|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
In a recent paper on the intermittent injection of gas in gas-liftoperations as opposed to continuous injection, Morgan Walker presented acomparative table showing the effect on oil and gas production from the samewells under the two methods of operation. This table carried a column showingthe formational gas...oil ratios under each method, computed by deducting thevolume of input gas from the volume of trap gas.
During the entire spring and summer of 1928, the present writer had beenengaged on some extensive test car tests of rich gas in Glenpool, seeking,primarily, for an explanation of the apparent loss of gasoline during thesummer season between the field meters and the plant master meter. Tests atseveral of the field meters where the richest gas was obtained regularly failedto check with other tests on the same gas made farther along the line, and thedifference was not made up by the quantity of drip gasoline collected betweenthe points. In the course of these tests, the final outcome of which isimmaterial to the present purpose, an attempt was made to construct a curverepresenting the relationship between gasoline content of the gas in gallonsper thousand cubic feet and the shrinkage in the volume of gas treated as aresult of removing the condensible fractions. Such a curve was made, expressedin units of gallons per thousand cubic feet on the abcissa and the ratio vaporvolume to liquid volume on the ordinate. This curve is a rectangular hyperbolaand shows that 3 gal. of gasoline taken from 1000 cu. ft. of raw gas as a vaporoccupied much less space, per gallon, than did 1 gal. of gasoline taken from1000 cu. ft. of another sample of gas. This is, in general, corroborated byconverting to the same units the residue settlement curves prepared by theTidal Oil Co. and by the Natural Gasoline Manufacturers Assn., though the threecurves, when plotted together, occupy entirely different positions on thepaper, as shown in Fig. 1.
Early in the development of every oil field, the operators of relativelylarge tracts of land must decide upon a spacing plan for the wells to bedrilled upon their respective tracts. Usually it is desirable to have a systemof regular spacing laid out over the entire tract, making such adjustments asmay be necessary near land boundaries.
Unless the dip of the producing formation is very steep, the equilateraltriangular arrangement is the one most generally preferred, for the very goodreason that it gives the maximum amount of drainage area for the wells up tothe limit at which mutual drainage interference begins.
Assuming that the equilateral triangular pattern is to be used, there stillremains to be decided the matter of distance between wells. Many variablefactors enter into the problem of proper spacing, some of which are physicaland some economic. It is not the purpose of this paper to discuss thesefactors, as that has been done in several comprehensive articles during thepast few years. In the end the decision must be based upon an estimate of therelative importance of various factors which apply to the field in question andto economic conditions in the oil industry at the time the development isgetting under way. Even if the best possible spacing distance should be chosenwhen the problem arises, changing economic conditions and variation in physicalconditions in different parts of the field may cause the original choice tobecome more or less unsatisfactory.
Obviously, it is advantageous, once a spacing plan has been adopted, to avoidalteration of it as the field becomes better known or as conditions in theindustry change, and yet be able to vary somewhat the acreage per well inextending the area of development. Or, it may be desired to provide for anultimate spacing considerably closer than conditions seem to warrant early inthe development of the field. With these contingencies in mind, it is ofinterest to examine the possibilities which the equilateral triangular systemaffords.
Orientation Of Coordinate System
In this system the well locations may be considered as the inersectionst ofthree sets of coordinate lines, the lines of each set intersecting those of theother two sets at 60?.
When a well strikes an oil-bearing layer, the oil has a pressure which isgenerally sufficient to enable it to rise to near the surface (sometimes abovethe surface). As soon as a well begins to produce, however, the liquid movesthrough the pores of the reservoir bed and the pressure in the well becomesmuch lower than the pressure originally prevailing there. At some distance fromthe well, however, the pressure in the reservoir bed remains unaltered; thusthe pressure of the oil has not only to lift the oil, but also to overcome thefriction resistance in the pores. The fact that so many oil wells are gushersis a consequence of the energy accumulated in the gas.
In gushing the well acts as a gas-lift. A mixture of liquid and gas (the latterpartly dissolved in the former) rises vertically from the oil-bearing layerthrough a cylindrical casing to the surface. In time conditions alter and thewell ceases to gush regularly, then the gushing can be further promoted byinserting a narrower tube in the well and connecting the top of the oil stringto the tubing. If the action in time becomes irregular, the gushing can be keptup for a further period by forcing gas between the two tubes. In the oil fieldsthe term" gas-lift" is used actually only where extraneous gas isapplied, as in the last of the stages mentioned. The action, however, is justthe same whether the gas exclusively originates from the formation, or ispartly applied artificially. Thus by gas-lift we simply mean a vertical tube inwhich the energy of gas under pressure, and of dissolved gas, is utilized forraising a liquid.
In gushing oil wells the pressure is frequently very high and the absorptioncoefficient 0.4 (expressed in vol. ratio) of the coexisting gas is notparticularly high, so that in reality it should be assumed that a considerableportion of the gas, at any rate at the bottom of the gas-lift, is dissolved inthe oil. For water-producing wells this is not usually of such importance.
It is recognized that in the early days of the petroleum industry oil wasproduced with practically no scientific or fundamental knowledge of the lawsand principles governing its extraction from the ground. Only a few, if any, ofthose exploiting the oil resources made any effort to collect accuratescientific information. There was little need for it, for as is frequently thecase where the supply of a natural resource appears inexhaustible and isgreater than the demand, large profits were possible from merely skimming thesurface with little regard to the efficiency or thoroughness with which the oilwas obtained. With this policy in force, great quantities of gas were wastedand when the reservoirs ceased producing by natural means they still containeda large fraction of the oil originally in them.
The present tendency in the industry is toward conserving existing oilreserves. In order that greater percentages of the oil present in the groundmay be recovered, efforts are being made to improve production methods and torework depleted fields. If the production methods are to be improvedintelligently it is essential that the fundamental laws be known andunderstood, consequently large amounts of money and effort are being spent ingathering data that will serve to define and interpret these laws.
A great deal of attention is being focussed on the study of the motion of oilthrough an oil-bearing sand, the forces that cause it to move and the factorsthat affect its motion. It is recognized that one of the important factors ispressure, but its quantitative effect is not known. In a given field the rateof production declines with time and so does the rock pressure.
The spacing of wells in Long Beach oil field has caused much discussion fromthe earliest days of its development, on account of the closely drilledtown-lot areas which have been as intensively developed as any productive areasin the world. The purpose of this paper is to present the results obtained fromthis type of development in comparison with the more widely spaced developmentin this field, now that a sufficient length of productive life has elapsed. Theintensity of development in Long Beach is indicated by the fact that up toSept. 1, 1929, there were 1736 wells either drilled or drilling. Of these 165failed to get production and 121 were drilling. Thus 1450 wells have producedoil from an area of 1350 acres. Up to July 1, 1929, these wells produced atotal of 355,047,913 barrels.
Four groups of offset areas were chosen, in each group the wells in one areaare widely spaced and in the other, closely spaced. The areas in each group arecontiguous and the groups are spread over the field from the southeast end tothe northwest, so that representative data are assured. The comparison has alsobeen extended to shallow and deep zones, like zones in their respective areasbeing compared.
Even though these productive zones are very thick, a fair picture of results isobtained, because the drilling of the wells in the two types of areas wasconcurrent in the different groups and, as a rule, the thickness of sand openedup at anyone time was about the same. As deeper sands were found, wells inoffset areas were deepened or new ones were drilled in both types of areas. Thedivision of zones was made in the same manner in contiguous areas.
The total cumulative production up to July 1, 1929, of the wells in theirrespective zones in each area was obtained. From these figures were computedthe recovery per well, recovery per acre, acreage per well, and the ratios ofthese factors. Thus July 1, 1929, is the end of the period considered in allareas.
Production engineering has continued its rapid progress during the pastyear. Many engineering efficiencies long practiced in other industries arebeing rapidly accepted by the oil industry and every branch of development andproduction work is undergoing study that is reducing the time and hazard in thedrilling of wells even for greater depths, is securing better service fromequipment, and is obtaining greater ultimate recovery from thereservoirs.
Deviation of Drill Holes
Deviation of drill holes probably has attracted greater and wider study thanany other subject. Until about one year ago it had been given little attention.Some California operators knew the common magnitude of deviation, butMid-Continent operators hesitated to believe that holes could be drilled socrooked. The great emphasis and publicity given to the seriousness of theproblem resulted in the design of many instruments to record the deviation bychart or photograph. Instruments which would give the direction as well asdeviation were not generally favored because of the grave fears of the legalcomplications which might result if the bottom of a producing well should proveto be on a neighboring lease. With all the instruments available fordetermining deviation, the use of etching on a glass bottle by hydrofluoricacid is still the most common method employed.
The beneficial result of this earnest effort to drill straighter holes isproved by a comparison of deviation of wells completed recently with thosecompleted over a year ago. The time of this change is arbitrarily taken as Jan.1, 1929, as at that time the operators were aware of the deviation whichusually occurred, and most of them were making some effort to reduce it. Table1 shows averages of wells completed by several companies.
Natural gas influences the movement of oil throughreservoir rock by affecting the physical properties of the oil and the pressurewithin the reservoir. The presence of gas bubbles changes the laws of flow andthe distribution of forces.
The problem of flow of oil and gas through a porousrock is complex, and most authors, in order to analyze their problems, havefound it necessary to make or accept certain generalized assumptions. Some ofthese assumptions, particularly the ones dealing with capillary phenomena andmolecular forces, have not received sufficient attention and by lack ofunderstanding have often been misinterpreted.
Herold has called attention to the resistance offeredby gas bubbles in capillary spaces and has drawn some interesting conclusionswith regard to the action of natural gas in a reservoir rock. He has duplicatedan experiment of the French physicist Jamin, has interpreted his findings interms of molar mechanics, and, from experimental observations, has made anumber of assumptions which he later applies in discussing problems of naturalflow and recovery of oil. The Jamin action is an assumption because it lacks aphysical proof and has never been determined quantitatively.
Tickell attempts to develop a formula expressing thework performed by a distorted bubble. H. A. Wilson, in the recent publicationof the American Petroleum Institute on the function of natural gas, discussesthe phenomenon in terms of classical physics. Both contributions are incompleteand offer no experimental data.
This paper includes a study of the static condition ofequilibrium of gas and liquid bubbles confined to capillary spaces. In order toavoid any possible confusion, it is proposed to restrict the term" Jaminaction"
Law of Flow for the Passage of a Gas-free Liquid through a Spherical-grainSand
The flow of a gas-free liquid through a spherical-grain sand has beeninvestigated by Slichter. By theoretical considerations involving a ratherlarge number of approximations he arrives at the following flow-formula:
v = CPd2At
where V is the volume of liquid delivered in time t through a cylindricalcolumn of sand of length L and cross-sectional area A by a pressure-drop P whenthe sand grains of diameter d are packed to a porosity m and the coefficient ofviscosity of the liquid is n. C and B are constants which were determined fromgeometrical relations; B was in reality a function of m. Values of B(l - m) forvarious values of m were given in a table.
In the same Annual Report of the U. S. Geological Survey, King reported theresults of his experiments on the flow of water through porous media. He foundthat there was a departure of about 86 per cent. from a linear relationshipbetween rate of flow and pressure, for a specimen of Dunnville sandstone forpressure drops not exceeding 60 cm. of mercury. The specimen was approximately4 cm. in diameter and 5.1 cm. in length. Over the same range of pressure hefound a departure of 45 per cent. for a specimen of Madison sandstone. Forunconsolidated sands he found departures as great as 49 per cent. for pressuredrops of 70 cm. of mercury; for pressure drops not exceeding a few inches ofwater he obtained a close approximation to a linear relation.
*This paper contains results obtained in an investigation on The Effect ofNatural Gas upon the Viscosity, Surface Tension, Adhesion and GeneralExtractability of Petroleum of Various Types, listed as Project No. 33 ofAmerican Petroleum Institute Research. Financial assistance in this work hasbeen received from a research fund of the American Petroleum Institute donatedby John D. Rockefeller. This fund is being administered by the Institute withthe cooperation of the Central Petroleum Committee of the National ResearchCouncil. Prof. H. C. George is Director of Project No. 33.
The Petroleum Experiment Station of the U. S. Bureau of Mines atBartlesville, Okla., has for the past three and a half years maintained alaboratory with the necessary personnel for conducting research on methods ofincreasing the recovery of oil. Some of the preliminary problems encountered instarting this work, with the data obtained, were discussed in A. 1. M. E.Technical Publication No. 144, "Oil Recovery Investigations of thePetroleum Experiment Station of the U. S. Bureau of Mines," which waspresented at the October meeting of the Institute at Tulsa, Okla., in1928.
Scope of Paper
The following discussion gives some of the results of the more recent work atthe Oil Recovery Laboratory. These experiments were conducted to obtainlaboratory data pertaining to the relative merits or efficiencies of variouspressure media in the recovery of oil from a sand reservoir. Two general typesof experiments were conducted. In one the various pressure media were passedthrough an artificial body of sand partly saturated with oil at a constantinput pressure; in the other the pressure media were passed through the body ofsand partly saturated with oil at a constant volume rate. No attempt will bemade to discuss the ramifications of economic problems which enter into thechoice of a pressure medium.
Apparatus and Materials
The two 6-in. by 6-ft. flow tubes described in Technical Publication 144 wereused in these experiments. They consist of 6-ft. sections of 6-in. casing witha blind flange at each end.
Part I Of Final Report Of A. P. 1. Project No. 33
The data and information compiled under Part I of this report are the resultsof experiments performed in the petroleum engineering laboratory under thesupervision of W. F. Cloud, Associate Professor of Petroleum Engineering andWilliam Schriever, Professor of Physics, University of Oklahoma.
Problems Being Studied
Most of the time has been spent in trying to determine the rate of flow andpressure gradient in a 4-in. flow tube 10 ft. long, which was packed with sandof various grain sizes. To date, only two kinds of sand have been packed inthis tube: 60 to 80-mesh Canadian river sand, 60 to 80-mesh and 80 to 100-meshSimpson sand. The original intention was to follow the same procedure and usethe same types of crude under identical saturation pressures of both air andgas, as well as unsaturated (dead) oil, flowing the various crudes throughseveral different sizes of both Canadian river and Simpson (Wilcox) sands, butlack of time has prevented such an intensive study of the problem.
Some additional time has been spent flowing saturated and unsaturated crudes,similar to those used in the 4-in. tube, through I-in. tubes, one of which was2 ft. long and one 5 ft. long. The results obtained have been checked againstthose obtained by using the 4-in. tube, to obtain the relation of diameter andlength of tube to rate of flow in similar sands.
Saturation Tank.-A heavy steel cylindrical saturation tank, capacity about 42gal. was mounted on a platform of bricks, then a galvanized tin temperaturebath was built around the saturation tank. This contained
* This paper contains results obtained in an investigation on The Effect ofNatural Gas on the Viscosity, Surface Tension, Adhesion and GeneralExtractability of Crude Oil, listed as Project No. 33 of American PetroleumInstitute Research. Financial assistance in this work has been received from aresearch fund of the American Petroleum Institute donated by John D.Rockefeller. This fund is being administered by the Institute with thecooperation of the Central Petroleum Committee of the National ResearchCouncil. Prof. H. C. George is Director of Project No. 33.