The present invention relates to a method of measuring the insulation provided across a material, particularly though not exclusively a smoke or fire barrier, especially a barrier incorporating intumescent material.

The current standard method of measuring insulation provided by a smoke or fire barrier is to take a sample of the barrier material, expose one side to a heat source, for example a furnace, and measure the temperature on the other, unexposed side. The barrier is deemed to have failed when the surface exceeds a mean temperature of 140°C above ambient within a certain period of time, or at any one spot where the temperature is above 180°C, again within a certain period of time. Usually the temperature will be measured with a thermocouple as these can be adhered or fixed to the material, but the temperate read at a distance. Typically thermocouples are positioned in each quartile of the sample and in the centre. A sample may be up to 3m by 3m. In addition thermocouples are placed in discrete areas and are set in from any gaps. A roving thermocouple, namely a thermocouple disc mounted on a pole, can be used to access "hot-spots" throughout the sample where fixed thermocouples cannot measure. Generally the thermocouples attached to the sample are discs approximately the size of a one pence piece, which is attached, and a tail, from which the measurement is taken.

This is not entirely satisfactory, as the attachment of the thermocouple to the sample is not standardised. Thus different attachment methods may produce different results, depending in part on whether the thermocouple is in close contact with the material or at a short distance.

In addition the testing of barrier materials incorporating an intumescent material causes difficulties. Intumescent material, by its nature, reacts to heat, swelling to reduce heat transfer through the material. However, this makes attachment of thermocouple discs to the material very difficult. Adherence to the surface of the material either results in the thermocouple becoming detached when the material expands, or prevents expansion of the material. Similarly, physically fixing them to the material, for example by stitching, prevents the intumescent process from occurring. As a result no accurate testing of intumescent material as a fire or smoke barrier can be carried out.

The object of the present invention is to provide an improved method of testing the insulation provided by a material, particularly for use in a fire or smoke barrier.

According to the invention there is a method of measuring the insulation provided by a sample of smoke or fire barrier curtain material, comprising the steps of:

connecting a temperature sensor to the sample by physical means whereby the physical means maintains surface contact of the temperature sensor against the surface of the sample;

exposing a side of the sample to which the temperature sensor is not held against to an elevated temperature; and

measuring the temperature time relationship on the unexposed surface of the sample with the temperature sensor.

The method being such that if the sample reacts to the heat, for example if the sample is an intumescent material, the temperature sensor will remain in contact with the non-exposed side, such that a temperature can be measured which indicates the insulation provided by the sample.

In some embodiments the temperature sensor may be urged into to contact with the sample by a wire passing through the sample connected on one side to the temperature sensor and on the other side to a weight. If intumescence occurs, the sample is free to expand, but the weight on the wire keeps the temperature sensor in contact with the surface of the sample. Alternatively the temperature sensor may be attached to the sample via a pair of wires forming a U-shaped connection, the length of the sides of the U being slightly longer than the maximum swelling of the sample. If intumescence occurs, the sample can expand due to the length of the wires, but the temperature sensor will remain in contact with the surface of the sample. In a further alternative, the temperature sensor may be connected to a sprung counterbalance provided on the thermocouple side of the sample. This allows the sample to swell, but ensures that the temperature sensor remains in contact with the surface of the sample.

Preferably the temperature sensor may be a thermocouple.

According to a second aspect of the invention there is provided means for measurement of the insulation provided by a sample, comprising, a temperature sensor and physical means for maintaining the temperature sensor in contact with the sample.

In one embodiment the means for maintaining contact may be a wire adapted to be passed through the sample and provided with a weight on one end. In another embodiment the means for maintaining contact may be a U-shaped section on wire, adapted to pass through the sample in two positions, and secured to the temperature sensor at both ends, such that swelling of intumescent material is accommodated in the U, while maintaining the contact between the temperature sensor and the sample. In a further alternative the temperature sensor may be attached to a sprung

counterbalance. This enables the intumescent to react and swell thus limiting transfer of heat through the sample within acceptable levels whilst measuring surface conduction.

Preferably the temperature sensor will be a thermocouple. Usually this will be in the form of a disc thermocouple, having a pad, which can be adhered to the sample, and a tail. Alternatively other forms of thermocouple can be used.

The means for maintaining the temperature sensor in contact with the sample may include adhesive in addition to the physical means.

According to a further aspect of the invention there is provided a fire or smoke curtain in which the lower end of the curtain, in deployment, is provided with additional insulation. Preferably the lower 1-1.5m of the curtain is provided with additional insulation. Alternatively the lower 0.1 -0.5m may be provided with the additional insulation. Preferably the additional insulation is provided at 1 100mm or 100 mm.

Typically the additional insulation or protection will be provided as additional layers of the curtain material from which the main curtain is made. Alternatively, additional layers of different materials may be used, for example material

incorporating intumescent materials. The materials must also be fire-resistant, flexible materials.

To help understanding of the invention, a specific embodiment thereof will now be described by way of example and with reference to the accompanying drawings, in which:

Figure 1 is a cross-sectional view through a sample being tested with a thermocouple;

Figure 2 is a sectional view of a thermocouple connected to a sample according to the invention;

Figure 3 is a view of the sample and thermocouple of Figure 3, after intumescence;

Figure 4 is a view of a thermocouple connected to a sample according to a second embodiment of the invention;

Figure 5 is a view of the sample and thermocouple of Figure 4 after intumescence;

Figure 6 is a view of a thermocouple held against a sample according to a third embodiment of the invention;

Figure 7 is a view of the sample and thermocouple of Figure 6 after intumescence;

Figure 8 is a perspective view of a fire or smoke curtain according to a further aspect of the invention; and

Figure 9 is a cross-sectional view of the fire or smoke curtain of Figure 8.

Referring to Figure 1, in order to measure the insulation provided by a sample 1 , in particular a sample of a fire of smoke barrier, the sample can be tested by exposing one side 2 to a heat source 4, for example a furnace, and taking temperature reading on the other side 6.

The temperature reading is usually taken with a thermocouple 8 as this can be attached to the sample, and the readings can be taken some distance away. Current test standards require a fire barrier to have a mean temperature of less than 140°C plus ambient on its unexposed surface, for a set period of time, and to have no one spot having a temperature of more than 180°C after a set period of time, when exposed to a particular temperature.

Generally a significant piece of the material will be provided with an array of thermocouples adhered or otherwise fixed at various positions. For example thermocouples may be provided in each quartile with a further thermocouple in the centre of the sample. The thermocouples commonly used are thermocouple discs having a disc approximately the size of a one pence piece. In addition the thermocouples are generally adhered to the surface of the sample with an adhesive.

While this is satisfactory for many samples, this is not satisfactory for samples that change as a result of the heat, in particular intumescent materials. These materials are regularly used in fire barriers as they reduce temperature after swelling, and yet before reacting to the heat of a fire, can be rolled and stored like any other fabric.

Such materials are often silicone coated and if a thermocouple is adhered to the surface of intumescent material in the normal way, it will usually become detached when the material heats and/or swells. The use of stronger adhesives, or physically attaching the thermocouple to the material by mechanical means, for example with stitching or staples, usually prevents intumescence occurring. This is clearly unsatisfactory as it does not enable a useful test to be carried out.

We have therefore sought to develop a new method of testing materials for their insulation properties, typically for use as fire or smoke barriers. The method involves ensuring that the thermocouple or other temperature sensing means remains in contact with the surface of the material throughout the test.

As shown in Figures 2 and 3, the thermocouple 28 can be held in contact with the surface of the sample 21 by attaching it to a wire 22 passing through the material and provided with a weight 24. As shown in Figure 2, at the start of the test before the material has swelled, the weight 24 maintains the thermocouple against the sample 21. On swelling, as shown in Figure 3, the wire connection allows for expansion of the sample 21 , but the weight still continues to hold the thermocouple in contact with the sample 2 . Thus the temperature of the surface of the sample can be taken throughout the test.

Referring now to Figures 4 and 5, an alternative arrangement is shown in which the thermocouple 48 is held against the sample 41 by a loop of wire 44 passing through the sample in two positions 42, 43, and connected to the thermocouple 48 at each of its ends, 49, 50. The loop is substantially U-shaped and is sufficiently long to accommodate the full width of the expanded intumescent material. At the start of the test, as shown in Figure 4, the thermocouple is secured to the sample via the loop of wire and adhered into position. The wire extends around the back of the sample, with the excess being in this position, As the material intumesces, as shown in Figure 5, the wire is drawn through the sample ensuring that the intumescence can occur, but also ensuring contact is maintained between the sample 41 and the thermocouple 48.

In these two embodiments, shown in Figure 2 & 3, and 4 & 5, the

thermocouple may initially be adhered to the surface of the sample, with adhesive 27, 47. The adhesive may release during the test, as shown in Figure 3, or may not, as shown in Figure 5, due to the thermocouple being maintained against the sample using additional means. The adhesive initially assists in the connection of the thermocouple to the sample and where is does not fail, it will assist connection between the thermocouple and the sample after intumescence.

The disadvantage of the two alternative embodiments shown above, is that in order to ensure contact between the thermocouple and the sample a cable, wire or the like is passed through the sample. This results in a passageway for the transfer of heat, and the cable or wire itself will also heat up and transfer heat. As a result the temperature measurement will be affected.

As shown in the embodiment of Figures 6 and 7, the thermocouple 68 is held against the sample 61 by a sprung counter balance 64, all provided on the

thermocouple side of the sample. The thermocouple is mounted on a spring 69, held by the sample holder 70, just above the sample. This urges a support 72, as shown a bent support, against the sample with sufficient force to hold the thermocouple against the sample but not sufficient to prevent intumescence or other movement of the support occurring, as shown in Figure 7. In an alternative, not shown, the sprung counter balance can be provided below the sample, either on the sample holder or on the ground.

This embodiment does not require any apertures in the sample and thus is preferred.

Now turning to Figures 8 and 9, the curtain 80 there shown is a standard smoke or fire carrier curtain. When not deployed the curtain is wound on a roller 82 or lifted in a pleated motion, provided in a headbox 84, mounted in a ceiling 86. The curtain is also provided with a bottom bar 88 to weight the curtain. When the curtain is drawn up onto the roller or lifted by a roller the bottom bar 88 can be used to close the headbox 4. When deployed the bottom bar provides stability to the descending curtain, and in deployment rests on the floor, where it may be locked into position, typically by electromechanical latches (not shown).

Side guides are also provided to retain the side edges 90 of the curtain. They extend from the head box on the ceiling to the floor, and may have in-turned lips (not shown) between which the edges of the curtain 80 are held or encapsulated, typically with projections provided along side edges of the curtains (not shown).

According to the invention the bottom section of the curtain in use is provided with additional insulation or protection. This is in the form of additional layers 92 of material. The additional layers will also be held within the side guides as the main curtain 80 is, as described above. This additional insulation will be provided to heights of typically 1 100mm or 100mm. Fires often starts at low level spreading to higher levels once established. Thus the lower sections of the curtain are likely to, initially at least, be exposed to the highest temperatures, thus it is these sections which are provided with addition insulation. While the additional insulation may be provided at any level, for example 1-1.5m from the bottom, the height at which the insulation starts will be determined by the position of the curtain and surrounding area. 1100mm is the height of a crawling person, and thus additional insulation from this level results the less heat passing though the barrier at lower levels, extending the period during which persons can evacuate the building. Alternatively some floor coverings, such as caipet compromise the fire protection provided by the barrier, being flammable materials. Thus the provision of additional insulation at very low levels, typically less that 0.5m and in particular at 100mm provides additional protection to the curtain barrier.

The additional insulation is provided in the form of additional layers of curtain material. This may be the same material from which the main curtain 80 is made, or may be other fire proof, flexible material. For example the additional layers may be of a material which include intumescent materials, including a woven glass fibre, perhaps stainless steel reinforced, fabric with an intumescent graphite silicone elastomer, such as EFP™ 2/1000/BI or EFP™ 2/1000/DGI available from Coopers Fire Limited, Havant, England.

The additional insulation may be a single layer provided on one side of the curtain, or two or more layers provided on one or both sides of the curtain. As shown in Figure 9, three additional layers are provided, one 94 on one side, and two 96, 98 on the other side. However, additional or fewer layers may be provided.

The invention is not intended to be restricted to the details of the above- described embodiment. For instance, the sample could be something other than intumescent fabric. The sample could be glass, to which the adherence of a thermocouple is not satisfactory.