Two numbers, pulling opposite ways
Every tray run gets settled by two numbers that want opposite things. Fill says how much cable the cross-section will carry before the code, the heat, or the next revision of the cable schedule says stop. Support spacing says how far apart the hangers can sit before the side rails sag, the splice plates start working, and a straight run turns into a skipping rope. Raise the first and the second has to come down. Stretch the second and the first gives ground. There is no third number that rescues you from that trade.
Drawings usually hand over one of them and leave the other implied. A note reading 24 in (600 mm) ladder tray, supports at 12 ft (3.6 m) has already committed to a weight per foot whether or not anybody calculated it. Cable schedules grow during a job — they always do — so fill creeps up quietly while the support spacing stays exactly as drawn, and the failure shows up months later as a visible dip at the third hanger off the riser.
Referee the argument at four moments: takeoff, before the first hanger is fixed, at the halfway mark of pulling, and at handover. Miss the second one and the fix is no longer a calculation, it is a scaffold, a de-cable, and a re-hang through an occupied ceiling.
What fill is actually counting
Fill is not one number, it is three wearing the same name, and they bind at different points. Code fill is the legal one — in North America the cable tray rules in NFPA 70, National Electrical Code, Article 392 set it, and the method changes with what you are laying in. Multiconductor cables of 4/0 and smaller are governed by summed cross-sectional area against a value tied to the tray's inside width; larger multiconductor cables are governed by summed diameters against that width, laid in a single layer. Single conductors follow their own path. Ladder and ventilated trough behave differently from solid bottom. Outside the NEC the same job is done by BS 7671, Requirements for Electrical Installations, by CSA C22.1, Canadian Electrical Code, Part I, or by AS/NZS 3000, Electrical Installations, and the permitted arrangement is not identical between them — the standard in force at the jobsite governs, not the one you learned first.
Thermal fill binds earlier than most crews expect. Power cables bundled in a tray lose the free air the ampacity table assumed, and the adjustment for cables installed in tray is a real reduction, not a paper one. A run that is legal on cross-sectional area can still be the reason a feeder derates below its design current, and the discovery usually happens after energisation.
Install fill is the practical one: what a puller can get through a wall penetration, around a 90 degree horizontal bend, and past firestop pillows without skinning a sheath. Communications pathways are held to lower percentages than power practice — ANSI/TIA-569, Telecommunications Pathways and Spaces, sets the target for those, and the edition in force matters — largely so that the moves and adds nobody has scheduled yet are still possible.
Work the number segment by segment, not run by run. The pinch point governs: a run that averages comfortable fill and reaches capacity only through one sleeve is a run at capacity.
The fill percentage worked out here becomes the input to every decision further down the run, so it belongs before hanger centres, rod size, or tray class are chosen rather than after.
The cable tray's usable internal width.
The cable tray's usable internal depth.
The sum of the cross-sectional areas of all cables in the tray.
Cable tray fill percentage
29.7 %
Maximum allowable fill percentage varies by tray type (ladder, solid-bottom, ventilated) and cable type (single vs. multiconductor) per NEC 392.22 — verify your specific tray/cable combination's allowable maximum against the code rather than a single universal limit.
- Tray usable cross-sectional area
- 47 in²
They open the calculator with your figures already in it
Cable Tray Fill Percentage Calculator: 29.68 % — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
What this calculation does not cover
- An area percentage is not always the rule that governs. For the larger multiconductor cables, and for the larger single conductors, NEC 392.22 stops being an area test and becomes a sum-of-diameters test with the cables laid in a single layer — a tray can sit well inside any percentage and still be non-compliant because the cables were stacked.
- Fill has nothing to say about heat. Cables bundled in a tray lose the free-air rating they were picked on and derate against the number of current-carrying conductors and how tightly they lie, so a legally filled tray can still be running conductors past their allowable ampacity — maintained spacing and a single layer are the cure, and neither shows up in a percentage.
- Cross-section is not weight. Armored and medium-voltage cable fills a tray slowly and loads it quickly, so a run sitting at a comfortable fill percentage can be over the tray's rated load class long before it is anywhere near its fill limit.
Fill converted into pounds per foot
Fill converts to pounds per foot, and that conversion is the moment the argument changes sides. Take the cable schedule for the segment, pull the published weight per unit length for each cable, sum them, and add the tray's own weight. Aluminum tray is lighter to hang and lighter to handle; steel is heavier before a single cable lands in it. Neither figure is negligible on a 24 in run at length.
Add what the catalogue span rating does not cover: a concentrated load for the person who will stand in the tray while running the next system — they will, whatever the sign says — plus ice and snow outdoors, plus any conduit, cleat, or messenger someone hangs off the side rail. Outdoors and in cold rooms the accumulated environmental case rather than the cable case often governs, and the local building code names it.
Leave headroom deliberately. A run designed to its exact present load has no answer for the addition that arrives before commissioning, and the cheapest spare capacity on the whole job is bought at hanger-spacing time, not later.
| Load contributor | Where the figure comes from |
|---|---|
| Cable bundle | Published weight per unit length from the cable schedule, summed per segment |
| Tray self-weight | Manufacturer catalogue for the section, material, and rung spacing |
| Concentrated live load | Allowance for a worker in the tray during later trades |
| Environmental load | Ice, snow, or wind where the run is outdoors, per the governing building code |
| Attached services | Conduits, cleats, messengers, or shared trapeze loads fixed to the rails |
The span the tray wants versus the span the building offers
Tray is sold with a load/span class — a span paired with a working load — and the class belongs to the section you received, not to the one specified. Substituted material is the commonest way a run ends up under-rated: the spacing on the drawing was drawn for a class the yard did not ship.
Deflection limits are set by catalogue, expressed as a fraction of span, and they are not uniform between manufacturers. Steel and aluminum of the same nominal size deflect differently under the same load. Check the limit for the tray in your hands rather than the one you memorised on the last job.
The building rarely offers the span the tray wants. Bar joists at their own spacing, purlins on a module, ductwork already claiming the obvious route — available anchor points are a fixed grid, and the tray class has to meet it or intermediate steel has to be added. Three moves exist and only three: heavier tray, closer supports, or new structure to hang from. Choosing the third late is the expensive one, because it needs a structural engineer's sign-off and a crew that has left site.
Continuous runs over several supports behave differently from a single simple span, and end spans deflect more than interior ones. The last support before a riser or a piece of equipment deserves a harder look than the ones in the middle of the run.
Span answers back at this point, so put the load per foot you just built against a real centre-to-centre distance before hanger positions are committed to the ceiling grid.
The uniformly distributed load on the tray, including cables and the tray's own weight, per unit length.
The distance between adjacent tray supports.
The tray rail material's modulus of elasticity.
The tray rail cross-section's moment of inertia, a measure of its bending stiffness.
Calculated center-span deflection
0.00553 in
The centre-span deflection this tray works out to is below the L/100 limit for the support span entered shown with it — The L/100 convention NEMA VE 1 load/deflection ratings are referenced against gives it. Being under one limit is not a design. Nothing else is checked here — not the other limit states, not the connections, not the member the load arrives from.
- L/100 allowable limit (NEMA VE 1)
- 1.2 in
They open the calculator with your figures already in it
Cable Tray Center-Span Deflection Calculator: 0.0055 in — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
Add the equipment this sizes
This result is a specification — 0.00553 in — not a quantity. Put the thing it sizes into your project: how many, what you call it, and your supplier’s price.
What this calculation does not cover
- Solves one simply supported span, which is the softest case and rarely the built one. Tray normally runs continuous over several supports, and where the splices sit decides which case applies: a splice landing near mid-span puts a hinge at the point of maximum moment and multiplies this figure, while splices kept near the quarter points let the run behave as the continuous member the manufacturer's load tables were written for.
- A uniform load per meter is not what bends trays on site. Someone stands on one to reach a fitting, a pull rope drags a concentrated load along it, a heavy feeder drops into a single bay — tray load classes are tested against a concentrated load at mid-span as well as the distributed one, and no part of 5wL⁴/384EI sees any of it.
- Clearing L/100 says nothing about what holds the tray up. Every span hands its reaction to a hanger rod, a trapeze and an anchor into concrete or steel, and those are sized in a separate exercise — a tray stiff enough for the deflection limit, hung on undersized rod or shallow anchors, fails at the ceiling rather than in the middle of the span.
Where the hangers actually land
Spacing is a maximum, not a target to hit everywhere. Fittings, changes of direction, and transitions all want support close to them: horizontal and vertical bends, tees, and crosses take load into the fitting from two directions, and a fitting hung from one side only will twist.
Splices belong away from midspan. Manufacturer installation practice, of the kind described in NEMA VE 2, Cable Tray Installation Guidelines, places them near the quarter point of the span where bending is low, and the reason becomes visible the first time a midspan splice opens under a full load of cable. Expansion connectors need their own support close on both sides, and the gap they are set to comes from the manufacturer's table against the temperature range the run will see. Aluminum moves further per degree than steel over the same length.
Prove line and level with a string line before anything is loaded, and record what you find. Once the tray is full, a sag caused by one hanger that never got its second nut is indistinguishable from a sag caused by overfill, and both get blamed on the wrong trade.
Vertical runs carry their load in a different direction and need supports rated for it. A horizontal hanger turned on its side is not a riser support.
Everything above the tray belongs to someone else
Everything above the tray belongs to somebody else. Threaded rod, trapeze channel, beam clamps, and anchors are the part of the system a structural engineer will ask about, and the part most often decided by whoever had hardware on the cart.
Rod works in tension and does very little in compression — a trapeze hung on long unbraced rod will sway, and in a jurisdiction with seismic requirements it needs bracing designed under the loading standard in force, in the United States ASCE/SEI 7, Minimum Design Loads and Associated Criteria for Buildings and Other Structures, as adopted by the local building code. Post-tensioned slabs are not drilled without scanning and written permission. Bar joists get attached at panel points, not wherever the rod happens to land.
Match corrosion class through the whole assembly, not just the tray. Hot-dip galvanizing to ASTM A123/A123M, pre-galvanized sheet, stainless, and fiberglass systems to NEMA FG 1, Fiberglass Cable Tray Systems, each have an environment they belong in, and hanging a galvanized tray on plated hardware in a wash-down area means the hardware fails first. Dissimilar metals in contact outdoors are a maintenance ticket waiting to be written.
Set it empty, prove it, then load it
Set the run empty and finish it before a single cable arrives. Align it, torque the splice hardware, install the bonding jumpers, and take the inspection that needs an empty tray while the tray is still empty.
Bonding deserves its own pass. Metallic tray may serve as an equipment grounding conductor only under specific conditions and only where the system is marked and installed for it — the electrical code in force sets those conditions, and where they are not met a separate equipment grounding conductor runs in the tray. Splice plates are not automatically bonding devices; jumpers go across every joint unless the plate is listed for the purpose.
Load the tray by laying in from above wherever access allows, which is most of what tray exists for. Where cable must be pulled, rollers and sheaves carry it — rungs and side rails are not a pulling surface, and a sheath dragged across a rung edge fails an insulation resistance test that then takes a day to trace. Dress cables in the layers the fill calculation assumed: a single layer that gets piled into a bundle at the bend has moved both numbers at once.
Cable ties hold, they do not squeeze. Over-tightened ties deform insulation and defeat the spacing the ampacity adjustment assumed; on vertical runs, cleats carry weight and ties do not.
Where the conditions for using the tray itself as the equipment grounding conductor are not met, a separate one rides inside it, sized from the overcurrent device protecting the circuit rather than from the cable running alongside.
The rating of the fuse or circuit breaker protecting the circuit ahead of the equipment grounding conductor.
An illustrative scaling constant used only to produce a rough proportional estimate — not a code value.
Approximate grounding conductor sizing factor
2.3 (illustrative scaling units)
This proportional calculation is illustrative ONLY — NEC Table 250.122 defines equipment grounding conductor size using specific breakpoints (e.g. 60A→10AWG copper, 100A→8AWG copper, 200A→6AWG copper, 400A→3AWG copper), NOT a continuous formula. Always look up the required size directly from NEC Table 250.122 (or your local code) for the actual OCPD rating — never use a proportional/interpolated calculation for the real conductor size.
They open the calculator with your figures already in it
Equipment Grounding Conductor Sizing Reference Calculator: 2.25 (illustrative scaling units) — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
What this calculation does not cover
- One conductor of several that get confused with it. Table 250.122 sizes the equipment grounding conductor — the one running with the circuit back to the panel. The grounding electrode conductor out to the rod or the concrete-encased electrode comes from 250.66 off the service-entrance conductors, the main and system bonding jumpers from 250.28(D), and a supply-side bonding jumper from 250.102(C). Every one of those keys off conductor size rather than device rating, so entering a breaker rating for any of them returns an answer from the wrong table entirely.
- The overcurrent device rating is the index in the NEC and almost nowhere else. BS 7671 and the IEC-derived codes size the protective conductor from the LINE conductor — a table keyed to the phase cross-section, or the adiabatic equation run against the prospective fault current and the device's let-through energy — so the same circuit is sized on a different quantity in most of the world. A reader outside the United States will not find the breaker rating in the method at all.
How the argument gets lost
Most tray failures trace back to one of the two numbers moving without the other being told. Fill grows: a late addition of six power cables goes in on a Friday, nobody recalculates, and the run sits over its class for the rest of its life.
Support goes missing quietly. A hanger omitted where ductwork got in the way, a beam clamp on a flange it was never rated for, a trapeze shared with a plumbing line added afterwards — each one lengthens an effective span the tray class never covered. The dip appears in the same place every time: adjacent to the omission, and worse where that segment also carries a splice.
Pinch points do the rest of the damage. Sleeves and firestops squeeze fill locally, horizontal bends concentrate cable against the outside rail, and drops off one side load the tray eccentrically and roll it. A run compliant along its length and non-compliant through one 12 in (300 mm) penetration is non-compliant, and the penetration is the part that gets photographed.
Field cuts left unfiled cut sheaths months later. Ends left unbushed at equipment do the same. Neither shows up until the cable does.
Six cables added on a Friday move the correction factor as well as the fill percentage, so the question worth asking here is whether those feeders still carry what they were drawn to carry.
The conductor's tabulated ampacity before any correction or adjustment factors are applied.
The multiplier for the ambient temperature the conductor will actually operate in, from the applicable NEC temperature correction table.
The multiplier for the number of current-carrying conductors bundled together in the same raceway or cable.
Derated conductor ampacity
91.5 A
Verify both correction factors directly against the current NEC tables for your specific ambient temperature and number of current-carrying conductors — this calculator applies factors you supply, it does not look them up for you.
They open the calculator with your figures already in it
Feeder Conductor Ampacity Derating Calculator: 91.52 A — shown in imperial, US market. The link sets both, so the result they see is the one on your screen.
Add the equipment this sizes
This result is a specification — 91.5 A — not a quantity. Put the thing it sizes into your project: how many, what you call it, and your supplier’s price.
What this calculation does not cover
- Derating and terminations are two separate ceilings and the lower one governs. NEC 110.14(C) holds the conductor to the 60°C or 75°C column for whatever the breaker and lugs are listed at, so a run derated down from the 90°C column can still not be used above its 75°C table value — the number this page returns may legitimately be higher than the ampacity you are allowed to claim.
- The adjustment factor you type depends on a conductor count this page cannot verify. Equipment grounding conductors and the neutral of a balanced circuit are not current-carrying, while the neutral of a three-phase, four-wire wye feeding significant nonlinear load is — reclassifying one conductor can drop you a whole band in Table 310.15(C)(1). The adjustment also does not apply at all through a nipple of 24 in (610 mm) or less.
- A derated ampacity proves nothing by itself. It still has to cover the load it serves at 125 percent of the continuous portion, sit correctly with the overcurrent device protecting it, and survive a voltage drop check over the length of the run. This page stops one step before all three.
Closing both numbers out on paper
Close the argument on paper. Record achieved fill for each segment against that segment's rated capacity, with pinch points identified by location, so the next contractor knows what they may add before opening the ceiling.
Record support spacing as installed, with deviations noted and the reason beside them. A run at 8 ft (2.4 m) centres where the drawing said 12 ft is not an error to hide; it is evidence that somebody did the arithmetic when the cable schedule changed.
Photograph the empty run and the loaded run before the ceiling closes, test bonding continuity end to end, and label each segment with its tray designation and remaining capacity. Every one of those is cheap while the lift is still on the floor and awkward afterwards.
Spare capacity is the number an owner asks for first, and the only one that makes the next project on the same building simpler than this one.
Before the first hanger goes in
Both numbers are decided at takeoff, so bring the cable schedule and the tray catalogue to the same table and settle fill and spacing together.
- Tray class and material — Load/span class for the tray actually delivered, not the one specified — substitutions change the span you are allowed to use.
- Cable weight per foot — Published weights pulled from the cable schedule and summed per segment, before hanger centres are chosen.
- Support hardware set — Trapeze channel, rod, beam clamps or anchors matched to the deck type, plus the extra pair each expansion connector needs.
- Splice and expansion kits — Splice plates counted per joint and expansion connectors for long or outdoor runs; gap settings come from the manufacturer's temperature table.
- Bonding jumpers — One per joint unless the splice plate is listed for bonding, plus terminations into every enclosure.
- String line and level — The only way to prove line, level, and spacing while the run is still empty and reachable.
Opens the calculators above on one screen with the dimensions from this article already filled in. Quantities only — this site publishes no price list, because local prices vary too much to publish honestly.