The code — IS 14687:1999
IS 14687:1999 is the Indian standard for falsework for construction works — guidelines. It covers design principles, load cases, material requirements, and safety provisions for temporary works. The word "guidelines" in the title is misleading — the loads and tolerances inside are hard numbers. Treat them as mandatory.
I see formwork treated as a trade craft rather than engineering on too many Indian sites. The carpenter decides tie spacing based on experience, the supervisor checks by eye, and nobody runs the pressure calculation. This works — until it doesn't. A blown formwork failure dumps cubic metres of wet concrete on the slab below, with people underneath. The math isn't hard. Let's walk through it.
Lateral pressure of fresh concrete
Fresh concrete behaves like a fluid before it sets. It pushes outward against your formwork, and the faster you pour, the higher the pressure climbs. IS 14687 lays out the factors: rate of pour (R in m/hr), temperature (T), concrete consistency (slump), and vibration depth.
The worst case is full hydrostatic pressure: P = density × height = 24 × H kN/m². This happens when the concrete stays fluid through the full pour height — slow-setting mixes, cold weather, or very rapid pours where the concrete hasn't started stiffening at the base by the time you've filled to the top.
CIRIA Report 108 (an international reference, since IS 14687 references general principles) gives the practical formula for walls:
Where C is a coefficient depending on temperature and cement type (lower temperature = higher C), Cw is the concrete density coefficient (1.0 for normal weight), and R is the pour rate in m/hr. For columns, the formula is fuller because the cross-section restrains the concrete differently.
The pressure envelope on the form is typically triangular — maximum at the base, decreasing upward as concrete starts setting. For a 3 m wall pour at 2 m/hr in Mumbai at 30°C, Pmax works out to roughly 40–50 kN/m² at the base. That's a lot of force on a plywood panel.
Dead loads, live loads, lateral loads
Self-weight of wet concrete at 24 kN/m³ + formwork self-weight (typically 0.5–1.0 kN/m² for timber/steel systems). Don't forget the weight of reinforcing steel — adds 1–1.5 kN/m³ to the wet density.
Workers, equipment, material storage: 2–3 kN/m² minimum. Add impact loads for pumped concrete placement — sudden surges when the pump pushes through a blockage.
Wind on the formwork (especially tall wall and column forms), equipment vibration (concrete pumps, vibrators), and the impact of concrete placement. Provide diagonal bracing for all vertical forms.
Deflection limits — what the eye can see
IS 14687 sets the deflection limit for formwork: 3 mm for a beam span below 3 000 mm, and the least of 30 mm and L/1000 above it, for visible concrete surfaces. For architectural or premium finishes, the limit tightens further.
Why does this matter? Because formwork deflection is permanent. The concrete takes the shape of the form. If your plywood sags 8 mm between studs, you'll see an 8 mm ridge on the wall surface after stripping. You can't grind it out without cosmetic patching.
The rule of thumb: your formwork must be stiffer than the structure it's supporting. If a slab span deflects 10 mm at service load, your formwork deflection must be tighter — or the concrete will set with that deflection baked in.
Tolerances — IS 14687
Even well-designed formwork isn't perfect. IS 14687 specifies acceptable tolerances:
| Deviation | Tolerance |
|---|---|
| Positional (plan location) | ±25 mm |
| Cross-sectional dimensions | ±6 mm to ±12 mm (size dependent) |
| Verticality (per metre height) | ±6 mm per m, max ±25 mm total |
Reference: IS 14687:1999 — Tolerances
Common failures — the patterns I see
Wall and column forms blow out or tip over under wind or concrete placement impact. Diagonal bracing is the cheapest insurance against formwork collapse. Don't skip it.
Tie rods snap or pull through walers when spacing is too wide for the concrete pressure. Always check tie capacity against the pressure envelope — especially at the base where Pmax occurs.
Props punch through the slab below, or settle into soft ground. Use sole plates — timber or steel spreaders — under every prop base. Check bearing capacity of supporting slab.
The formwork was designed for 1.5 m/hr, but the pump operator is delivering 4 m/hr. Pressure exceeds design and the form blows. Communicate pour rate limits to the pump operator before starting.
Use the formwork pressure calculator to run the pressure envelope for your specific wall height, pour rate, and temperature.
Field-checklist blocks
A short questionnaire, an inspector's checklist, and three worked examples that show how a formwork design is checked against the pressure envelope, the deflection limit and the stripping schedule, and resolves into a go or no-go on the pour.
Questionnaire — ask before the pour is authorised
- What wall height, pour rate and temperature was the form designed for, and what rate is the pump actually set to? Full hydrostatic pressure is 24 × H kN/m², and CIRIA Report 108 gives P max = C × C w × √R for walls.
- What is the pressure at the base of the form, and do the tie rod spacing and the tie capacity on site match that pressure envelope, or a carpenter's judgement of it?
- Has the formwork been checked for deflection under its own dead load, the wet concrete and the construction loads, against IS 14687 Cl 7.5?
- Is the pour rate limit written down and read back by the pump operator before the first bucket goes in?
- Are all vertical forms diagonally braced for wind, equipment vibration and placement impact under Cl 7.3.3?
- Is there a sole plate under every prop base, and has the bearing capacity of the slab or the ground below been checked?
- What stripping time applies to this member under IS 456 Cl 11.3, and which cube result releases the soffit and which one releases the props?
- Does the load take-off include the reinforcement (1–1.5 kN/m³ on top of 24 kN/m³ of wet concrete) and the formwork self-weight (0.5–1.0 kN/m²)?
Inspector's checklist — what to verify on site
- Pressure envelope on file for the wall: height, pour rate, temperature, and the P max at the base, run through the formwork pressure calculator for your own inputs.
- Tie rod schedule with spacing and capacity at each lift, checked against the envelope, with the bottom row carrying the highest load.
- Deflection calculation for the actual span against Cl 7.5: 3 mm below 3 000 mm, and the least of 30 mm and L/1000 above it.
- Load take-off showing the Cl 7.3.1 dead load (24 kN/m³ wet concrete, formwork 0.5–1.0 kN/m², reinforcement 1–1.5 kN/m³) and the Cl 7.3.2 live load of 2–3 kN/m².
- Bracing and bearing: diagonal bracing on every vertical form for Cl 7.3.3, and a sole plate under every prop base with the bearing capacity of the slab below recorded.
- Written pour rate limit issued to the pump operator, with the acknowledgement on the pour card.
- Cube results on file before any soffit is stripped, and the props held to the IS 456 Cl 11.3 schedule: 16–24 hours for walls and columns, 3–7 days for slab soffits with the props left, 7–14 days for beam soffits, 14+ days for cantilevers, and props to slabs out at 7 days minimum or 70% fck.
- Tolerance record against IS 14687: positional ±25 mm, cross-section ±6 mm to ±12 mm, verticality ±6 mm per metre and ±25 mm in total.
What happens if… three worked examples
What happens. The design pressure was worked out for 1.5 m/hr. At 4 m/hr the pour rate is 4 ÷ 1.5 = 2.67 times the design rate, and because P max = C × C w × √R the pressure scales with the square root of the rate: √(4 ÷ 1.5) = 1.63. The wall carries about 63% more pressure at the base than the panels and ties were sized for. The absolute ceiling for a 3 m wall is the full hydrostatic 24 × 3 = 72 kN/m², and a 4 m/hr pour walks a long way toward it. Nothing on site tells you any of this has happened, because the form looks identical at 10 am and at 2 pm.
Likely outcome. The ties pull through the walers or the panels bow, and you lose the wall and the concrete behind it. I've seen this fail more often than I can count, and it is always the pump operator who was doing exactly what they were asked to do. Put the rate limit in writing on the pour card and get it read back before the first bucket.
Preventive correction. Make the pour rate a stated hold point in the pre-pour checklist, not a note in the method statement that nobody sees at 4 pm.
What happens. IS 14687 Cl 7.5 caps the deflection at 3 mm for a span below 3 000 mm, and above 3 000 mm at the least of 30 mm and L/1000. Take the article's own case, a plywood panel that sags 8 mm between studs. That is 8 ÷ 3 = 2.7 times the 3 mm cap that applies below 3 000 mm. Above 3 000 mm the cap opens up, but only to the least of 30 mm and L/1000, so an 8 mm sag clears the L/1000 rule only once L reaches 8 mm × 1 000 = 8 000 mm, and it is non-compliant at every shorter span.
Likely outcome. The concrete sets in the shape of the form, so an 8 mm ridge is now an 8 mm ridge in the beam soffit, and you can't grind it out without cosmetic patching. Stiffen the panel or add a prop line before the pour, not after it.
Preventive correction. Measure the actual panel deflection under a test load and record it, so the check is a number on the file rather than an opinion about plywood.
What happens. IS 456 Cl 11.3 puts slab soffits at 3–7 days with the props left in, so a 5-day release takes the soffit off inside the window. It does nothing for the props: those come out at 7 days minimum or once the concrete has reached 70% fck, and 5 days is short of the 7-day minimum. The cube results are the only other route out.
Likely outcome. The slab carries its own weight and the construction loads through props that were struck early, and the soffit deflection that follows gets recorded as a concrete defect instead of a sequencing failure. Hold the props for the full 7 days minimum or until the cubes show 70% fck, and in cold weather extend every one of these times.
Preventive correction. Issue the prop release on cube strength, not on the calendar, and keep the back-shuttering on the pour record so the two dates are never confused.
Frequently Asked Questions
Related articles
Types of Formwork and Form Liners — Timber, Steel, Aluminium
The companion piece — materials, systems, and architectural finishes.
Why Concrete Cracks: 7 Types, Causes, Prevention & Repair
Formwork-related cracks — early stripping, settlement, and deflection patterns.
Amit Haridas
Founder & Proprietor, ConcreteInfo. 25+ years experience in construction QA/QC, concrete technology, and RMC plant operations. NRMCA Certified Trainer (USA) and ISO Lead Auditor.
Technical updates, not sales emails
One email when a new article goes out on concrete technology, QA/QC and construction quality. No promotions, and your address is never sold or passed to advertising networks. Subscribing is separate from making an enquiry.