Designing a fluid mechanics and hydraulics teaching laboratory: rigs, water services and a purchase order that follows the syllabus

How to plan an undergraduate fluids laboratory from the experiment list backwards: choosing bench-mounted against floor-standing rigs, getting supply and drainage right, and phasing the buy across an academic year.

A department orders a hydraulics bench, a Bernoulli apparatus and a pipe friction rig. The crates clear customs, the technician signs for them, and only then does anyone ask where the water goes. The bench is fine because it recirculates from its own sump. The friction rig needs a drain. The flume, when it is eventually funded, needs a supply and a drain at a flow rate no tap in the corridor can deliver. Nothing here is a fault of the equipment. It is a fault of ordering in the wrong sequence.

A fluids laboratory is the wettest and most services-dependent teaching space in most engineering faculties, and the one where a good layout keeps paying back for twenty years while a bad one is never really fixed. What follows is the order in which the decisions should be taken.

Work backwards from the syllabus, not forwards from the catalogue

Before any supplier is contacted, write out the experiments the course outcomes oblige you to run. Not the experiments you would enjoy having: the ones a student must have performed to be assessed honestly. For a mechanical or civil undergraduate programme, that list is fairly stable across universities, and it drives every other decision on the page.

Lecture topic	Laboratory experiment	Usual rig format
Hydrostatics	Centre of pressure on a submerged plane surface; metacentric height of a floating body	Bench-mounted accessory
Continuity and energy	Bernoulli's theorem, pressure distribution along a converging and diverging duct	Bench-mounted accessory, for example the EDIBON FME03
Flow measurement	Venturi meter, orifice plate and rotameter, discharge coefficient by comparison	Bench-mounted accessory
Laminar and turbulent flow	Osborne Reynolds dye injection, critical Reynolds number	Free-standing, needs a steady head tank
Pipe friction	Major losses and friction factor, minor losses across bends, valves and sudden changes of section	Frame or wall mounted, larger footprint
Momentum	Impact of a jet on flat, inclined and hemispherical targets	Bench-mounted accessory
Turbomachinery	Centrifugal pump characteristic curve, pumps in series and in parallel	Floor-standing
Hydraulic machines	Pelton wheel and Francis turbine performance	Floor-standing, high flow and head
Open channel flow	Flow over notches and weirs, sluice gate, hydraulic jump	Notches on the bench, the rest in a tilting flume
Unsteady flow	Tank discharge and time to empty, surge and water hammer	Free-standing

Two things fall out of this table. First, more than half the list can share one hydraulics bench, which is why an integrated set such as the EDIBON LIFLUBA basic fluid mechanics laboratory, or an FME00 bench with a family of accessories, is the sensible core purchase. Second, the four or five items that cannot share a bench are the expensive ones, and they dictate the room.

Bench-mounted against floor-standing

The hydraulics bench is a pump, a sump, a volumetric measuring tank and a working surface. Accessories drop onto it and connect with quick couplings. The trade-off is real and worth stating plainly to whoever signs the purchase order.

Consideration	Bench plus accessories	Dedicated floor-standing rig
Floor area	One bench footprint serves many experiments in sequence	One footprint per experiment, permanently occupied
Cost per experiment	Falls sharply after the first bench is paid for	Full cost every time
Flow and head available	Limited by the bench pump, which is fine for most demonstrations	Sized for the experiment, which is why turbines and flumes are separate
Changeover	A technician swap between accessories, which eats into a timetabled session	None
Single point of failure	A failed bench pump stops every accessory that depends on it	A failed rig stops one experiment
Suits	Hydrostatics, flow measurement, jet impact, notches, small friction loops	Reynolds apparatus, pump test sets, turbines, tilting flume, water hammer

The last row is the one to argue from. A department with two benches and one flume can teach a full syllabus. A department with six floor-standing rigs and no bench has spent more money and covers less ground.

Water: supply, sump, quality and where it drains

Municipal supply in most Bangladeshi campuses is intermittent and pressure varies through the day. Design as though the mains will be off during the afternoon lab session, because at some point it will be.

• Recirculate by default. Every experiment that can run from a closed sump should. Mains water should be a top-up, not a feed. This single decision removes the laboratory from the campus water argument.
• Size the make-up from the overhead tank, not the street. A dedicated draw-off from the building tank with an isolating valve and a hose union at each bench position is inexpensive at first fix and awkward later.
• Fit a non-return valve on every make-up point. A teaching laboratory with dye, glycerol and occasional tracer chemicals must not be able to back-feed the potable line.
• Plan for algae. Warm standing water in an open sump will grow biofilm within weeks, and it fouls manometer tappings and rotameter floats before anyone notices. Specify covered sumps, drain down at the end of each semester, and give the technician a written flush procedure.
• Drainage falls, not drainage holes. The floor should fall to trapped gullies. An untrapped gully in a Dhaka ground-floor laboratory is an invitation for smell and for insects in the monsoon.
• Assume splash. Chemical-resistant epoxy or a similar sealed floor finish, coved skirting so water cannot get under the screed, and no untreated timber anywhere near the wet zone.

If the department also teaches hydrology or environmental engineering, a rainfall and catchment unit such as the EDIBON ESHC hydrologic systems and rain simulator belongs in the same wet zone and on the same drainage strategy. Deciding that late means cutting a new channel through a finished floor.

Electrical supply in a room full of water

Pump motors, variable speed drives and computer-controlled rigs all live in a space where the floor is regularly wet. The specification is not exotic but it must be written down, because a contractor working to a generic office standard will not provide it.

• Residual current protection on every laboratory circuit, at 30 mA, with the sockets themselves mounted well above bench level rather than at skirting height.
• One local isolator per rig position, labelled, so a technician can work on a pump without killing the whole room.
• Suitably rated enclosures for anything within splash distance, and drip loops on every flexible lead.
• Voltage conditioning ahead of the sensitive loads. Supply voltage on many campuses wanders, and variable speed drives and data acquisition hardware are the first things to complain.
• A decision on what runs during load-shedding. A small uninterruptible supply on the instrumentation, even if the pumps stop, at least preserves the data already logged.

Where rigs are logged through a system such as EDIBON EDAS/VIS, run the data cabling in its own containment. Sharing a trunking run with drive cables is the standard way to acquire noise on a pressure transducer signal and then spend a semester blaming the sensor.

How many rigs for how many students

This is arithmetic, and it is worth doing on paper before the tender rather than in the middle of the first semester.

Take the largest section that will occupy the laboratory in one timetabled slot. Divide by the group size at which every student still gets to open a valve and read a manometer, which in practice is four or five, not eight. That gives the number of stations that must be running at the same moment. Suppose a section of sixty students in groups of five: twelve stations.

You do not need twelve of anything. You need twelve different experiments running in parallel, with groups rotating week by week, so the rotation closes in twelve weeks. If your experiment list only has eight entries, you have three options: split the section across two slots, duplicate four rigs, or accept groups of eight and the passengers that come with them. Choose deliberately.

Two rigs are worth duplicating whatever the arithmetic says: whichever experiment appears in the practical examination, because a breakdown in examination week is not recoverable, and the pipe friction rig, which gets the roughest handling and has the most joints to leak.

Sequencing purchases across an academic year

Most departments buy in tranches because that is how the budget arrives. The order below wastes the least money.

1. Services first, in the vacation. Drainage falls, gullies, floor finish, the make-up water ring, the distribution board and the containment. None of this is glamorous and all of it is disruptive once benches are in place.
2. The bench and its accessory family next. This is the purchase that covers the largest number of syllabus items per taka. Buy the bench and at least four accessories together so the technician learns the changeover routine on real teaching load.
3. Long lead-time floor-standing items in the same financial year, even if delivery is later. A tilting flume and a turbine test set are made to order and shipped as heavy freight. Place them early and let them arrive while the bench work is bedding in.
4. Data acquisition after the mechanical rigs are proven. Adding logging to a rig whose manometers you already trust is straightforward. Commissioning both at once means every anomaly has two possible causes.
5. Spares and consumables on the same order as the parent equipment. Seals, impellers, spare tappings, manometer fluid, dye, quick-coupling inserts. Ordering these separately six months later means a second import cycle, a second letter of credit and a second customs clearance for parts that would have travelled free in the original crate.
6. Hold a retention against commissioning with students present. Acceptance should mean a full class has run a full experiment and produced results that match theory within a stated tolerance, not that the crate was opened and the pump started.

Mistakes that recur

• Buying the flume before the drain exists. It then sits crated in a corridor for a year, and the warranty clock runs anyway.
• Specifying the wrong mains voltage and frequency on the order. Confirm it in writing on the purchase order. A transformer bolted on afterwards is an eyesore and a maintenance liability.
• Accepting a rig whose pump has no locally available equivalent. Ask, before signing, what happens when the pump fails in year four. If the answer is a long import cycle, negotiate a spare into the original order.
• Skipping the technician handover. The person who will change accessories two hundred times should be in the room during commissioning, with the manual open, taking notes in their own words.
• Leaving no dry bench space. Students need somewhere to put a notebook that is not the wet surface. Every layout that fills the room with rigs produces soaked lab sheets.

The cheapest change to a fluids laboratory is the one made on the drawing. The most expensive is the one made after the floor has been laid.: Vvon Technologies installation team

Back to all Insights