What are BS 476-20 - 23?

  • Est Time: 10 mins
Read our blog below and watch the video above to learn more about BS 476-20 - 23, don't forget to test you've understood everything by taking the short quiz afterwards.
BS 476 Part 20: 1987 sets out the general principles for assessing the fire resistance of construction elements. It sets clear criteria by which an element’s loadbearing capacity, fire containment (integrity) and the thermal transmittance (insulation) can be assessed. It is supported by three further standards which provide specific guidance for testing different categories of construction elements:

1. Loadbearing elements (BS 476 Part 21: 1987) - This is applicable for:
  • beams;
  • columns;
  • floors and flat roofs (up to a 10-degree pitch); and
  • walls. 
2. Non-loadbearing elements (BS 476 Part 22: 1987) - This is applicable for:
  • partitions;
  • fully insulated, partially insulated and uninsulated doorsets and shutter assemblies;
  • ceiling membranes; and
  • glazed elements.
3. Elements that contribute to the fire resistance of a structure (BS 476 Part 23: 1987) - This is applicable for:
  • suspended ceilings which protect steel beams; and
  • intumescent seals for use in conjunction with single-acting, latched timber fire resisting door assemblies.
How are elements assessed under these standards?

The construction elements are tested within a large gas fired furnace following a procedure which simulates a fire within an enclosed room. The test specimens can either be assessed vertically or horizontally (depending on their final application). Vertical separating elements (such as walls and partitions) are exposed to heat and pressure from one face whilst horizontal separating elements (such as floors and flat roofs) are exposed to heat from the underside.

Wherever possible, the test specimen (and any associated construction) should be full sized. If no suitably sized furnace is available, then the minimum size of the separating element exposed to fire must be:
  • 3 m high by 3 m wide for vertical elements (or 2.4 m by 3 m where load bearing forces are applied); or
  • 4 m span by 3 m wide for horizontal elements.
Loadbearing elements are subjected to either mechanical or deadweight loading which simulates the loads they are likely to experience in their final application.
The chamber depth for vertical furnaces (distance between the exposed face of the specimen and the furnace lining) must be between 600 mm and 1300 mm. In horizontal furnaces, the depth (distance between the soffit and the furnace floor) should be between 1000 mm and 2000 mm.

Test Process

During the test, the temperature within the furnace rises sharply then gradually starts to level out as shown by the temperature curve graph below:
Thermocouples (temperature sensors) within the furnace monitor the temperature during the test to ensure the average temperature follows the correct curve.
The furnace is designed to ensure that the neutral pressure plane is established 100 mm above the notional floor level. The pressure gradient within the furnace is expected to be 8.5 Pa per metre height but pressure at the top of the specimen must not exceed 20 Pa.

For example, if a glazing panel is typically installed in a wall so that the top of the panel is 2.8 m above floor level then this would be expected to experience an overpressure of 15.3 Pa. The pressurisation is monitored by a pressure sensing probe within the test chamber and should not exceed a maximum value of 20 Pa.

The test is either run until failure in all performance criteria or until a time agreed by the testing centre and sponsor prior to the tests.

Performance Criteria

The specimens are assessed against one or more criteria depending on their practical use:

Loadbearing capacity (BS 476 Part 21: 1987 only)
Loadbearing vertical elements are adjudged to fail when there is a rapid change in the deformation rate of the element. This is judged visually, based on indicators such as lateral deflection at the centre of the element. Columns are assessed solely against this criterion within BS 476 Part 21: 1987.

For loadbearing horizontal elements, the specimen is deemed to fail when either of the following values is exceeded:
  • a deflection equal to the span of the specimen in mm (L)/20; or
  • the rate of deflection (in min/mm) calculated in 1-minute intervals from 1 minute after the test began exceeds the following equation: rate of deflection = L2/ 9000 x the distance from the top of the structural section to the bottom of the design tension zone in mm.
Beams are only assessed under this criterion within BS 476 Part 21: 1987.

Failure is judged to occur when collapse or sustained flaming is observed on the unexposed face of the element or when the criteria for impermeability are exceeded. To judge this, a cotton pad is held near to any gaps, cracks or fissures which develop. The impermeability criteria are judged to be exceeded if there is any flaming or glowing on the cotton pad.

The cotton pad test is stopped if temperatures around the gap exceed 300 degrees Celsius. At this point, failure is deemed to occur if:
  1. a 6 mm diameter gap gauge can penetrate through the gap into the furnace and can be moved within the gap for at least 150 mm; or
  2. a 25 mm gap gauge can penetrate through a gap so that its end projects into the furnace.
Fully insulated, partially insulated and uninsulated doorsets and shutter assemblies are assessed under separate integrity criteria within BS 476 Part 22: 1987.

Failure is judged to occur when one of the following criteria is met:
  • if the mean temperature recorded by thermocouples on the unexposed face of the specimen rises by more than 140 degrees above its initial value;
  • if the temperature recorded by any of the fixed thermocouples or a roving thermocouple exceeds 180 degrees above the initial mean value for the unexposed face; or
  • when integrity failures take place.

The test results are stated as the elapsed time from commencement to failure rounded down to the nearest minute as shown in the example below:

Loadbearing capacity      83 mins
Integrity                            83 mins
Insulation                          70 mins

The period given for the insulation failure cannot exceed that for integrity failure. If no integrity or insulation failure occurs, then the result is the time up until the agreed end of the heating period. The results are given as part of a full report with additional observations and measurements.

Tested Kingspan Systems
Kingspan Kooltherm K12 Framing Board and the Kingspan TEK Building System have been tested in specific configurations under BS 476 Part 21: 1987.
Kingspan Thermawall TW55 has been tested in specific configurations to BS 476 Parts 21 and 22: 1987.

Full details of its test results are available within the product literature.

You can download the full testing standards from the British Standards Institution shop.