Fixing cables to ladder and tray
When a cable needs a cleat, when a cable tie is enough and when the fixing has to be metal. Written for engineers sizing submain containment in the UK.
Regulation numbers here are quoted from manufacturer and IET reproductions of BS 7671. Check them against your own copy of BS 7671:2018+A4:2026. Cleat test figures are the manufacturers' own. This guide does not replace the project specification or the designer's fault level calculation.
Two questions decide the fixing
The first question is about fire. If the fixing burns away, does the cable stay where it is? A cable lying on a horizontal steel ladder or tray does, because the containment carries it. A cable hanging under a tray, running vertically or sitting on the side of a ladder does not, so its fixing has to be metal.
The second question is about short circuits. During a fault, single-core cables push each other apart with a force of several tonnes per metre. Only a cleat tested for that fault level is proven to hold them. Armoured multicore cables keep most of that force inside their own sheath, which is why they are normally tied rather than cleated. Answer both questions and use the stricter result.
| Cable and position | Usual fixing | FillSizer mode |
|---|---|---|
| Multicore SWA lying on a horizontal steel ladder or tray | Cable ties at regular spacing. Nylon ties keep the cables tidy and the steel containment carries them in a fire. Many specifications ask for some or all ties in stainless steel. | OD |
| Multicore SWA under a tray, on a vertical run or on the side of a ladder | Metal fixings: stainless steel ties, steel P clips, metal SWA clips or cleats | OD |
| Single-core cables, for example 4 x 1c 240 mm² XLPE/AWA | Cleats tested for the fault level, usually trefoil cleats, at a spacing worked out from the peak fault current | Cleat |
| Fire-resistant cables for sprinkler pumps, firefighting lifts and smoke fans | Fixings that last as long as the cable, normally fire-rated stainless steel clips or cleats | OD or Cleat, depending on the fixing |
Fire: cables must not fall
Firefighters died after getting tangled in fallen cables at Harrow Court in Stevenage in 2005 (two firefighters), in a warehouse fire at Atherstone-on-Stour in 2007 (four) and at Shirley Towers in Southampton in 2010 (two). At Shirley Towers the heat melted plastic trunking and the cables dropped across the flat. The coroner asked for fire-resistant support of all cables, not only fire alarm cables.
BS 7671 now requires wiring systems to be supported so that they do not collapse early in a fire. The rule first covered escape routes only (Amendment 3 to the 17th Edition, from July 2015) and since the 18th Edition it covers the whole installation as regulation 521.10.202. The notes to it, as reproduced by Prysmian and the IET, make three points that matter on containment. Cables in or on steel containment are deemed to comply, which leaves out aluminium and GRP ladder. Plastic clips and ties must not be the only support where cables are fixed straight to a surface or hang under a tray. Suitably spaced steel or copper clips, saddles or ties comply.
In fire tests by BRE the metal clips themselves held, but the cables still fell at 212 to 486 °C because the plastic wall plugs behind the clips failed. So check what holds the fixing to the building as well as the fixing.
Short circuit: single cores push apart
Current in a conductor makes a magnetic field. That field pushes on the conductor next to it. In normal running the force is small. In a short circuit the current jumps to tens of kiloamps. The force grows with the square of the current, so twice the current gives four times the force. Single-core cables in trefoil or flat formation take this force directly and try to fly apart. Cleats exist to stop that.
The force on each metre of a touching trefoil group is
F = 0.17 × ip² / S
where ip is the peak fault current in kA, S is the distance between cable centres in metres (for touching cables, the cable diameter) and F comes out in newtons per metre. The formula is from Annex B of IEC 61914 as reproduced in the CMP cleat catalogue. For the outer cables of a flat formation the factor is 0.16 instead of 0.17.
Peak current, not the rms figure on the drawings
UK designs usually quote the fault level in kA rms. The formula needs the peak, which is higher because the first half cycle of a fault is asymmetric. Ellis Patents use peak = 2.5 × rms by default. IEC 61439-1 gives lower factors of 2.0 for 10 to 20 kA rms, 2.1 for 20 to 50 kA and 2.2 above 50 kA. Agree the factor with the system designer and write it down.
A worked example
Take three 240 mm² single-core XLPE/AWA cables of 32.8 mm diameter, touching in trefoil on a ladder, fed from a board with a fault level of 50 kA rms. With the IEC 61439-1 factor the peak current is 50 × 2.1 = 105 kA. The force is 0.17 × 105² / 0.0328 = 57,142 N per metre, about 5.8 tonnes on every metre of the group. The force on each cleat is that figure times the cleat spacing.
Compare that with what the cleats were tested to. These are Ellis Patents figures from their own tests.
| Cleat | Tested force at 300 mm | At 600 mm and over | For this example |
|---|---|---|---|
| Ellis Alpha (aluminium trefoil) | 9,500 N | 15,000 N | Too weak at any spacing |
| Ellis Vulcan+ trefoil | 25,000 N | 36,000 N | OK at 300 mm. At 600 mm only just (34.3 kN against 36 kN) |
| Ellis Emperor trefoil | 51,000 N | 63,000 N | OK at 600 mm with margin |
A cable 30 m further from the board sees a lower fault current, so using the board figure errs on the safe side.
Who decides
The designer works out the prospective fault current at each board and should give it as a peak value, or say which peak factor was used. The specification (NBS or the M&E specification) names the fixing method. For single-core cables it should ask for cleats to BS EN IEC 61914 together with the maker's short-circuit test certificate and force calculation.
The cleat maker, often free of charge, or the contractor's designer then picks the cleat and its spacing. Whoever does that is the designer of that part of the installation. The installer fits what was approved. A change to a cheaper cleat or to ties needs the designer's agreement, because the cheaper part may never have been tested at that fault level.
Site practice varies. Some specifications ban plastic ties on metal containment altogether. Others accept nylon ties on horizontal runs with a stainless tie every 800 to 1000 mm. Where the specification says something, follow it.
The fixing methods side by side
| Method | Good for | Fire, used alone | Short circuit | Width beside the cable | Watch out for |
|---|---|---|---|---|---|
| Nylon cable tie | Keeping multicore SWA tidy on a horizontal steel ladder | No | None declared | Next to nothing | Softens from about 80 °C. White ties break down in sunlight, so use black UV-stable ties outdoors. Cut tails flush, as sharp tails cut the next cable. |
| Stainless steel tie (304 or 316) | Multicore SWA where the fixing must be metal | Yes | None declared | Lock head about 6 to 11 mm | A tensioning tool set too high cuts into the sheath. Some types have a plastic lock. Use coated ties against galvanised steel. |
| Steel P clip | A single smaller SWA on strut or wall | Yes | None declared | A few mm | One cable per clip |
| Strut cable clamp | Cables on strut channel | Yes | None declared | About 6 mm each side | Check the clamp range against the cable diameter |
| Steel banding | Holding a group of cables to the containment | Yes | None declared | Band thickness | Slower to fit, sharp edges |
| Polymer cleat (single) | Single cables at low fault levels indoors | No, unless backed by metal | Tested rating, lower range | About 30 mm more than the cable | Melts in a fire |
| Aluminium claw or trefoil cleat | Most LV single-core submains on ladder | Metal | Tested rating, middle range | Trefoil: one cleat for three cables | Lower melting point, so check it is acceptable on fire-fighting circuits |
| Stainless steel cleat with liner | High fault levels, outdoors, corrosive areas | Metal | Highest tested ratings | Single: about 55 mm more than the cable | Cost. Isolate from aluminium and galvanised parts. |
| Strap cleat and intermediate restraint | Large trefoil groups with fewer fixed cleats | Stainless | Tested as a system | Strap thickness | Needs the right tool and some practice |
| Fire-rated stainless cleat | Fire-resistant cables on life safety circuits | Fire tested | Often low | About 30 mm more than the cable | Low fault and weight rating on some types |
Single-core cables and cleats
A single cleat holds one cable. A trefoil cleat holds three cables in a triangle and is the usual choice for single-core submains, because the triangle keeps the magnetic fields balanced and takes far less width than three single cleats. A four-way cleat holds three phases and the neutral. An intermediate restraint is a strap that sits halfway between two fixed cleats and holds the group together without being bolted to the rung, which lets the fixed cleats sit further apart.
Choosing a cleat
- Read the overall cable diameter from the cable datasheet and check it falls inside the cleat range, including at the edge of the cable's tolerance.
- Take the fault level at the board and convert rms to peak with the agreed factor.
- Work out the force per metre, F = 0.17 × ip² / S for trefoil.
- Multiply by 0.3, 0.6 and 0.9 m to get the force on each cleat at each spacing.
- Pick a cleat whose test force at the same or a shorter spacing is higher. A stronger cleat on every second rung often costs less in total than a weaker one on every rung.
- Choose the material. Aluminium and polymer are non-magnetic. Stainless steel suits outdoors, high fault levels and corrosive areas. Life safety circuits need fire-rated cleats.
A kA rating with nothing else attached tells you very little. Two cleats both sold as "140 kA" can differ a lot if one was tested with a 35 mm cable at 600 mm spacing and the other with a 45 mm cable at 300 mm. Ask for the cable diameter and spacing used in the test. A result from a longer test spacing may be used at a shorter spacing, never the other way round. Ellis has an online cleat calculator.
Spacing along the run
On straight runs use the calculated spacing in steps of 300 mm to match the rungs. CMP keeps trefoil cleats no more than 900 mm apart and adds an intermediate restraint halfway when the spacing is between 900 and 1800 mm. At bends the cleats go every 300 mm. At a gland or joint, fit one cleat as close as possible and another before returning to the normal spacing. Leave enough room between cleats to open them fully during installation.
Steel around a single cable
A single-core AC cable must not sit alone inside a closed loop of magnetic steel, because the loop heats up from induced current. Around a complete trefoil group the fields cancel, so a steel trefoil cleat is fine there. For a single cable on its own use aluminium, polymer or non-magnetic stainless steel. Gland plates should be non-magnetic too.
Multicore SWA
In a multicore cable the cores are held together by the sheath and the armour. Prysmian, answering an engineer about 400 mm² 4-core SWA at 43 kA, said that old ERA tests showed an armoured multicore cable can be used up to its full short-circuit rating. A non-armoured cable should be limited to 25 kA for 1 second. The engineer posted that reply on electriciansforums.net. This is why multicore SWA is normally tied.
Cleat makers argue that some multicore installations still need cleats at high fault levels. There is also no test standard for cleats on multicore cable. Some specifications ask for a cleat on every cable anyway. In that case size the run in Cleat mode.
For tied cables the IET On-Site Guide table of fixing spacings, as reproduced by BEAMA, gives these maximum spacings for armoured cable:
| Cable diameter | Up to 15 mm | 15 to 20 mm | 20 to 40 mm | Over 40 mm |
|---|---|---|---|---|
| Horizontal | 350 mm | 400 mm | 450 mm | Cable maker's figure |
| Vertical | 450 mm | 550 mm | 600 mm |
Many large sites simply fix at every rung, which is every 300 mm. Most submains from about 95 mm² 4-core upwards are over 40 mm in diameter, so for them the cable maker's guidance applies.
Vertical runs and risers
On a vertical run the fixing carries the cable's weight as well, so it has to be metal. For single cores the spacing is set by whichever is worse, the fault force or the weight. On LV submains the fault force almost always governs: three 240 mm² cables weigh about 8.8 kg per metre, which is only 26 N per cleat at 300 mm. On tall risers each trefoil cleat is turned through 180° from the one below to spread the load. Prysmian recommends an offset loop on vertical runs over about 32 m and a short horizontal section of at least 1 m on runs over 100 m.
Life safety circuits
Cables feeding sprinkler pumps, firefighting lifts and smoke control fans have to keep working in a fire, so their fixings have to last as long as the cable. In practice that means fire-rated metal clips or cleats, for example CMP Solace or Ellis Phoenix in stainless steel. BS 8519 also asks for these cables to be kept apart from ordinary cables.
Two points are worth checking with the designer. Aluminium cleats, common on ordinary submains, have a low melting point and may not be acceptable on these routes. Some fire-rated cleats also have a low short-circuit rating (Ellis Phoenix is 31 kA peak at 300 mm by Ellis's own test), so close to the transformer the fire and fault requirements can pull in different directions.
How the fixing changes the ladder width
Here the fixing method feeds straight into containment sizing. The same three 35 mm cables need very different widths depending on how they are held.
| Three cables of 33 to 38 mm | Width on the ladder |
|---|---|
| Laid touching, fixed with ties | about 105 mm |
| One trefoil cleat (Ellis Vulcan+, Emperor, CMP Sovereign) | 85 to 110 mm |
| Three polymer single cleats (CMP Sabre) | 201 mm |
| Three stainless single cleats (Ellis Emperor, CMP Sapphire) | 273 to 288 mm |
A trefoil cleat is also taller than the cables it holds (95 mm against about 66 mm for the bundle), which matters when two runs are stacked.
Which FillSizer mode to pick
Use OD when cables lie touching and are held with ties, clips or bands. This is the normal choice for multicore SWA. Use Cleat when each cable sits in its own cleat. FillSizer then counts one single cleat per cable (and per core of a single-core circuit), chosen by the cable diameter from the cleat range set for the project: CMP Solace (the default), CMP Sabre, CMP Sapphire or Ellis Emperor single. Use Spaced 2x OD only when the cables are kept apart on purpose so the designer can take no grouping factor. Cleats cannot be combined with that mode. A different mode can be set on a single run, for example Cleat on the run out of the main switchboard and OD everywhere else.
Trefoil is not modelled yet. Cleat mode lays single cleats side by side, so for single cores in trefoil cleats it gives a ladder roughly two to three times wider than the real cleat width. That errs on the safe side but may oversize the containment. A trefoil option is under review.
Common mistakes
- Choosing cleats by eye, or by a kA number without the test spacing and cable diameter.
- Using a test result from a longer spacing at a shorter one.
- A cleat too big for the cable, so it is loose in a fault, or one tightened until the sheath deforms.
- Trefoil cleats fitted so close together that they cannot be opened.
- Magnetic steel around a lone single core, including steel gland plates.
- Different metals touching without isolating washers, for example stainless cleats on aluminium ladder.
- White nylon ties outdoors.
- Plastic ties as the only fixing under a tray, on a vertical run or on fire-rated cables.
- Cables tied to the outside of the ladder or to other cables.
- Metal fixings held up by plastic wall plugs.
- Sizing the ladder for bare cables when the cables will be cleated, so the ladder turns out too narrow.
Sources
- CMP Products, Cable Cleats catalogue 2016 and Cable Cleats catalogue 2023 (single cleat sizes in FillSizer)
- Ellis Patents, electrical brochure November 2024 and Black Book v3 April 2024
- BEAMA, Premature Collapse Guide 2022
- BRE, Fire performance of cable supports (published by the IET)
- IET Wiring Matters, Autumn 2015
- Prysmian, Cable Support Distances
- Panduit, cable cleats against cable ties (manufacturer test)
- Practitioner discussions on electriciansforums.net, IET Engineering Communities and Reddit