Relief Tents: Name the Catalogue Item, Not the Agency

A requisition that says UNHCR standard family tent, or IFRC family tent, reads like a part number. It is not one. Each of those agencies catalogues more than one family tent, built from different fabric, tested to different fire performance, and packed to different limits. Two suppliers can quote against the same phrase, both cite the agency honestly, and offer tents that behave differently in a fire. The requisition did not choose between them. Somebody downstream will.

What the catalogue items actually say

UNHCR publishes an item sheet for its Family Tent, item number 05353, last updated June 2014. It describes a tent for a family of five with a 16 m² main floor plus two vestibules of 3.5 m² each, 23 m² in all. The outer roof canvas is a polyester and cotton blend, roughly 50 to 70 per cent polyester, at 350 g/m² within 15 per cent; the outer wall canvas is 275 g/m² within 10 per cent. Roof canvas must reach at least 30 hPa under ISO 811, the hydrostatic pressure test for resistance to water penetration, and wall canvas 20 hPa. The master bundle may not exceed 2250 mm in length, gross weight is about 62 kg and gross volume about 0.23 m³. Fire performance is stated plainly: EN 13823 and CPAI-84, class B-s1-d0 for the outer roof and wall canvas, and class B-s1-d0 or above for the inner tent.

The IFRC and ICRC Standard Products Catalogue lists a ridge family tent, item code HSHETENTF16C, a 16 m² double fly tent with groundsheet, shipping weight 55 kg and shipping volume 0.28175 m³, the page last updated on 18 March 2019. Its specification, updated 9 December 2014, uses the same 350 g/m² roof canvas but a lighter 200 g/m² wall canvas, and the same 30 and 20 hPa under ISO 811. On fire it names no standard at all. The single fire requirement in the document is that the chimney reinforcement be heat-resistant fabric rated to a minimum of 900 °C. The catalogue page also records that this is the 2008 type and that it has been replaced by the geodesic family tent, 2018 type, item code HSHETENTF18.

The replacement is a different product, not a revision of the same one. Its unified technical specification, version 1.0, validated on 1 March 2025, describes a hexagonal 18 m² tent whose base fabric is not canvas but woven high-density polyethylene with a low-density polyethylene coating, at 170 g/m² within 10 g. It packs as one bundle, length not exceeding 1.2 m, weight 50 kg maximum. Fire performance is EN 13823+A1 at a minimum class of D-s2-d2, and the document prohibits bromine and antimony flame-retardant additives.

Three fire positions behind one phrase

So one phrase covers polycotton canvas at 350 and 275 g/m², polycotton canvas at 350 and 200 g/m², and a polyethylene sheet at 170 g/m². It covers 16, 18 and 23 m² of covered area. On the attribute where a mistake cannot be corrected later it covers three positions: class B-s1-d0, no named standard, and a minimum of class D-s2-d2.

Those classes are not degrees of the same thing. Commission Delegated Regulation (EU) 2016/364, in force since 4 April 2016, sets the classification system, and classes B, C and D are all assessed with EN 13823, the single burning item test, together with EN ISO 11925-2. A supplier who writes tested to EN 13823 has named the test and withheld the result. The suffixes carry the rest. For smoke, s1 caps the smoke growth rate index at 30 and total smoke production at 50 m² over 600 seconds, where s2 allows 180 and 200 m². For burning debris, d0 requires no flaming droplets or particles in the single burning item test within 600 seconds, while d2 is defined as neither d0 nor d1, which is to say flaming droplets are permitted. Treating a D-s2-d2 tent and a B-s1-d0 tent as alternates changes the fire behaviour of a shelter that sleeps five people.

The edition is part of the specification

The second thing that goes wrong is a good report against the wrong document. Both IFRC specifications require the whole-tent rain penetration test to ISO 5912:2003 and state explicitly that ISO 5912:2011 does not apply. ISO 5912:2011, the fourth edition, has itself since been withdrawn, and the current text is ISO 5912:2020, the fifth edition, published in March 2020 and confirmed as current in 2025. A laboratory asked for a rain test will reach for the edition on its shelf. A report against ISO 5912:2020 is a competent report against a document the catalogue item does not accept. The same trap sits behind CPAI-84, which the Advanced Textiles Association updated in 2021 for the first time since 1995. A specification written in 2014 naming CPAI-84 and a test report issued in 2026 naming CPAI-84 are not necessarily describing the same test.

Area, and the arithmetic nobody runs

UNHCR sets a minimum of 3.5 m² of covered living space per person in warm and tropical climates, excluding cooking space, and 4.5 to 5.5 m² where cold, wind, rain and snow persist for three to five months, with a minimum ceiling height of 2 m; its guidance page carries a last-updated date of 21 January 2026 and says the entry is itself under revision. Run those figures against a household of five. The ridge tent's 16 m² gives 3.2 m² a person. The geodesic 18 m² gives 3.6. The UNHCR tent's 16 m² main floor gives 3.2, and reaches 4.6 only once both 3.5 m² vestibules are counted as living space. Whether a vestibule counts is a programme decision that the item sheet does not make for the buyer, and it is the difference between meeting the minimum and missing it.

The agencies verify by inspection, not by certificate

Notice how the agencies themselves prove a tent. The ICRC quality control instruction for family tents, PU-I-014 at revision 1.4, has samples taken at random by the buyer from the delivered items and then inspected, with the buyer choosing between the ICRC quality control laboratory and an inspection company. Sampling is drawn from ISO 2859-1. Nonconformities are classed as critical at an acceptance quality limit of 0, major at 4.0 and minor at 6.5, and fibre composition is tagged critical, which is zero tolerance on what the fabric is made of. Anyone reproducing that regime should note that ISO 2859-1 reached its third edition in January 2026, replacing the 1999 edition and its amendments, so a clause still citing the 1999 text cites a withdrawn document. A mill certificate collected at order placement reproduces none of this.

That is the discipline we apply before a factory is asked for a price. We name the item, which means the agency, the item code, the specification document and its version date, so that what is quoted is one document rather than a category. We separate what the document fixes from what it leaves to the buyer, and fire performance is settled in writing first, because the ridge tent's silence on fire is not permission to supply an untreated tent into a camp. We tie every test to an edition, so reports arrive against ISO 811:2018, against the ISO 5912 edition the specification actually names, and against a fire class written out with its smoke and droplet suffixes. Then we put the verification in the order rather than in a promise: sampling standard and edition, acceptance quality limits by class, where samples are drawn, and a laboratory accredited against ISO/IEC 17025:2017 to do the testing.

What to send us

If you are buying relief shelter, send the agency and item code you are working to, or the specification document itself. Send the number of households and the climate the shelter has to hold through, the fire class you require with its suffixes or a note asking us to recommend one, the destination and whether packed length or packed weight is constrained by your transport, and the quantity and required-on-site date. We will come back with the specification read line by line, what it fixes and what you still have to decide, and the inspection regime we propose to prove it. That work happens before anyone quotes a price, because a price against an unnamed specification is not a quotation. It is a guess with a number on it.