Electrical
Cable pull planning and the tension card for feeders and MV
Plan the pull, calculate the max tension and sidewall pressure, and hand the crew a tension card so the feeder goes in without damage you cannot see.
Direct answer
A cable pull plan calculates the maximum pulling tension and sidewall bearing pressure before the pull, because a pull that exceeds the cable's tension or sidewall limit damages insulation and shielding you cannot see, and it fails later in service. The cable manufacturer's instructions, ICEA, and IEEE guidance set the limits.
Key takeaways
- Max conductor pulling tension on a pulling eye is about 0.008 lbf per circular mil for copper, 0.006 for aluminum, summed across conductors.
- Sidewall bearing pressure equals tension out of the bend divided by bend radius in feet; common jacketed limit is near 300 lb/ft.
- Jam ratio (1.05 x conduit ID / cable OD) between about 2.8 and 3.0 is the danger zone where three same-size cables wedge and stall.
- Each 90-degree bend multiplies incoming tension by about 1.37 lubricated versus about 2.19 dry, so pull toward the bends.
- Stop the puller on any dynamometer spike and find the cause; the cable manufacturer's cut sheet governs all limits.
Why you plan the pull before you pull it
A cable pull plan is the set of numbers and the setup you work out before the cable comes off the reel: the maximum tension the conductor can take, the sidewall bearing pressure at every bend, the conduit fill and jam ratio, the pull direction, and the rollers and grips that hold it all inside the limits. You produce that plan as a tension card the crew pulls against.
You do it before the pull, not after, because the damage from an overpull is invisible. Stretch a conductor past its tension limit and you neck the strands and tear the insulation bond. Crush a cable at a bend past its sidewall limit and you flatten the insulation and the shield against the conduit wall. Neither shows on the outside. The jacket looks fine. The megger may even pass on day one. Then the weak spot sits in the pipe under voltage stress and lets go months or years later, usually in service, usually at the worst time, and the fault is buried in a conduit you cannot get to without tearing the run apart.
On a data center job the runs are long, the conduit banks are dense, and the bends stack up between the utility room and the distribution gear. That is exactly the geometry that builds tension and sidewall pressure fast. The pull that goes in smooth because somebody calculated it and set the rollers is the pull nobody ever thinks about again. The one that gets yanked in on a hunch is the one that comes back as a megafault after the building is occupied.
Maximum allowable pulling tension
The maximum tension you can put on the conductor is set by the metal, not the puller. The common rule of thumb for the conductor itself is about 0.008 times the conductor area in circular mils per conductor for copper, and about 0.006 for aluminum, summed across the conductors you are pulling on. So three 500 kcmil copper conductors with a pulling eye work out to roughly 0.008 times 500,000 times 3, around 12,000 lbf. Verify the constants and any cap against the cable manufacturer's pulling instructions and ICEA guidance, because the published limit on the cable can be lower.
How you attach to the cable changes the limit. A pulling eye factory-attached to the conductors, or a field eye made up on the bare copper or aluminum, pulls on the metal and gets the full 0.008 or 0.006 figure. A basket grip, the woven mesh sleeve that grabs the jacket, does not. It pulls on the jacket and the outer layers, so the tension is capped well below the conductor figure. Common practice limits a basket grip to roughly 1000 lbf per grip, or to a value the jacket can take without tearing, whichever is less. The grip manufacturer and the cable manufacturer give the real number.
There is also a count factor. With more than three conductors pulled together, the load does not share evenly, so the published guidance commonly knocks the total down, often by a factor around 0.8. The point underneath all of it is the same. The pulling eye lets you use the strength of the metal. The basket grip protects the jacket and pays for it with a much lower ceiling. Pick the attachment for the tension you actually expect, then size the rest of the setup so you never approach it.
Tmax = K × n × CMA- K
- Tension constant, commonly about 0.008 lbf per circular mil for copper and 0.006 for aluminum; verify against the manufacturer and ICEA
- n
- Number of conductors pulled on a common eye
- CMA
- Circular mil area of one conductor, for example 500,000 for 500 kcmil
What is sidewall bearing pressure?
Sidewall bearing pressure, SWBP, is the crushing force the cable feels as it is dragged around a bend, equal to the tension coming out of the bend divided by the radius of the bend. Write it as P equals T over R, with tension in pounds and the bend radius in feet, and the answer comes out in pounds per foot. It is not the same thing as pulling tension. A run can be well under its tension limit and still crush a cable flat at one tight bend, because the bend concentrates all that tension against a small arc of conduit wall.
Bends are where cable gets damaged, and SWBP is why. On the straight, the cable rides the bottom of the pipe and the load is spread out. At a bend, the cable is pulled hard against the outside wall of the elbow, and the tighter the radius the smaller the patch of cable taking the load. Tension that was harmless on the straight becomes a crushing pressure at the elbow. The last bend before the puller is the worst, because tension is highest there, having built up over the whole run.
The limit depends on the cable construction. Common figures are around 300 lb/ft for standard jacketed building and feeder cable, higher for lead-sheathed or interlocked-armor cable, often quoted near 500 lb/ft for lead and up around 1000 lb/ft or more for armor. Confirm the number on the manufacturer's cut sheet, because it varies by insulation and construction. With three single conductors the load does not share evenly across them, so the center cable in a cradled configuration sees more than the simple T over R figure, and the published multi-cable correction applies. When in doubt, calculate it for the worst cable and the tightest bend, and keep margin.
P = T / R- P
- Sidewall bearing pressure in pounds per foot at the bend
- T
- Pulling tension coming out of the bend, in pounds
- R
- Inside radius of the bend, in feet
How do you calculate pulling tension?
Tension builds in two ways down the run: friction on the straights, and multiplication at the bends. On a straight horizontal section, the tension added is the weight of the cable per foot times the length times the coefficient of friction, with a weight correction factor when you pull more than one conductor. You feed the tension out of one section into the next, so it accumulates from the reel end to the puller end.
Bends are the part people get wrong, because a bend does not add tension, it multiplies it. The capstan equation governs it: the tension out of a bend equals the tension into the bend times e raised to the coefficient of friction times the bend angle in radians. A 90 degree bend is pi over two, about 1.571 radians. With a lubricated coefficient of friction around 0.2, the multiplier is e to the 0.314, roughly 1.37, so every 90 degree bend multiplies the incoming tension by about 1.37. Stack four bends and the tension is multiplied by 1.37 to the fourth, more than three and a half times, before friction on the straights is even counted.
That multiplication is why pull direction matters so much and why the order of the bends matters. A bend early in the pull, when tension is still low, multiplies a small number. The same bend late in the pull, when tension is already high, multiplies a big one. You feed the cable so the high-tension bends are as few and as gentle as you can make them, and you calculate the tension section by section all the way to the puller. The final number at the puller is the one your maximum allowable tension has to beat, and the tension coming out of each bend is what feeds your sidewall pressure check at that bend.
T = wc × w × μ × LTout = Tin × e(μ × θ)- w
- Weight of the cable in pounds per foot, summed for all conductors pulled
- w_c
- Weight correction factor for three conductors in the conduit, often around 1.4 when they ride cradled in the bottom, and closer to 1.2 for a true triangular (trefoil) arrangement
- mu
- Coefficient of friction between the cable jacket and the conduit, roughly 0.5 dry and 0.2 lubricated
- theta
- Bend angle in radians, where a 90 degree bend is about 1.571
Field example: three 500 kcmil copper through a two-bend run
Take a real feeder pull. Three 500 kcmil copper conductors, roughly 1.1 lb/ft each, so 3.3 lb/ft together, pulled into 4 in conduit with a lubricated coefficient of friction of 0.2 and a weight correction factor of about 1.4 for the cradled bundle. The route runs 100 ft straight, a 90 degree sweep at a 2 ft radius, 50 ft more, a second 90 degree sweep at 2 ft, then 50 ft to the puller.
Work it from the reel. The first 100 ft straight adds 1.4 times 3.3 times 0.2 times 100, about 92 lbf going into the first bend. The bend multiplies by 1.37, so it comes out near 127 lbf. The next 50 ft straight adds about 46 lbf, reaching 173 lbf into the second bend. That bend multiplies again to about 237 lbf, and the last 50 ft adds another 46 lbf, so the cable reaches the puller at roughly 283 lbf.
Now check it against the limits. The conductor maximum on a pulling eye is about 0.008 times 500,000 times 3, near 12,000 lbf, so 283 lbf is nowhere close. The sidewall pressure at the second bend, the worst one, is the 237 lbf coming out of it divided by the 2 ft radius, about 118 lb/ft, well under a 300 lb/ft jacketed limit. The lesson on big copper is that tension is rarely what bites you. The bends, the sidewall pressure, the jam ratio, and the bend radius are what bite, which is exactly why you calculate all of them and not just the pulling tension.
| Step | Calculation | Tension |
|---|---|---|
| 100 ft straight | 1.4 x 3.3 x 0.2 x 100 | 92 lbf |
| First 90 deg bend | 92 x e^(0.2 x 1.571) | 127 lbf |
| 50 ft straight | 127 + (1.4 x 3.3 x 0.2 x 50) | 173 lbf |
| Second 90 deg bend | 173 x 1.37 | 237 lbf |
| Final 50 ft to puller | 237 + 46 | 283 lbf |
| SWBP at second bend | 237 / 2 ft | 118 lb/ft |
| Max allowable (eye) | 0.008 x 500,000 x 3 | 12,000 lbf |
Coefficient of friction and pulling lubricant
The coefficient of friction is the single biggest lever you have on tension, and it is the one number in the calculation that the crew can actually change at the pipe. Dry, the friction between a cable jacket and the conduit runs high, often around 0.5 for the common combinations, and when you do not know it, 0.5 is the conservative value to assume. A good pulling lubricant cuts that to somewhere around 0.2, sometimes lower, and because friction sits in the exponent of the bend equation, that cut compounds at every bend down the run.
Lube matters because it changes the answer, not because it makes the pull feel easier. Drop the coefficient from 0.5 to 0.2 and a 90 degree bend multiplies by about 1.37 instead of about 2.19. Run that through three or four bends and the difference is the gap between a pull that goes in clean and one that stalls or overtensions. The lube also keeps the jacket from grabbing and chattering, which is what gouges a jacket and spikes the dynamometer.
Pick the lube for the cable and the conduit, not whatever is on the truck. The jacket compound, the conduit material, and the temperature all change which lubricant works, and the wrong one can swell or attack some jackets. The cable manufacturer and the lubricant manufacturer both publish compatibility, and on MV and high-value feeders that compatibility is worth checking before the pull rather than after. Apply it generously and keep applying it as the cable feeds in, because a dry stretch in the middle of a long run is where the tension you calculated stops matching the tension you get.
What is jam ratio?
Jam ratio is the conduit inside diameter divided by the cable outside diameter for three same-size cables, and a value in the danger zone means the cables can wedge against each other and the conduit wall at a bend and stop the pull cold. The common form accounts for the slightly oval shape a conduit takes at a bend, so it is written as 1.05 times D over d, where D is the conduit inside diameter and d is one cable outside diameter.
The danger zone is a ratio between about 2.8 and 3.0. In that band three cables that traveled down the straight in a triangle get forced side by side at a bend, and their combined width is just enough to jam against the pipe like a wedge. Above 3.0 the cables have room to stay stacked and jamming is effectively impossible. Below about 2.5 they cannot line up three across, so jamming is again impossible, though now you have to check clearance instead. The trap is the middle, and 3.0 sits right at the edge of it, so a clean-looking three-in-one fill can be the one that jams.
A jam is not a soft failure. When three cables wedge, the pull does not get harder, it stops, and the next thing that happens is the crew pulls harder and either overtensions the lead cable or crushes the jammed ones. The fix is at the planning stage, not the pipe. Check the jam ratio when you size the conduit, and if it lands in the 2.8 to 3.0 band, change the conduit size or the cable configuration so it does not. This is one of those numbers you run once on paper and never have to think about again, as long as you actually ran it.
Jam = 1.05 × D / dConduit fill, and how fill, jam, and clearance interact
Conduit fill is the percentage of the conduit's cross-sectional area the conductors are allowed to occupy, and the NEC sets it in Chapter 9, commonly 40 percent for three or more conductors in a single raceway, with different limits for one or two. The fill table is a code requirement, not a rule of thumb, so it is the floor you start from. But fill alone does not tell you a pull will work, because fill, jam, and clearance are three different checks on the same geometry and a run can pass one and fail another.
Fill answers whether the cables fit by area. Jam ratio answers whether three same-size cables can wedge at a bend even when the area is legal. Clearance answers whether there is enough room left over for the cable to feed without binding, usually expressed as the gap between the top of the cable bundle and the top of the conduit. A fill that passes the 40 percent table can still have a jam ratio sitting in the danger zone, or so little clearance that a long pull binds even though it technically fits.
Run all three when you size the conduit. Start with the NEC fill table because it is enforceable, then check jam ratio for any three same-size cable pull, then confirm clearance for the bundle. The interaction is the point. Going up a conduit size to clear a jam ratio also improves clearance and drops fill, which is usually the right move on a long or heavily bent run even when the smaller pipe is code-legal on fill alone. Confirm the fill rules against the adopted code edition, because Chapter 9 and the 300-series installation rules are amended by cycle and by jurisdiction.
Planning the pull direction
Pull direction is a decision you make before anyone touches the cable, and it usually decides whether the pull is easy or marginal. The rule that holds up: pull toward the bends, not away from them, and feed from the end that builds the least tension. Because every bend multiplies the tension going into it, you want the bends to happen while tension is still low, near the feed end, not after a long straight has already loaded the cable up.
Set the puller at the end where the cable arrives with the most bends behind it pulled at the lowest possible tension. If a run is straight for a long way and then turns hard near one end, feed from the turn end and pull toward the straight, so the worst bend sees the cable when tension is lowest. If the bends are clustered at one end, that is usually the feed end. The reel should sit where the cable feeds in with slack and alignment, not where it has to be dragged up and over to start.
On a long run with a midpoint vault or pull box, you also have the option to pull in two stages or to set up a figure-eight and back-feed, which resets the tension to near zero partway through. That trades setup time for a much lower peak tension, and on a long MV pull it is often the difference between a pull you can do at all and one you cannot. The plan names the feed end, the puller location, and any midpoint, before the reel is staged.
Setup: rollers, sheaves, swivels, and the dynamometer
The setup is what keeps the real pull inside the numbers you calculated. On the straights, cable rollers carry the cable off the floor and the conduit lip so it is not dragging and gouging. At every bend and every transition, sheaves or radius blocks hold the cable to a radius at least as large as the cable's minimum bend radius, so the cable turns the corner without crushing against the conduit mouth or the pull box wall. Undersize a sheave at a bend and you have built the exact overbend the plan was supposed to prevent.
The connection between the rope and the cable carries the load, so it has to be right. A pulling eye attaches to the conductors for the full conductor tension. A basket grip grabs the jacket for the lower jacket-limited tension. Either way you put a swivel between the grip and the pulling rope. The swivel lets the rope turn without transmitting twist into the cable, because a cable that winds up under tension kinks, and a kink is a permanent bend-radius violation you cannot undo.
The dynamometer is what turns the plan into a live check. It sits in line between the puller and the rope and reads the actual tension in pounds as the pull happens, so the operator is watching the real number against the calculated maximum instead of guessing from the sound of the puller. On any pull where the calculated tension gets within reach of the limit, and on every MV pull, the dynamometer is not optional. It is the only thing that tells you, in real time, that the cable is still inside the limits you set.
What do I do if tension spikes during the pull?
If the dynamometer spikes, stop the puller. A sudden jump above the trend is the cable telling you something changed, and pulling through it is how a stall becomes a damaged cable. The number you watch is not just the peak against the maximum, it is the shape of the curve. A pull that is going right shows tension climbing smoothly and stepping up at each bend in line with the calculation. A spike that does not match a bend is a problem.
Read what the spike means before you react. A sharp jump on a straight section usually means a roller failed, the cable dropped onto the conduit lip, or the lube ran out and the jacket is grabbing. A jump at a bend can mean a sheave is undersized or seized, or that three cables are starting to jam. A steadily rising baseline higher than calculated means your friction assumption was optimistic, often a dry stretch or a wrong lube. Each has a different fix, and none of them is pull harder.
Back off, find the cause, fix it, and re-lube before you resume. If the cable jammed, you may have to back it out and change the configuration, which is miserable but recoverable, where forcing it is not. Log the peak tension you saw against the calculated maximum, because that number is part of the record that says the cable went in within its limits. The whole reason the dynamometer is in the line is so that decision is made on a reading, not on a feeling in the rope.
The pre-pull checklist and the tension card
The tension card is the one-page plan the crew pulls against. It carries the calculated maximum allowable tension, the calculated peak pulling tension at the puller, the sidewall bearing pressure at each bend, the conduit fill and jam ratio, the chosen feed end and puller location, the attachment method, the lube, and the rollers and sheaves the setup needs. It is the difference between a pull that is engineered and a pull that is improvised, and it lives at the puller where the operator can see it.
Build the card before the reel is staged, and walk it against the actual run. The pre-pull check confirms the route matches the calculation, the rollers and sheaves are set and sized, the bends have radius blocks, the swivel and grip are rigged, the dynamometer is in line and zeroed, the lube is on hand and the right type, and the maximum tension is written where the operator pulling the trigger can read it. Anyone on the crew should be able to call the stop if the dynamometer passes the number on the card.
The card is also the start of the record. The same sheet that carries the calculated numbers gets the measured peak tension written on it when the pull is done, so the plan and the result live on one page. That page is what proves, later, that the cable went in inside its limits, which is the question that gets asked when a feeder faults and nobody can remember how it was pulled.
Medium voltage cable: the extra care
MV cable raises the stakes on every number in this guide, because the failure mode is worse and the cable is less forgiving. A 600 V feeder that gets nicked usually survives. A 15 kV or 35 kV cable with a crushed shield or a strained insulation screen carries a voltage stress that finds the weak spot and turns it into a fault, often after the cable has been in service long enough that the pull is forgotten. The insulation, the semiconducting shields, and the metallic shield all have to come through the pull intact, and none of that damage is visible from outside.
So you tighten everything. The sidewall pressure limit is the one to respect hardest, because crushing the shield against a bend is the classic MV pull failure and it is exactly what SWBP measures. The bend radius is larger for shielded cable and has to be held at every sheave and every pull box. The jacket and shield protection during handling matters as much as the pull itself, so you keep the cable off the ground, off sharp edges, and out of the sun on a hot day where the jacket softens. The dynamometer is mandatory, and so is recording the peak.
MV pulls are also where the manufacturer's pulling instructions stop being a reference and become the governing document. The published maximum tension, the minimum bend radius, the maximum sidewall pressure, and the lubricant compatibility for that specific cable override any rule of thumb in this guide. On MV, you pull the cable the manufacturer's numbers allow, you hold the radius their cut sheet calls for, and you write down what you measured. The cost of doing it by feel is a fault in an energized MV feeder, and that is not a callback, it is an outage.
Minimum bend radius limits
The minimum bend radius is the tightest the cable can be bent without damaging it, given as a multiple of the cable's overall diameter, and it applies during the pull and after, in the final routing. Bend a cable tighter than its minimum and you strain the insulation on the outside of the bend and buckle the shield on the inside, the same invisible damage that an overpull causes, just from geometry instead of force. The radius has to be held at every sheave, every pull box, every elbow, and every place the cable is dressed into gear.
The multiplier depends on the cable construction, and shielded MV cable needs the most room. Common figures run around 12 times the overall diameter for shielded or lead-covered single-conductor cable, with multi-conductor and nonshielded cables allowed somewhat tighter, often in the range of 6 to 8 times the diameter. These come from the cable standards and the manufacturer's instructions, and the NEC also sets bending-radius rules for conductors, including specific requirements for cable over 1000 V. Confirm the exact multiplier on the cut sheet and against the adopted code edition, because the construction drives it and the manufacturer can require more.
The number people miss: the minimum radius is measured to the inside surface of the bend, not to the centerline of the conduit, so a sheave or block sized to the conduit centerline can still violate the cable radius. When the manufacturer's radius is larger than the code minimum, the manufacturer governs. Size the sheaves and the pull box geometry to the cable's radius before the pull, because a radius violation at a pull box is just as permanent as one at a sheave, and far easier to create when the cable is being muscled into a tight box at the end of a hard pull.
| Cable type | Common minimum bend radius | Governing source |
|---|---|---|
| Shielded MV, single conductor | ~12x overall diameter | Manufacturer, ICEA, NEC over 1000 V |
| Lead-covered | ~12x overall diameter | Manufacturer, ICEA |
| Multi-conductor shielded | Greater of 12x individual or 7x overall | Manufacturer, ICEA |
| Nonshielded 600 V | ~6 to 8x overall diameter | Manufacturer, NEC |
What to document
A cable pull that nobody recorded is a pull you cannot defend. When a feeder faults, the first question is whether it was ever pulled within its limits, and the only thing that answers it is the record made at the time. The tension card is the start of that record, and the measured values close it out.
Record the cable and its construction, the conduit size and the run, the routed length, the number and angle of the bends, the calculated peak tension, the sidewall pressure at the worst bend, the maximum allowable tension, the lubricant used, and the actual peak tension the dynamometer measured. If the pull was staged or back-fed at a midpoint, record that too, because it changes how the tension built. The point of the record is that someone who was not there can reconstruct the pull and see that the cable went in inside every limit that matters.
| Field to record | Why it matters |
|---|---|
| Cable type, size, construction | Sets the tension, SWBP, and bend-radius limits |
| Conduit size and route | Drives fill, jam ratio, and clearance |
| Routed length | Plan distance understates the real friction |
| Number and angle of bends | Each bend multiplies tension and sets SWBP |
| Calculated peak tension | The number the pull was planned against |
| SWBP at worst bend | Where the cable is most likely crushed |
| Max allowable tension | The conductor or grip limit the pull must beat |
| Lubricant used | Justifies the friction coefficient assumed |
| Measured peak tension | Proves the pull stayed within the limit |
Common mistakes
- Pulling with no calculation at all, then blaming the cable when a feeder faults months later.
- Pulling without a dynamometer, so nobody knows the actual tension until the cable is damaged.
- Ignoring sidewall bearing pressure at the last bend, which carries the highest tension of the whole pull.
- Letting the jam ratio sit in the 2.8 to 3.0 danger zone on a three same-size cable pull.
- Using a basket grip at the conductor's full tension limit instead of the lower jacket limit.
- Sizing a sheave or pull box to the conduit centerline instead of the cable's minimum bend radius.
- Running out of lube partway through a long pull and assuming the calculated friction still holds.
- Pulling harder when the dynamometer spikes instead of stopping to find the cause.
Field checklist
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
Standards and references
The governing document for any specific pull is the cable manufacturer's pulling instructions. The maximum tension, the maximum sidewall pressure, the minimum bend radius, and the lubricant compatibility for that exact cable come from the cut sheet, and they override every rule of thumb in this guide. When the manufacturer's number is stricter than the general guidance, the manufacturer's number wins.
The general guidance sits in a few places. ICEA, the Insulated Cable Engineers Association, publishes the cable construction and installation guidance the manufacturers build on, including the tension and sidewall-pressure framework, often through the joint ANSI, NEMA, and ICEA cable standards. IEEE covers cable installation methods, with IEEE 525 for substation cable design and installation and IEEE 1185 for generating station cable installation methods, and AEIC publishes guidance on pulling extruded power cable. For conduit fill and the installation rules, the NEC, NFPA 70, sets the fill limits in Chapter 9 and the raceway and bend-radius requirements in the 300-series articles, including bending requirements for cable over 1000 V.
Treat the code parts as enforceable and the engineering guidance as the method. The NEC fill and bend-radius rules are requirements, adopted by edition and amended by jurisdiction, so confirm the exact articles against the adopted code before you cite a number on a submittal. The 0.008 and 0.006 tension constants, the SWBP limits, the friction coefficients, and the jam ratio band are widely used engineering values, not code mandates, so verify them against the manufacturer and the cable standards for the cable you are actually pulling.
Units, terms, and conversions
Cable pulling carries a handful of units that show up across the calculation, the cut sheet, and the dynamometer, and mixing them is how a clean calculation goes wrong.
Pulling tension is in pounds-force, lbf, and the dynamometer reads the same. Sidewall bearing pressure is pounds per foot, lb/ft, which is tension in pounds over bend radius in feet, so the radius has to be in feet, not inches, when you run it. Conductor size is in kcmil, thousands of circular mils, where 500 kcmil is 500,000 circular mils for the 0.008 constant. The coefficient of friction, mu, is dimensionless, roughly 0.5 dry and 0.2 lubricated. The bend angle in the capstan equation is in radians, where a 90 degree bend is pi over two, about 1.571. Jam ratio is dimensionless, conduit inside diameter over cable outside diameter with the 1.05 factor for the oval at the bend.
- Pulling tension (lbf)
- The force on the cable along the conduit, calculated to the puller and read live on the dynamometer
- SWBP (lb/ft)
- Sidewall bearing pressure, tension out of a bend divided by the bend radius in feet
- kcmil
- Thousands of circular mils of conductor area, the basis for the 0.008 copper and 0.006 aluminum tension constants
- mu
- Coefficient of friction between jacket and conduit, about 0.5 dry and 0.2 lubricated
- Jam ratio
- 1.05 times conduit ID over cable OD; about 2.8 to 3.0 is the danger zone for three same-size cables
- Minimum bend radius
- Tightest allowed bend as a multiple of cable diameter, measured to the inside of the bend
FAQ
How do I calculate maximum pulling tension?
For the conductor, multiply about 0.008 by the circular mil area per conductor for copper, or 0.006 for aluminum, summed across the conductors on the eye. Three 500 kcmil copper conductors give roughly 12,000 lbf. A basket grip caps lower, near 1000 lbf. Verify against the manufacturer's pulling instructions.
What is sidewall bearing pressure?
Sidewall bearing pressure is the crushing force a cable feels at a bend, equal to the tension out of the bend divided by the bend radius in feet, in pounds per foot. Common limits run near 300 lb/ft for jacketed cable, higher for lead or armor. The tightest bend with the most tension is worst.
What is jam ratio and why does it matter?
Jam ratio is 1.05 times the conduit inside diameter over one cable's outside diameter for three same-size cables. A value between about 2.8 and 3.0 is the danger zone, where the three cables wedge at a bend and stall the pull. Change the conduit size or configuration to get out of that band before pulling.
What do I do if tension is too high during the pull?
Stop the puller, do not pull harder. A spike means a roller failed, the lube ran out, a sheave seized, or the cables started to jam. Find the cause, fix it, re-lube, then resume. If the cables jammed, back the cable out and change the configuration. Forcing it damages the cable permanently.
Which way should I pull the cable?
Pull toward the bends and feed from the end that builds the least tension, because every bend multiplies the tension going into it. Bends pulled while tension is still low cost far less than the same bends pulled after a long loaded straight. On a long run, a midpoint figure-eight back-feed resets tension to near zero.
How much does lubricant reduce pulling tension?
Pulling lube typically drops the coefficient of friction from around 0.5 dry to around 0.2, and because friction sits in the exponent of the bend equation, that cut compounds at every bend. A 90 degree bend multiplies tension by about 1.37 lubricated instead of about 2.19 dry. Match the lube to the cable jacket and conduit.
Do I need a dynamometer for every pull?
Use a dynamometer on any pull where the calculated tension approaches the maximum allowable, and on every MV pull. It reads the actual tension in line so the operator watches the real number against the limit instead of guessing from the puller. Without it, you only learn the cable was overtensioned after it has failed.
What is the minimum bend radius for MV cable?
Shielded MV single-conductor cable commonly needs a minimum bend radius around 12 times its overall diameter, held during the pull and in the final routing. Multi-conductor and nonshielded cables allow tighter, often 6 to 8 times. The radius is measured to the inside of the bend, and the manufacturer's cut sheet governs.
Why does sidewall pressure peak at the last bend?
Tension builds from the feed end to the puller, so the last bend before the puller carries the highest tension of the whole pull. Sidewall pressure is that tension divided by the bend radius, so the last bend usually sees the highest pressure. Calculate SWBP at every bend, but check the last one hardest.
People also ask
Codes cited in this guide
This guide is written and reviewed against the published standards below. Always confirm the current adopted edition with the authority having jurisdiction.