PREVIOUS   NEXT   CONTENTS   HOME 

A simple machine for bulk processing sediments

1. Introduction

Anyone who has had the task of extracting fragile fish bones from clay or silts will be familiar with the problems involved. The commonly used procedure is to dry the sediment thoroughly and then to soak it in hot water. The disaggregated sludge is then sieved through an appropriate mesh. Most clays can be cracked in this way. The disadvantages are that:

An alternative method is to deep-freeze the matrix and then immerse it in hot water. This does not compare favourably with drying but will crack some silts and silty clays.

2. Adjuvants

There are a number of reagents that assist the disaggregation of sediments; these add to the expense of the technique, but can crack clays and shales that water alone will not affect.

1) Quaternary 0

This is a wetting agent used to disaggregate small samples of compacted clays. I have no experience of this, and use automatic-washing-machine powder as a joint detergent and water softener to help with sticky sediments. The use of Quaternary 0 is described by Zingula (1968).

2) Hydrogen peroxide

Used as a 20-60 volume solution, hydrogen peroxide rapidly breaks up clays and indurated silts, particularly those containing manganese salts. It also rapidly oxidises pyrite which in turn can corrode calcitic and phosphatic remains. To avoid this, it has been found that the addition of caustic soda (NaOH) to make a 5% solution is beneficial. This, however, destroys some plant remains.

3) Sodium hexametaphosphate

This compound, in solution, breaks up silts and clays, presumably by breaking up their colloidal state. I find it difficult to dissolve, expensive and relatively ineffective when compared with simple drying.

4) Kerosene (Paraffin)

Thoroughly dried clays and some indurated shales and sands crack very rapidly when soaked for 24 hours in kerosene and then washed in very hot, preferably boiling water. Most of the kerosene can be poured off the matrix before adding the water. The remainder can be salvaged later from above the water, thereby cutting the cost. It is probably the cheapest and most effective of all the methods listed here. Residue obtained by this method should not be treated with hydrogen peroxide.

5) Acid

Calcareous clays can be rapidly disaggregated by hot, dilute (2.5%) acetic or formic acid. Generally, little damage is done to the calcareous macrofauna. A variant of this method is to soak dried clays in a solution (10%) of washing soda (sodium carbonate) prior to treatment in acid. This is a harsh but effective method with clays and weak shales that break into pellets rather than totally disintegrate.

3. Bulk processing

None of these methods, however, solve the basic problem of getting a large amount of sludge through a sieve and extracting reasonably intact remains. This problem was partially solved by Henkel (1966) and Freudenthal (1976) who both used large static sieves and a power pump to bulk process large volumes of silt and clay in the field. This method can be very effective but is quite labour intensive. It is no faster with finer fractions or coarse silts and sands, but is less exhausting than hand sieving. The disadvantages of this technique include the bulk (weight) of the equipment and problems of transporting the pump, sieves, suction and delivery hoses et al. to the site. The main advantage is that the residue is prepared in the field and available for immediate sorting. It is probably most effective where a relatively coarse and bulky residue is being washed from a fine-grained sediment.

4. A new technique

Where it is possible to bulk sample the matrix and the results are not urgently needed, the following machine technique has been found most effective:

The machine (Figure 1) is in two parts; a polyethylene tank and a hinged lid assembly. The lid contains three lawn sprinklers, two oscillating and one rotary, both of which can be bypassed by a hose. The sides and part of the top of the lid are made of wood with a sloping, transparent perspex top. Three taps control the relative amounts of water to each sprinkler. The tank is rectangular and holds 350 litres when full. From the drain at the base a pipe is looped up to form an automatic syphon. By trial and error it was found that 25mm. (internal diameter) was suitable for the drainage pipe (30mm. external diameter). A larger pipe fails to syphon, merely overflowing with the tank in full position. A smaller bore pipe often fails to break the syphon when empty. As the bore of the drain pipe needed is determined by the local water pressure, this must be established experimentally. The drainage rate of the tank depends on both the drain pipe diameter and the length of pipe below the base of the tank. A drop of 80cms. causes the tank to empty one and a half times faster than it fills, thus its outflow rate is two and a half times faster than the inflow. A longer drop would be desirable. To syphon efficiently the drop must be as near vertical as possible and the end of the pipe must not be submerged.

Ward machine for bulk soil processing
Fig.1 A. Front, side and plan views of clay processing machine, B. Plumbing diagram for the lid assembly. Scale approx. x1/10

The problem of the silting up of waste pipes can be overcome by allowing the effluent to pass through a separate settling tank or by trapping in the washing tank by a modification shown in Figure 1A (side view). The sieve used is made from a plastic crate 75 x 25 x 17cm deep with 500 micron stainless steel mesh heat-welded on the inside. This is supported on blocks about 15cm from the base of the tank.

Most of the materials used in this machine are available from 'Do-It-Yourself' and hardware shops. Brass screws are used throughout to avoid corrosion. The plumbing is constructed from plastic hosepipe, standard hose connectors and 1/2" bore plastic taps. A hose clip holds the hose on to tap T1 where the water pressure is directed whilst the machine is in operation.

N.B. The mains water pressure rises considerably overnight, and during the weekends; however, the plumbing is inside the lid assembly, so a blown joint does not lead to flooding.

A rubber strip is used on the inside of the base of the lid to prevent spray from the rotating sprinklers escaping; both of these are standard lawn sprinklers. It is recommended that good quality ware is used, particularly in the case of the oscillating sprinkler. The stainless steel mesh can be purchased from:
Cadisch Precision Meshes Ltd.
Unit 1, Finchley Industrial Estate
879 High Road
North Finchley, London N12 8QA
Telephone: (0181) 492 0444
Fax: (0181) 492 0333
e-mail:
cadmesh@compuserve.com

Usually, several meshes will be available with the same aperture size, differing in the wire diameter and thus a percentage open area. A high percentage open area implies a thin wire diameter and a fragile mesh. The converse, a low open area, gives a robust sieve, but a slow flow rate. The author usually opts for a 45-55% open area. Nylon bolting mesh may be an appropriate and cheaper alternative.

5. Operating the processor

The sediment is stacked into the sieve and the sprinklers set in action. Tap 1 is closed, taps 2 and 4 open and tap 3 to the rotary sprinkler partially closed in order to create a heavy mist. In this mode the oscillating sprinklers wash backward and forward disaggregating, whilst the rotating sprinkler damps down the sieve contents and reduces any splatter. The machine is then left to its own devices. The tank will slowly fill with water, flood the base of the sieve and then empty. The equipment used by the author has a cycle time of about seven minutes. Progress can be followed through the perspex lid. Before opening the lid, tap 1 is opened and tap 2 closed. This diverts all the water flow through the hose to the bottom of the tank and allows the sieve contents to be inspected without the operator getting wet. The hose can be used to clean the sieve mesh and tank and to pile up the residue in one corner before emptying the sieve. The best results have been obtained by inclining the sieve a few centimetres towards the left and leaving the lowermost (left) ten centimetres free of matrix. This allows clean residue to accumulate and still be removed whilst a large batch of sediment is still being processed. Originally the author intended to use this machine to process dried sediments. However, it coped quite adequately with wet clay, but at a much slower rate. Dried silty clay can be processed at 10-15kg an hour, which compares with about 50kg an hour with a hand sieve. The advantage is that this machine operates 24 hours a day and does not demand coffee breaks. Wet clay is processed at about 1-2kg/hour. The slowness of the wet process is more than offset by the energy saved in not having to dry the clay and the fact that fossil seeds are recovered intact. Whether the sediment is processed wet or dry, it has been discovered that considerably fewer breakages occur with this method than with any other. The above description is of a system that has been evolved and the author does not expect it to remain static. However he would be very glad to hear of the experiences of others with this or similar systems.


 PREVIOUS   NEXT   CONTENTS   HOME 

Internet Archaeology
Last updated: Thu Aug 14 1997