What a soil mechanics laboratory needs when the ground outside is delta clay: the apparatus in the order the syllabus uses it, why demonstration rigs and testing apparatus are separate purchases, and the bench time consolidation quietly eats.
The ground under most Bangladeshi campuses is the reason the geotechnical laboratory matters more here than the equipment budget suggests. Soft alluvial clay, silt that liquefies when you look at it, a water table that comes up to within a metre or two of the surface for half the year: a student who has only tested a nice remoulded sand has not met the material they will actually be asked to build on. That has consequences for what the laboratory buys, in a direction most equipment catalogues do not point.
This is the room-level view. The department-wide split between soils, concrete, surveying and transportation is in equipping a civil engineering department room by room.
Almost every result a soil mechanics laboratory produces reduces to moisture content and dry density. Classification, compaction, consolidation, shear strength and permeability all rest on those two, and both depend entirely on two pieces of apparatus that rarely get argued about at procurement: the drying oven and the balance.
For the oven, the specification that bites is not the maximum temperature. It is how many moisture tins fit on a shelf, how many shelves there are, and whether uniformity across the chamber is good enough that a tin at the back dries at the same rate as one at the front. A class of sixty working in pairs generates thirty tins per practical, and a small oven quietly becomes the thing that decides how many practicals can run in a week. What setpoint does and does not tell you about a chamber is set out in the article on autoclaves, ovens and incubators.
For the balance, the mistake is usually the opposite: buying more readability than the room can support. A soil laboratory does not need a microbalance, it needs a balance that settles quickly in a room with a ceiling fan and grit in the air, and it needs the readability chosen against the smallest mass actually weighed rather than against a specification sheet. That argument is in why your analytical balance will not settle.
This is where most first purchase orders go wrong. There are two quite different families of equipment on the market and they are easy to confuse in a catalogue, because both are described as soil mechanics.
The first family is testing apparatus: the compaction moulds, the sieve set, the liquid and plastic limit apparatus, the direct shear box, the oedometer, the triaxial cell. These produce numbers that a student reports against a standard. They are what an examiner expects to see, and they are the backbone of the course.
The second family is demonstration and research rigs: instrumented units that show the physics rather than certify a sample. In the EDIBON range that Vvon supplies, PDFDC is a computer controlled glass sided drainage and seepage tank with piezometers along its length, so a class can plot a flow net for a dam, a sheet pile or a field drain and watch the phreatic surface move. PEFPC is a transparent permeameter that measures permeability under a controlled head and then keeps raising the upward flow until the bed fluidises. PAHSC drains a sand column in stages and reads suction, giving the soil moisture retention curve that unsaturated soil work depends on. EFAS does the same thing for groundwater flow between adjustable head tanks, and PPD runs constant head and falling head permeability tests on a packed column.
Both families are worth having. What a department must not do is buy only the second and assume the first is covered, because the rigs are photogenic, arrive on wheels and demonstrate beautifully at an inspection. A laboratory of demonstration rigs with no shear box cannot support a single final year project on bearing capacity.
Given a choice between a direct shear apparatus and a triaxial system on the same budget, buy the direct shear apparatus, and buy more than one. The reasoning is unromantic.
The same argument applies to any apparatus whose test time is measured in days rather than minutes. In a teaching laboratory, throughput per bench per week is a more useful specification than the capability of any single instrument.
An oedometer test runs for days with load increments applied on a schedule, which means the bench it stands on is occupied by time. Two consequences follow, and both should be settled at layout stage rather than discovered in the second semester.
First, put the consolidation frames where nobody has to walk past them with a trolley. A knock during a load increment is not recoverable, and in a shared room somebody will eventually park a mixer next to them. A dedicated run along a wall, ideally on a bench that is not continuous with the general working bench, is worth the floor space.
Second, count the frames against the class, not against the practical. If a batch takes a week and a class of sixty works in pairs, thirty tests through one frame is a term. Departments solve this by running consolidation as a demonstration with one sample, which is defensible, or by buying enough frames to run a rota, which is better and rarely budgeted.
A third point that only shows up later: the load hangers need a level floor and a wall that does not vibrate. A frame on a floor that flexes when a class walks in produces settlement readings with a signature that matches the timetable.
A geotechnical laboratory puts more solids into its waste line than any other room in a civil department. Sieve washing, sample preparation, permeameter discharge and general cleaning all carry fines, and fines settle in traps and horizontal runs.
Specify a settling pit or a grit interceptor on the laboratory waste before the floor is poured, with a lid that can be lifted by a technician rather than a contractor. Without it, the drain blocks in the first year, and the standard local remedy is to stop washing sieves in the sink and start washing them outside, which means the department has lost control of where its samples end up.
On the supply side, constant head work needs a stable head, which means an overflow tank rather than a tap. Campus water pressure in Dhaka varies with the pump schedule, and a permeability result taken straight off mains pressure records the building's pumping cycle. A small header tank with an overflow to drain fixes it for the cost of a plumber's afternoon.
A short list of the things that most often stop a Bangladeshi soils laboratory a year or two after commissioning, all of which are avoidable at specification stage.