We know that the 240milligal Cloudcroft calibration range value is slightly low. When the L and R meters are used on calibration ranges determined by modern absolute gravimeters, the values determined by the Model G meters are usually low by a few parts in 10,000. We hope to have an absolute meter observe gravity at the Cloudcroft range in the future.
The LaCoste and Romberg land gravity meter is based on the zerolength spring suspension. It allows an instrument to be very sensitive to small changes in gravity yet be very compact in size. The size of the basic element is a cube about 5 centimeters on each side. The concept of the zerolength spring was developed by Lucien LaCoste as a graduate student at the University of Texas in 1932. He and his faculty advisor, Dr. Arnold Romberg utilized the concept first to construct longperiod vertical seismographs. Their instrument exceeded the period of existing instruments by an order of magnitude (LaCoste 1934).
While teaching at the University of Texas, Drs. LaCoste and Romberg developed the first two gravity meters during 1937 and 1938. The first weighed about 125 pounds (57 kilos) and the second about 100 pounds (45 kilos). Following spring semester in 1939, they took leave of absence from the university to develop and build gravity meters on a fulltime basis.
The first production meters weighed about 75 pounds (34 kilos). About 45 were produced from 1939 through 1941. These meters were not barometrically sealed but were compensated for buoyancy effects of changes in air pressure. Their range was 100 milligals and they could be reset to work anywhere on the surface of the earth.
Next came the "25pound" meters (11 kilos). They had a range of 200 milligals and about 80 were built between 1941 and 1957. There are a few of these that are still in service.
The geodetic meter was first built in 1957. These meters had a worldwide range without resetting. Eight were constructed using the 25pound meter size and several of its parts. Their range was 6,000 milligals.
The meter was miniaturized and several improvements made. Thus the Model G meter was born in 1959 . Since there were a few places atop the Andes not reachable by a 6,000 milligal range, the meters were designed to have a 7,000 milligal range. Their weight was about 8 pounds (3.6 kilos). They were mounted in white fiberglass boxes with rigid foam insulation and their power consumption was reduced. They were sealed so barometric pressure changes could not effect the buoyancy of the mass. By the end of 1990, over 980 Model G meters were constructed and almost all are still in service.
The Model D meter uses many parts in common with the Model G: the same spring, beam and mass. However, there are two screws for nulling the meter: a coarse reset screw with worldwide range and a fine reading screw with a 200 milligal range. The first Model D was built in 1968 and over 170 were constructed by the end of 1990.
Both the Model G and Model D meters have undergone gradual evolution in design details. As sources of error or potential difficulties were recognized, solutions were gradually developed and implemented. Several changes are under way at this time. Recently linear electrostatic feedback nulling has been added as an option. Model D meters are available with both fine and coarse screws calibrated, allowing greater accuracy in geodetic work. Development is under way for an internal microprocessor to work in conjunction with electronic levels, shaft encoder, calibration factors in memory, temperature sensor, clock and a data memory.
The L and R gravity meter has gone through some major changes and numerous minor changes during its evolution. Dr. LaCoste is the inventor and source of most of the important ideas that make the meter work so well. His partner, Dr. Romberg, was a resourceful contributor that worked at the firm till he was 85 years old.
Carroll Montigue's careful work on micrometer screws has been crucial and the patience and care of the several meter builders has been essential.
