State of FloridaDepartment of Environmental Protection
David B. Struhs, Secretary
Division of Resource Assessment and ManagementEdwin J. Conklin, Director
.\Florida Geological-Survey
WalteLScmnidl, State Geologistand Chief
~ Open.FiilYReport No. 88
¥J1%w('. ~
7i ."~",,.., ;..A -Mechanically Simple and Low Cost
SUbaqueouSSu,rface ~ediment Sampler.$:,.. 1. ",
~:.Ii'
Florida Geological SurveyTallahassee, Florida
2003
ISSN 1058·1391
CONTENTS
Page
ABSTRACT 1INTRODUCTION 1SAMPLING THE SEDIMENTATION UNIT 1SAMPLE SIZE AND SPLITTING 2MATERIAL SPECIFICATIONS, DIMENSIONS, AND CONSTRUCTION TIPS 2
Main Barrel 2Sliding Barrel Stop 3Sliding Barrel Bushing 5Sliding Barrel 5Bullet Points 5
BULLET SAMPLER OPERATION 6ECONOMIC ANALYSIS 7CONCLUSION 8ACKNOWLEDGEMENTS 8NOTE 8REFERENCES 8
LIST OF FIGURES
Figure 1. Basic components of bullet surface sediment sample showing first fivefoot segment (left) with fittings, second five-foot segment main barrelsegment (right) that threads onto the first segment for sampling In waterdepths of up to 10 feet, sliding barrel (top middle), and 50·gram and 100gram bullet points housing the sediment collection chamber (bottommiddle) 3
Figure 2. Schematic of dimensions of the bUllet sampler; dimensions are ininches 4
Figure 3. The author with a 10-foot length (two five-foot segments) of the bulletsurface sediment sampler (left) with an additional five-foot segment of mainbarrel (right) that can be threaded onto the top of the 10-foot length forsampling in a water depth of 15 feet 6
Figure 4. Detailed image of tip of the bullet surface sediment sampler showing thesliding barrel in position for sample collection (to be rotated in a clockwisedirection only), one of two sediment collection slots, and the 100-grambullet point sediment collection chamber. Upon extraction of the sampler,the sliding barrel slips down being stopped by the screw heads In the bulletpoint, thereby protecting the sample from being disturbed 6
Figure 5. Bullet surface sediment sampler in its carrying case with storage boxopen 7
A MECHANICALL Y SIMPLE AND LOW COSTSUBAQUEOUS SURFACE SEDIMENT SAMPLER
by
James H. Balsillie, P.G. 167Florida Geological Survey, 903 W. Tennessee St.. Tallahassee FL 32304-7700
ABSTRACT
Over .the yea~ the author has de,,:,elop~ a subaqueous surf~ce sediment sampler that is simpleto operate and mexpenslVe to construct. It IS desIgned to be operated In water ranging from wading depthsto a. water ~epth up to 20 f~et when operated from a boat. This paper describes (1) sampling rational of thesedlmentatton umt for which the device has been designed. (2) sample size constraints for which thesamp/~r has been configured, (3) sampler specifications, dimensions and construction tips, and (4) sampleroperation.
INTRODUCTION
In 2000, the author completed thedesign and construction of a surface sedimentsampling device that can be simply operated inwaist- to chest-deep wading depths or from aboat. Its origin grew from the desire of nothaving to dive for surf zone bottom sedimentsduring the winter when water temperaturesattain a highly uncomfortable mid-50's degreesFahrenheit. The sampling device is currentlydesigned to allow collection of unconsolidatedsediments from wading depths or from a boatup to a water depth of 20 feet. Samplinglogistics are eased from a water-depthperspective, since the device is comprised offive-foot threaded sections.
The device has specific merits. First. itcan ameliorate or mitigate the watertemperature problem for the sampling individualwhere total immersion would be required.Second, it constitutes a significantly quickermethod requiring fewer personnel for obtaininga sample than using divers. Third, it allows asample to be obtained from a boat in water notconducive to diving due to potential pollutionproblems. Fourth, it much more nearly samplesa sedimentation unit. Fifth. it procures asample of a size specifically suited to sievinganalytical procedures.
The purpose of this paper is todescribe the sampling device with thefollowing specific goals: (1) what it is capableof sampling, (2) design details anddimensions to assist the interested researcherin its construction, and (3) how it is operated.
It is standard practice in scientificwork to express length in S.l. units(International Systeme d'unites). However, inthe United States, a great many commerciallyavailable products are specified only in BritishImperial units. In fact, construction of thedevice described herein could not beaccomplished using S.l. units. Hence, thiswork constitutes the rare case where mixedunits are employed to insure clarity andprecision. Mass (weight) is reported in grams.
SAMPLING THESEDIMENTATION UNIT
Central to the design of a samplingdevice is the question: what is to besampled? An underlying assumption withsedimentologic studies is that the desired fieldsample is a lamina sample (Balsillie, 1995).This is the sedimentation unit of Otto (1938,p. 575) defined as ... that thickness ofsediment which was deposited underessentially constant physical conditions.Apfel (1938, p. 67) used the terminologyphase sampling in which a phase is defined
as ... deposition during a single fluctuation inthe competency of the transporting agency (seealso the work of Jopling, 1964).
Hence, the sedimentation unit orphase sample represents a narrowly definedevent. For example, it is not deposited by aflood occurring over a period of several weeks,but it might be deposited by one energy pulseoccurring over-and-over during the event. Justwhat a sedimentation unit, lamina, phasesample, or bedding plane is in terms of physicalprinciples, is not known. But, we do recognizethem to some general extent, and regardless ofthe unknowns one should strive to collectsedimentation unit samples (Balsillie, 1995).
Based on observations of beddingplane characteristics, it was decided that thesampling device should be designed to collect asample no more than 13/32 inches (onecentimeter) in depth. Moreover, the designpurpose of the device was to obtain a bedsurface sample.
SAMPLE SIZE AND SPLITTING
as 5% error per split, and recommended nomore than one split where sample archiving isof importance, say, for litigation purposes.
The bullet surface sediment sampleris designed to collect a sample of a sizesuitable for sieving analysis. No more than100 grams of sediment should be introducedto the sieve nest. A larger mass may result inovercrowding on one or more of the sieves(e.g., Carpenter and Dietz, 1950; Daescherand others, 1958; McManus, 1965; de Vries,1970, Jenke, 1973, Shergold, 1980: Socciand Tanner, 1980). A sample size of 45grams is ideal, but can vary from 35 to 50grams (Socci and Tanner, 1980; Balsillie,1995). The bullet sampler, then, has beendesigned to collect a 50-gram and a 100gram sample, bearing in mind that a certainpercentage of the sample will be fluid (salt,fresh, or polluted water).
MATERIAL SPECIFICATIONS,DIMENSIONS, AND CONSTRUCTION
TIPS
Main Barrel
The bullet sampler has five maincomponents: (1) main barrel, (2) slidingbarrel stop, (3) sliding barrel bushing, (4)sliding barrel, and (5) bullet points; see Figure1 for basic segments, and Figure 2 for designschematics. Material specifications anddimensions and construction instructions areas follows.
The main barrel is comprised oftwo-inch diameter schedule PVC well casingand bullet points with ASTM F-480 threads(i.e., two lA-inch threads per inch for bothmale and female threaded ends and an Dring for each male threaded end). The mainbarrel has an inside diameter (ID) of 1 9/16inches, and an outside diameter of 1 29/32inches. It is recommended that well casingsegments five feet in length be used for thesampler. Well casing can be obtained from awell drilling supplier.
Sample size and sample splittingwhere necessary are other importantconsiderations.
Upon occasion I have prevailed uponfriends and co~workers to secure a sand-sizedsediment sample or two from some exoticlocale to which they were traveling. Even withspecific instructions as to its size, I haveinvariably been given a quart-sized or even,upon occasion, a gallon-sized zip-lock bag filledwith sand. What these individuals fail to realizeis that handling of the sample bag andvibrations when transported in a vehicle willfurther sort a sample. Any sub-samplecollected from such a large sample may wellnot represent the original distributivecharacteristics of in-place natural material.Mechanical splitting procedures can be used,but they too introduce error (e.g., Wentworth,1926; Swineford and Swineford, 1946;Sengupta and Veenstra, 1968; Sanford andSwift, 1971; Emmerling and Tanner, 1974;Socci and Tanner, 1980; Balsillie, 1995).Emmerling and Tanner (1974) found as much
2
mainThe first five-footbarrel requires
segmentspecial
of thedesign
Figure 1. Basic components of bulletsurface sediment sampler showing firstfiv~foot segment (left) with fittings,second fiv~foot segment main barrelsegment (right) that threads onto thefirst segment for sampling in waterdepths 01 up to 10 leel. sliding barrel (lopmiddle). and 50-gram and 10o-grambullet points housing the sedimentcollection chamber (bottom middle).
considerations. Since the bullet points havemale threads. modifications are made to thefemale end of the first main barrel segment.The lineal length of the threaded sections forboth male and female threads is 1 Y2 inchesfrom the casing ends. Two slots approximately1 1A inches length are cut across the mainbarrel (Figure 1), one each on opposite sides ofthe main barrel, at a distance of from 1 lhinches to 1 19/32 inches from the female
3
thread end of the first casing length. Thisresults in a slot about t 3/32 inches (i.e., 1.0cm) in width. These slots are appropriatelybeveled to facilitate sediment samplecollection when the main barrel is rotated ina clockwise direction. A rubber stopper isinserted into the main barrel flush with the topof the slots. If snug enough, friction fit willsuffice. If not, it can be secured using epoxytwo-part glue. This insures that upon barrelremoval water draining from the main barrelwill not wash out the collected sample.
All main barrel segments areperiorated with Y2-inch holes at 1 ¥i-footdistances along the main barrel length (allfive-foot segments) to facilitate samplerimmersion and drainage of water when themain baTTel is removed from the water. Themain barrel is also labeled for distances inBritish Imperial or S.1. units as desiredbeginning at the top of the sediment collectionslots. Horizontal lines perpendicular to themain barrel axis are made using a finetriangular file to create shallow grooves whichare then filled with indelible black ink from afine felt tip marker. Adhesive-backednumerals can then be applied to appropriate,easily recognizable intervals so that waterdepths can be easily measured when thesampler is at its design collection depth.
The first length of the main barrel isalso fitted with the sliding barrel stop andsliding baTTel bushing whose descriptionsfollow.
Sliding Barrel Stop
The sliding barrel stop is a piece oftwo-inch PVC 1120 schedule (SHD) 40.ASTM D-178S pipe (I. D.: 2 3164 inches; O.D.: 2 3/8 inches) that is available from anyhardware retailer. A one- to two-inch piecewill suffice. Cut a vertical kerf with a hacksaw parallel w~h ~s length (Figure 2). Thiswill allow one to easily slide it up the mainbarrel to a position where it will precisely stopthe sliding barrel at the top of thesedimentation collection slots. The slidingbarrel stop is attached to the main barrel
FemaleThreaded
End
SLIDING BARREL BUSHING
Sediment Collection Slot
MAIN BARREL
(1-' 5-foot Section)2" PVC SCH 40 withASTM F-480 threads
SLIDING BARRELSTOP Depth
Marks
r-----------
Rubber Stopper Screws Dralna~eHole
o 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
SLIDING BARREL
EndView
Direction of Rotation
o-..
Bevel ends ofSEDIMENT
COLLECTIONSLOT
BULLET POINT(loa-gram)
0 4
Foot
Center FilledWith Epoxy Sediment
CollectionChamber
Area Drain Plate
151413121110987
2- PVC 1120SCH 40 ASTM 0·1785
6
Balsillie Bullet
5432
Area DrainScrews
1o
MaleThreaded
End
Stainless SteelScrew Stops
Figure 2. Schematic of dimensions of the bullet sampler; dimensions are In Inches.
first segment using PVC glue and fourstainless steel sheet metal screws.
Sliding Barrel Bushing
Main barrel well casing and two-inchPVC 1120 SCH 40 ASTM D-1785 used for thesliding barrel leaves a gap of about 1/16inches and a sloppy fit. I fitted a piece of clearplastic tubing with an 10 of 1 59/64 inches and00 of 2 1132 inches, 10 inches in length toserve as a bushing to reduce the gap. Theclear plastic tubing was found at a well drillingsupplier and patiently applied by tapping it withsome force into place. The fit in the bulletsampler was a friction fit. If one uses epoxyglue, only use it at the top 14 inch of the bushingand apply it to the main barrel only (i.e., not toboth surtaces).
Sand-sized particles in the bulletsampler tend to sometimes get lodged betweenthe bushing and sliding barrel, reducing theeffectiveness of the sliding barrel to protectthe sample from being washed out of thesample container. From time-to-time the slidingbarrel should be removed and both it and thebushing washed clean of any adheringsediment.
Sliding Barrel
The sliding barrel, 14 3/8 inches inlength, is constructed of 2.0-inch PVC 1120SCH 40 ASTM D-1785 tubing, and a 2 x 3 inchround PVC area drain with a foot diameter ofclose to 6 inches, both readily attainable at yourlocal hardware retailer. The PVC tubing mustextend entirely through the area drain and beflush with the 6-inch diameter surtace (Figures1 and 2). The plate of the drain can beremoved and a 2 3/8- inch diameter preciselycentered hole cut using a jig-saw; somesanding or filing using a rasp may be needed to"true-up" the fit. Do not use PVC glue to securethe two pieces, because it will set before thepieces can be properly aligned. Instead,position the pieces so that a flush fit is obtainedand at 1200 spacings secure the pieces using5/16-inch full thread stainless steel sheet metalscrews.
5
The sliding barrel serves twoimportant purposes. First, it collects a surtacesample by ensuring that the sampler remainsat the sedimentary bed surface due to the 6inch diameter area drain head. Second, uponextraction the barrel slides downward, tocover the sample collection slots so thatsample material will not be washed out of thebullet point sediment collection chamber.
Bullet Points
Bullet points house the samplecollection chamber. I purchased severalpoints 6 5/8 inches in length. They arecomposed of two units, a point and a malethreaded barrel (ASTM F-480 threads) joinedby rivets. Drill out the rivets to separate theparts. Next, fill the point inside the chamberwith epoxy and let it set and cure; this shouldtake about eight hours.
Now, the bullet points can beconstructed to collect specific samples sizes(i.e., masses). I designed two sample sizes, a50-gram sample and a 100-gram sample.The first is the optimal size for sieving, thesecond if a split is necessary forarchivaillitigation purposes. The depth of thesample collection chamber is made byshortening the barrel from the non-threadend. One will need to calibrate the barrellength required for sediment volumes to becollected. I use a plastic 35-mm film canisteras a guide. It holds close to slightly less than50 grams of dry medium-sized quartz sand.The researcher will have to decide, based onregional characteristics of average grain sizesto be sampled and their mineralogic contentand attendant mass densities, the appropriatesediment collection chamber size.Construction of two or more chamber sizes isrecommended. Also, the researcher mustinclude in volume collection considerations,the volume of fluid that will be part of thesample collected.
When points and barrels areassembled they are joined using PVC glueand two 1;2 in long stainless steel round headscrews. It is the screw heads which providethe lower stop for the sliding barrel.
Figure 3. The author with a 1G-foot length(two five-foot segments) of the bulletsurface sediment sampler (left) with anadditional five-foot segment of mainbarrel (right) that can be threaded ontothe top 01 the 10-Ioot length lor samplingin a water depth of 15 feet.
BULLET SAMPLER OPERATION
The sampler is deployed in the verticaldirection (Figure 3). Vertical pressure is applieduntil the sliding barrel is felt to make contactwith the sliding barrel stop (Figure 4). For
6
I~L: .
Figure 4. Detailed image of tip of thebullet surface sediment sampler showingthe sliding barrel in position for samplecollection (to be rotated in a clockwisedirection only), one of two sedimentcollection slots, and the 10o-gram bulletpoint sediment collection chamber. Uponextraction of the sampler, the slidingbarrel slips down being stopped by thescrew heads in the bullet point, therebyprotecting the sample from beingdisturbed.
sample collection the sampler main barrelmust be rotated in the clockwise directiononly. If not, one risks unscrewing the bulletpoint sample container, thereby losing both itand the sliding barrel. (Please note thatPVC is neutrally buoyant).
Upon extraction, the user should feelthe sliding barrel make contact with the stopscrews embedded in the bullet point. Ifcontact is not felt, exert a greater verticalamount of thrust. It is possible that sandsized material caught between the slidingbarrel and sliding barrel bushing isrendering operation inefficient. In either case,
Figure 5. Bullet surlace sediment sampler in its carrying case withstorage box open.
continue to extract the sampler, keeping itvertical. In a secure posture, i.e., so that thesliding barrel with not drop into the water,remove the bullet point and secure the samplein a suitable container. A narrow spatula maybe required to remove the sample. Next,remove the sliding barrel and wash both it andthe sliding barrel bushing clean. Feel thesliding barrel bushing from time-to-time. Ifburrs are felt, lightly sand it using wet-dry 180grit or finer sand paper until smooth.
In wading depths (waist- to chest-deepwater), the first five-foot length of the samplershould be sufficient for sample collection.When employed from a boat, vertically insertthe first two segments (Le., a 10-foot mainbarrel section comprised of two five-footsections). Thereafter, successively thread oneach additionally required five-foot sections.Following sample collection, any section greaterthan 10 feet in depth (or length) should beremoved while the sampler is in the verticalposition. Bear in mind that the PVC well casingwall is quite thin at the threads. A 10-foot
7
section is quite substantial unless mistreated.Any length beyond that may be vulnerable tofailure if handled in any but the verticaldirection.
ECONOMIC ANALYSIS
The author has, during the pastyear, donated the use of the Bullet Sampler intwo Florida Geological Survey projects. Todate 218 samples have been collected in oneproject during seven days (6 hours/day) offield work along Florida's Panhandle and BigBend coasts. Shallow water sampling indepths of 1.5 m (5 feet) or less can beaccomplished by free surtace diving.However, in greater water depths, particularlywhere currents are moderate or strong, scubadivers are required. Even if both shallow anddeeper sampling were to be anticipated in aproject, one should employ scuba divers toinsure continuity in the sampling effort. Localcommercial dive shops were contacted forquotes on diving rates. While they can varyconsiderably, an equitable, competitive, and
relatively low quote of $62.50 per hour wasfound.
Two activities sap scuba divers'strength: (1) working in a current, and (2)exiting the water and climbing back aboard thehost water craft. Four such episodes constitutethe diver's working day. Hence, for the workconducted 14, not seven, days of field workwould have been required. In addition, twodivers working together would be required tosatisfy safety standards. Using divers, the 218samples would have resulted in a commercialtotal value of $10,500 dollars or $48.00 persample.
During the seven field sampling days,the author operated the Bullet Sampler. Basedupon his approximate salary, the collection of218 samples during 42 hours of field workresults in a total cost of $1,050 or $4.80 persample. For an entry level employee operatingthe Bullet Sampler, the cost could easily be aslow as $2.40 per sample. Hence, the BulletSampler is far more cost effective at less than1/10 to 1120 of the standard cost, resulting incompleting the sampling task in half the time.In one instance, three samples were extracted15 feet from a sewage processing facility outfallin 16 feet of water, a task that divers would nothave been particularly happy to periorm.
CONCLUSION
This paper describes samplingrational, sampler specifications, dimensions andconstruction, and sampler operation for asurlace sediment sampler costing less than $70dollars (year 2000 expenditure levels) and amodicum of tools for its construction. Acarrying case is an additional concern notcovered in this work (Figure 5). The case wasconstructed of materials left over frompreceding wood working projects and extractedno other costs other than design, construction,and painting.
ACKNOWLEDGEMENTS
The author thanks Dr. William C.Parker, Department of Geological Sciences,Florida State University for his encouragement
8
that this paper be compiled and forconversations on the topic, and to Dr.Stephen A. Kish, Department of GeologicalSciences, Florida State University for thephoto images. The author also thanks hisFGS colleagues, Carol Armstrong, Jon Arthur,Paulette Bond, Ken Campbell, RonHoenstine, Jackie Lloyd, Guy H. Means,Frank Rupert, Walt Schmidt, and Thomas M.Scott, for their reviews and comments.
NOTE
The sampling device described inthis work was designed and constructed onthe author's free time external to any time asa State of Rorida employee. All materialcosts for its construction were incurred by theauthor.
REFERENCES
Apfel, E. T., 1938, Phase sampling ofsediments: Joumal of SedimentaryPetrology, v. 8, p. 67·78.
Balsillie, J. H., compiler, 1995, William F.Tanner on environmental clasticgranulometry: Florida GeologicalSurvey, Special Publication No. 40,144 p.
Carpenter, F. G., and Deitz, V. R., 1950,Methods of sieve analysis withparticular reference to bone char: U.S. National Bureau of StandardsJournal of Research, v. 45, p. 328346.
Daeschner, H. S., Siebert, E. E., and Peters,E. D., 1958, Application ofelectroformed precision micromeshsieves to the determination of particlesize distribution: Symposium onParticle Size Measurement, AmericanSociety on Testing and MaterialsSpecial Technical Publication No. 234,p.26-47.
de Vries, N., 1970, On the accuracy of bed-material sampling: Journal ofHydraUlic Research, v. 8, p. 523-534.
Emmerling, M., and Tanner, W. F., 1974,Splitting error in replicating sand sizeanalysis: Program Abstracts, GeologicalSociety of America, v. 6, p. 352.
Jenke, N. C., 1973, Sieve load equations andestimates of sample size: Journal ofSedimentary Research, v. 43, p. 518520.
Jopling, A. V., 1964, Interpreting the concept ofthe sedimentation unit: Journal ofSedimentary Petrology, v. 34, p. 165172.
McManus, D. A., 1965, A study of maximumload for small diameter screens:Joumal of Sedimentary Research, v. 35,p.792-796.
Otto, G. H., 1938, The sedimentation unit andits use in field sampling: Journal ofGeology, v. 46, p. 569-82.
Sanford, A. B., and Swift, D. J. P., 1971,Comparison of sieving and settlingtechniques for size analysis, using aBenthos rapid sand analyzer:Sedimentology, v. 17, p. 257-264.
Sengupta, S., and Veenstra, H. J., 1968, Onsieving and settling techniques for sandanalysis: Sedimentology, v. 11, p. 8398.
Shergold, L. Z., 1980, The effect of sieveloading on the results of sieve analysisof natural sands: Journal ofSedimentary Research, v. 65, p. 245249.
Socci, A., and Tanner, W. F., 1980, Little knownbut important papers on grain-sizeanalysis: Sedimentology, v. 27, p. 231232.
Swineford, A., and Swineford, F., 1946, Acomparison of three sieve shakers:Journal of Sedimentary Petrology, v. 16,p.3-13.
9
Wentworth, C. K., 1926, On mechanicalanalysis of sediments: University ofIowa Studies in Natural History, v. 2,52 p.
