5. MATERIAL TESTING AND PRACTICE

5.1 Compression test
5.2 Slump test
5.3 Taking a sample of cement
5.4 Sand testing and practice
5.5 Water:cement ratio (weight)
5.6 Cement:sand ratio (weight)
5.7 Additional tests

Material testing and practice plays a vital part to the integrity of the material and the satisfaction of all parties concerned in the production of a sound ferrocement craft. However, there are really two groups of test. The first are those that fall into the normal requirements of everyday production and are generally capable of being carried out on site, ref. Section 6.1-6.6. The second type are the tests far more searching to satisfy the requirements of classification societies, and any detailed investigation one wants to make into a particular type of construction prior to undertaking the work, ref Section. 5.7.

For the purposes of this document, we will concentrate on the first group as the requirements of classification societies vary from one to another. In most cases the type of construction initially chosen will have sufficient design and material strength data available to satisfy one that the method and type of construction chosen is suitable for the vessel to be built.

5.1 Compression test

The compression test is carried out on either cube or cylindrical samples taken from a cross section of mixes during the casting of the ferrocement hull, and any other section of construction that is cast on a different day. The size of the cubes or cylinders will be laid down by the local testing authority.

Test cubes

The moulds for test cubes should be made of steel or cast iron, with the inner surfaces parallel to each other and machine faced. Timber moulds should not be used. Each mould should have a metal base plate with a true surface to support the mould and prevent leakage. It is essential to keep the mould and base plate clean and both should be oiled lightly to prevent the mortar sticking to the sides. No undue strain should be used when the sides are fixed together.

A 100 mm cube should be filled in three layers, from three different mixes. Each layer should be rammed at least 25 times with a steel bar 600 mm long and having a ramming face of 16 mm square, the weight of which complies to the local standard. The surface of the cube should be trowelled smooth.

A variety of institutions, eg universities, civil engineering, etc. will normally have the facilities to carry out the tests.

It is usual to produce six cubes at a casting and send two 100 mm cubes for testing at seven days, 28 days, and 96 days, after casting. The minimum results obtained, from experience, should be in the order of the following:

	lb/in2	kg/cm2
7 days	4500	315
28 days	5500	387
96 days	6000	422

In the UK, over many castings and using 150 mm cubes, the results exceeded:

	lb/in2	kg/cm2
7 days	6000	422
28 days	7500	527
96 days	9000	633

Classification societies will have their own procedure and minimum strength requirements to satisfy their rules.

Test specimens should be cured at not less than 10°C and in the same way as the hull is cured, and for the same period. Specimens should be transferred to the testing station on the seventh and twenty-eighth day, wrapped in damp hessian/gunny cloth or similar.

A test cube data sheet (ref. Fig. No. 7) should be kept as a record for that craft and as a build-up of accumulated strength data over successive castings.

Figure 7. Example of test cube data form

Casting date ................... Hull type ...................

Cube No	Date tested	Age at test	Weight of cube	Load at failure kg/cm2	Load at failure lb/in2	Comments on manufacture and testing
/7
/7
/28
/28
/96
/96

5.2 Slump test

The slump test is a practical means of measuring the consistency of mix. Since changes in the values of slump obtained indicate material changes in the water content or proportions of the mix. It is therefore useful in controlling the quality of the mortar produced.

The apparatus consists of a steel mould 100 mm diameter at the top, 200 mm at the bottom and 300 mm high, complete with a 16 mm dia. steel tamping rod 600 mm long and rounded on one end. (Local standards may vary the size of the equipment.) The inside of the mould should be clean before each test, and the mould placed on a hard flat surface. The mould should be filled in four layers, each layer rodded 25 times with the tamping rod. After the top layer has been rodded, the surface of the mortar is struck off level. Any leakage is cleaned away from the base of the mould and the mould is lifted vertically from the mortar.

The slump is the difference between the height of the mix before and after removal of the mould. If any specimen shears off laterally or collapses, the test should be repeated.

By using the correct mix and water:cement ratio prior to undertaking any casting, as a sample test(s) the average slump achieved from several tests will give the range of slump acceptable when the actual casting takes place. Because the mix is a mortar mix, the slump can be exaggerated by a very little increase in the water:cement ratio. Therefore, it is a handy guide but should not be an over-riding conclusion when the practicalities of the construction and need for full impregnation of the reinforcement are of priority during a casting.

5.3 Taking a sample of cement

If there is any doubt as to the quality of the cement purchased, a sample can be sent for test, providing the facilities are available.

A sample of cement taken for test purposes must be representative of the consignment and be taken within one week of delivery. It should be a mixture of at least 12 equal sub-samples taken from 12 separate bags in the consignment. The sample should weigh at least 7 kg and be sealed in an airtight container, with the relevant particulars clearly marked on the outside.

5.4 Sand testing and practice

Bulking of sand

When mixes are specified by volume, the sand is assumed to be dry. The volume of a given weight of sand, however, varies according to its moisture content. Equal weights of dry and inundated sand have practically the same volume but the same weight of sand in a damp condition can occupy a volume as much as 40% greater. This phenomenon is known as 'bulking'.

It may be demonstrated by filling a gauge box with dry sand. If the sand is flooded with water the level will sink a little, but not to any great extent. When the box is similarly filled with damp sand and the surface is flooded the drop in level will be very much greater.

Unless allowance is made for bulking when batching by volume, the mortar may contain too little sand. This is one of the reasons why measurement by weight is preferable. Bulking occurs far more with fine sands.

Testing for impurities

Sands are usually washed by the suppliers to remove clay, silt, and other impurities which, if present in excessive amounts, result in poor quality mortar. A guide to the amount of clay and silt in sand can be obtained from the field settling test. An excessive amount recorded in this test will indicate that other more sensitive tests should be made.

The test involves placing about 50 ml of a 1% solution of common salt in water (roughly one teaspoon per pint/0.57 litre) in a 250 ml measuring cylinder. Sand as received, is then added gradually until the level of the top of the sand is at the 100 ml mark and more solution is added to bring the liquid level to the 150 ml mark. The cylinder is shaken vigorously and the contents allowed to settle for about three hours. The thickness of the silt layer is measured and expressed as a percentage of the height of the sand below the silt layer.

The amount of clay and silt in the sand may be considered acceptable if it does not exceed 10%.

If a measuring cylinder is not available, a jam jar filled to a depth of 50 mm with sand and to a depth of 75 mm with the solution, will give a comparable result if the contents are allowed to settle for three hours. The thickness of the silt layout in this case should not be more than 3 mm.

A simple check for organic impurities is to fill a medicine bottle with sand as delivered, to the 115 ml mark, and then add a 3% solution of sodium hydroxide (caustic soda) in water, until the level of the liquid after shaking is 200 ml. A solution of this strength may be purchased from local chemists. The bottle is then stoppered, shaken vigorously, and allowed to stand for 24 hours. If at the end of that time the colour of the solution above the sand is darker than the standard colour shown in BS 812, or similar local standard, laboratory tests should be undertaken to determine whether the sand is acceptable.

Sand sieve analysis

The sand sieve analysis is carried out as often as is required to maintain the correct grading of sand that is to be used. The grading of a sand aggregate for ferrocement is found by passing a representative sample of dry sand through a series of BS sieves Nos. 7, 14, 25, 52, 100 (or local equivalent standard), starting with the largest sieve. A record should be maintained (see Fig. No. 8) of the result and be compared to the required acceptable envelope (Fig. No. 6). The envelope may vary slightly from the one shown depending on working rules, but is one used by the author for many years.

Figure 8. Example of sand sieve test form

Sample No ......... Sample obtained from .................

Total weight of sample .......... grammes

Sieve No.	Weight retained on each sieve grammes	Total weight passing sieve grammes	% Passing each sieve	Ideal %
7				100
14				68-96
25				35-65
52				10-36
100				2-10
Tray

Comments on test:-
______________________________________________________
______________________________________________________
______________________________________________________

Date of test:- ............. Signed:- ..............

If the sieving is done manually, each sieve is shaken separately over a clean tray for not less than two minutes. If machine sieving is applied, a nest of sieves should be shaken for at least 15 minutes. The material retained on each sieve, together with any material cleaned from the mesh is weighed and recorded. The percentage by weight passing each sieve is then calculated. Sieving will not be accurate if there is too much material left on any mesh after shaking.

5.5 Water:cement ratio (weight)

The two essential properties of hardened mortar are durability and strength. Both of these are closely related to density. In general, the more the mortar is compacted, the stronger and more durable it is. The mortar must be dense to be impervious to water and to protect the reinforcement adequately.

The strength and durability of the mortar is governed by the amount of water used for mixing but the overall grading of the aggregate has an indirect effect. Fine gradings require more water than coarser grading to obtain the same degree of workability. It follows that in practice the grading of the aggregate influences the amount of water which must be added.

It has been established that the strength of the mortar depends primarily on the relative proportions of water and cement. The higher the proportion of water, the weaker is the mortar. An allowance for the moisture present in the sand should always be taken into account. A typical list of water:cement ratios are set out below:

Imperial gallons per 112 lb cement	litres per 50 kg	water:cement ratio by weight
3.5	15.5	0.31
4.0	18.0	0.36
4.5	20.0	0.40
5.0	22.5	0.45
5.5	24.5	0.49
6.0	27.0	0.54

The water:cement ratio for ferrocement construction which gives the required strength and workability will be in the region of 0.35-0.45, although 0.40 should not be exceeded under tentative rules set by Det Norske Veritas.

If, due to local conditions, a more workable mix is required then admixtures may have to be used to enhance the workability subject to their suitability. (Refer item 4.8.)

5.6 Cement:sand ratio (weight)

The cement:sand ratio has been established to fall between 0.4 and 0.6, for dry sand. From experience the value should be nearer 0.6, although many boats have been constructed using 0.5 as the ratio.

5.7 Additional tests

Further additional tests will almost certainly be required if a boat is built to classification. Indeed, many of these tests can be used by the builder to enlarge the information on the particular form of construction that may be undertaken even if the structure is not being built to rules. Out of necessity one will need to have the use or availability of the correct testing facilities and specimens will need to be of a size to suit the testing equipment.

Tensile testing and compression testing on reinforced material

Tensile tests can be carried out on the ferrocement in order to establish the relationship between tensile stress and elongation.

Tensile test on unreinforced specimens

The tensile strength can be determined by a split 'cylinder' test using similar apparatus as in the compression test. However, the true tensile strength will lie between 50 and 70% of the splitting 'tensile strength'.

Bending tests with reinforced material

Bending tests can be carried out on the ferrocement in order to establish the relationship between bending moment, tensile bending stresses and elongation.

Bending fatigue tests

Bending fatigue tests can also be carried out on test specimens of the hull construction.

Impact test

An impact test can be performed on representative reinforced panels by using a drop weight. Failure occurs when the test panel develops a leak

All test specimens should have references to:

a) identification number
b) dimension of the specimen
c) curing history and moisture condition at test
d) defects of specimen, if any, and age
e) ambient conditions at time of test