Key takeaways
- The SIA 358 standard requires the stiffness to be assessed without exceeding the elastic range of the steel.
- The jack or hoist is always installed with a calibrated dynamometer and a stable support point.
- Deflection measurement requires a dial gauge fixed on a tripod that is completely independent of the slab.
- The joint inspection record signed on site records the residual deformation and releases the final acceptance.
Preparing the pulling device and dosing the force according to SIA 358
On a metalwork construction site, the strength of a guardrail is sometimes the subject of a technical dispute between the metalworker, the civil engineer and site management. Discussions often concern the perceived stiffness of the handrail, the actual holding capacity of the anchors in recently cast concrete, or the anchorage on a slab whose effective thickness differs from the execution drawings. Instead of multiplying theoretical meetings or after-the-fact calculations, the static load test in place makes it possible to verify the mechanical behaviour of the structure without altering its components or damaging the final fixings.
The equipment used to carry out the test must be transportable and self-contained. The typical set-up comprises a lever hoist or a compact hydraulic jack with a through screw, connected in series to an electronic load cell with a valid calibration certificate. The point of application of the force is placed at the top of the post or at the centre of the most unfavourable handrail span, generally on the connection furthest from a rigid return corner. A wide padded textile strap prevents any localised marking on the powder coating or stainless steel.
- Prior check of the tightening of the anchors to the nominal torque with a torque wrench.
- Visual inspection of the base welds and verification of perfect flatness under the base plate.
- Fitting of a textile strap lined with felt to protect the finish coat of the sections.
- Strict alignment of the pulling axis perpendicular to the vertical plane of the guardrail.
- Checking the calibration of the dynamometer display before the first loading step.
Measuring the behaviour under load and the elastic recovery, and drawing up the test record
The validity of a load test rests entirely on the independence of the measurement reference. Installing a dial gauge or a laser sighting rule on the guardrail itself or on the slab immediately adjacent to the base plate distorts all the readings, because the concrete support deflects locally under the lever effect. The measuring device must be placed on a heavy, decoupled tripod, set at least one and a half metres from the deformation zone, bearing directly on the floor.
- Preloading and zeroingA minimal stabilising preload is applied to take up the mechanical play in the bolted connections and the fasteners. The mechanical dial gauge, reading to one hundredth of a millimetre, is then zeroed against the opposite face of the handrail.
- Progressive load stepsThe load is increased in successive steps of twenty-five per cent of the agreed maximum force. Each step is held for two minutes to monitor the stabilisation of the gauge needle and detect any anchor slippage.
- Holding at full test loadThe total loading defined according to SIA 358 is held for a duration of five minutes. The maximum elastic deflection under load is recorded before unloading.
- Complete unloading and check of the elastic recoveryThe force is released gradually down to zero. Wait three minutes for relaxation before reading the permanent residual deformation on the dial gauge.
| Measured parameter | Acceptable behaviour | Sign of mechanical failure |
|---|---|---|
| Deflection under nominal load | Proportional displacement without jerks, meeting the visual comfort criterion of SIA 358 | Sudden acceleration of the deflection indicating yielding of the steel or slippage |
| Elastic recovery after unloading | Recovery close to zero, with tolerance for the initial take-up of connection play | Visible permanent deformation indicating that the elastic limit of the section has been exceeded |
| Base plates and anchor rods | No detachment of the bedding mortar, no play under the washer and no cracking of the concrete | Spalling of the concrete cover, partial pull-out of the cone or stretched threaded rod |
The test legally protects the company only if it is immediately entered in a joint inspection record drawn up on the construction site. This document states the exact identification of the guardrail tested, its location on the installation drawing, the characteristics of the dynamometer used, the values recorded at each loading step and the final permanent deflection. The joint signature of the metalworking team leader, site management and the civil engineer attests to the conformity of the structure and allows the immediate lifting of the installation reservations.
Conducting the test step and reading the elastic recovery without damage
The test in place on an installed guardrail is not intended to test the structure to destruction, but to provide material proof that the assembly made up of the stringer, posts, handrail and fixings withstands the regulatory loads without abnormal residual deformation. A doubt raised by site management about the stiffness of a top rail or the holding capacity of a chemical anchor at the slab edge is dispelled once a reproducible physical protocol confirms the deflection criteria defined by the SIA 358 standard.
To avoid damaging the metalwork or the mineral substrate, the force must never be applied in jerks or by means of a manual winch without continuous monitoring. The horizontal force is introduced gradually using a hydraulic jack fitted with a calibrated pressure gauge or an electronic load cell with direct reading in kilonewtons. The point of application is on the handrail, at the axis of an intermediate post or midway between two posts, depending on the zone disputed by the parties.
The deflection must be measured from a geometric reference that is completely independent of the structure being tested. Placing a dial gauge or a laser target on the same slab that deflects under the action of the jack immediately distorts the acquisition. A heavy tripod wedged on a fixed point or a stabilised bracket outside the zone of influence of the guardrail is installed. The measurement is made in three stages: zeroing under an alignment preload to absorb the mechanical play, holding under static load for a stabilised duration, then complete unloading to establish the geometric recovery.
| Test step | Loading level | Holding time | Acceptance criterion |
|---|---|---|---|
| Setting the reference | Preload of 0.1 kN | 1 minute | Take-up of fixing play and zeroing of the dial gauges |
| Intermediate step | 50 % of the test load | 2 minutes | Linear progression of the deformation, no anchor noise |
| Nominal load SIA 358 | 100 % of the design value | 5 minutes | Horizontal deflection stabilised below the maximum permitted threshold |
| Recovery after unloading | Force brought back to zero | 3 minutes | Residual deformation nil or within the elastic recovery tolerance |
If the metal deflects under the nominal load but returns fully to its original position after the jack is released, the load-bearing capacity in service is demonstrated. A permanent deformation after unloading, on the other hand, reveals that the elastic limit of the steel has been exceeded, that the anchor studs have slipped or that the concrete has been locally crushed under the fixing plate.
Drawing up the joint inspection record and technical acceptance
A tensile or pushing test has technical value only if it is the subject of a joint inspection record drawn up on the spot. If the metalworker carries out the test alone in a corner, the results remain open to challenge and extinguish no legal reservation. The session must bring together around the structure the metal contractor, the civil engineer responsible for the structures and the representative of site management, in accordance with the general rules of the SIA 118 standard.
- Joint preliminary verificationVisual check of the welded connections, the bolting and the approval sheet of the chemical or mechanical anchors before any mechanical loading.
- Recording the calibration dataEntry in the document of the validity date of the calibration certificate of the load cell or pressure gauge and of the exact position of the displacement sensor.
- Recording the raw valuesNumerical notation, in millimetres, of the instantaneous deflection under load and of the residual state after the push is removed, for each span examined.
- Joint signature on the spotImmediate validation initialled by the metalworker, the engineer and site management, attesting to conformity or setting out corrective measures.
The record drawn up on the construction site is authoritative for lifting the execution reservations. It includes the location sketch with the precise identification of the spans tested, the photographs of the test rig and the reference to the requirements of the SIA 358 standard. By annexing this document to the project file handed over to the client, the contractor definitively closes any dispute concerning the stiffness of the structure and validates the conformity of its finishing work services.
What to do in the event of a non-conforming result or a technical disagreement?
When a horizontal pushing or tensile test reveals behaviour that does not comply with the requirements of the SIA 358 standard, the increase in load must be stopped before the structure is irreparably damaged. The finding of non-conformity shows itself in two distinct phenomena: excessive deflection under load, indicating insufficient stiffness of the rail, or a permanent residual deformation read on the dial gauge after complete unloading. This residual deflection indicates that the yield strength of the steel has been exceeded, or that the anchorage has undergone an irreversible displacement in its substrate.
Once these values have been entered in figures in the joint inspection record, those involved avoid any improvisation and start a rigorous technical sequence:
- Secure the drop: if the element tested shows worrying play or a partial failure at the slab edge, temporary collective protection complying with the Ordinance on construction work is put in place without delay.
- Diagnose the exact origin: check for the absence of pull-out of the anchoring resin, crushing of the neoprene shims under the plate, deformation of the base or local micro-cracking of the balcony edge.
- Check the conformity of the installation: verify whether the tightening torques applied, the class of fasteners and the distances to free edges match exactly the anchor manufacturer's instructions.
- Decide on a corrective measure: consider fitting through counter-plates with threaded rods, reinforcing the post with an internal steel sleeve, or adding welded triangular gussets.
Any structural repair necessarily requires a second joint load test on the modified module. In the event of a formal dispute by site management over the overall stiffness, the joint use of an independent civil engineer makes it possible to arbitrate the deflection criteria within the meaning of the SIA 358 standard, thus protecting the craftsman against unfounded rework.
Frequently asked questions
To what force value should the load test be calibrated for a dwelling according to SIA 358?
For private residential premises, the characteristic service load defined by the SIA 358 standard is generally 0.8 kN per linear metre on the top rail. The test in place must never push the structure to failure: the service load agreed with site management is applied, increased where applicable by a test factor defined in the specifications, while carefully checking the elastic behaviour of the structure.
How should the pushing point be fixed during the test without damaging the powder-coated finish?
The force of the jack or dynamometer is never applied directly to bare metal. The metalworker inserts a sacrificial block of hardwood or a dense elastomer pad, cut to the profile of the handrail or post. This piece spreads the pressure over about ten centimetres in length, avoiding any localised indentation, flaking of the powder coating or deformation of the hollow section during the progressive increase in force.
What percentage of elastic recovery is considered acceptable after the test?
Perfect elastic recovery does not exist in real construction site conditions. A slight residual deformation, generally less than a few millimetres depending on the span and the geometry of the structure, is tolerated. It results from the take-up of mechanical play at the base plate holes and from the prestressing of the anchors. The decisive factor remains the absence of continuous creep under sustained load and the complete absence of visible plastic deformation of the post.
Who must sign the on-site load test record?
The record must be drawn up and initialled jointly on the spot by the installation manager of the metalworking company and by the representative of site management or of the client. If a civil engineering firm has prescribed the load test, its technical delegate must also countersign the gauge and dial gauge readings. This joint signature formally validates the measuring method and prevents any later challenge.
Texts and standards cited
- SIA 358Guardrails - Construction provisions and imposed loads
