Which Torque Margin Does a Servo Feeder Need at 300 SPM?
Most servo sizing arguments on a press floor end with the same sentence: the motor is big enough, it has been running for years.
That is true until the line is asked to run 300 SPM on a shorter feed, and then the same motor starts losing position on the fourth stroke of a run.
Torque margin is a number you can calculate in ten minutes, and it explains both outcomes.
The calculation below is worked on a 2.0 mm by 300 mm mild steel job, with every input written down so you can substitute your own.
Three Forces the Motor Has to Beat
A feed drive fights three things on every stroke, and only one of them is the one engineers usually quote.
Strip tension. Back tension from the loop or the brake pulls against the feed direction. On a coil line it is rarely zero, and it is highest when the coil is full.
Plucking force. The strip is clamped in the die when the pilots enter. Any drag between the feed rolls and the die adds a load the motor has to overcome.
Inertia. The rolls, the drive train and the strip itself must be accelerated and stopped inside the feed window. At 300 SPM that window is a fraction of a second.
Tension and plucking produce a steady torque. Inertia produces a peak that scales with the square of line speed, which is why a machine that is comfortable at 180 SPM can be marginal at 300.
Working the Numbers on 2.0 mm Stock
Four inputs are enough: strip section, back tension, feed length and strokes per minute.
On this job the strip is 2.0 mm by 300 mm, back tension is set at 3 kN, feed length is 120 mm and the press runs 300 SPM. Feed roll radius is 60 mm.
Peak strip velocity works out near 0.9 m/s, and the drive has about 120 ms to complete the move. That gives an acceleration around 22 m/s², or 375 rad/s² at the roll.
| Term | How it is derived | Value on this job |
|---|---|---|
| Back tension torque | 3 kN times 0.06 m roll radius | 180 Nm |
| Plucking torque | 1.5 kN measured at the die, same radius | 90 Nm |
| Inertia torque | 0.045 kg m² total, 375 rad/s² | 17 Nm |
| Peak demand | Sum of the three | 287 Nm |
| Motor continuous rating | Rated torque, not peak | 200 Nm |
A 200 Nm motor with a threefold peak capability delivers 600 Nm for a few hundred milliseconds, so the margin here is about 2.1 to 1.
Two to one is the working floor for a press line. Below 1.5 to 1 the drive spends part of every stroke at its current limit, and position error starts to appear as feed length variation.
Why the Margin Shrinks as Speed Rises
Doubling strokes per minute halves the feed window and doubles the acceleration. Inertia torque therefore quadruples while tension torque stays flat.
At 180 SPM the same job needs about 190 Nm peak, giving a margin near 3.1 to 1. At 300 SPM it needs 287 Nm and the margin falls to 2.1.
At 400 SPM the window is 90 ms, acceleration passes 60 m/s², and the peak demand crosses 400 Nm. The margin drops under 1.5 and the drive starts missing the end of the move.
This is why an NC servo feeder that performs perfectly on a 180 SPM press can be reported as "losing length" after a speed increase, with no fault code and no mechanical change.
The fault is arithmetic. Nothing broke; the demand moved past the capability.
A drive that logs a position error at the end of the stroke is describing this calculation, not a mechanical failure.
Torque Against Strip Thickness and Width
Run the same arithmetic across a thickness range and the crossover becomes visible before a machine is ordered.
| Strip thickness | Peak torque at 120 mm feed | Margin on a 200 Nm motor | What fails first |
|---|---|---|---|
| 1.0 mm | 160 Nm | 3.7 to 1 | Nothing; inertia is small, tension dominates |
| 2.0 mm | 287 Nm | 2.1 to 1 | Nothing at 300 SPM, marginal at 400 |
| 3.0 mm | 410 Nm | 1.5 to 1 | Grip, before torque |
| 4.0 mm | 545 Nm | 1.1 to 1 | Both; the machine needs larger rolls, not a larger motor |
The last column is the part a torque calculation alone will not tell you, and it is the subject of the next section.
Grip Force Usually Fails Before Torque
Torque is delivered through the roll surface, and the surface can only push as hard as friction allows.
On this job the feed has to move 4.5 kN of strip resistance. With an oily surface friction coefficient near 0.15, the rolls need 30 kN of clamping force to transmit it.
That clamping force is what marks thin or soft strip. It is also what limits heavy gauge, because the frame and the roll shafts have to carry the load without deflecting.
Roll diameter is the lever that changes the answer. Going from a 60 mm radius to 75 mm raises transmitted force by a quarter at the same clamping load.
So a bigger motor on the same head buys nothing past a point. The machine that runs 4 mm strip has larger roll diameter and a heavier frame, not simply a larger drive.
Practical check: if the drive is at its current limit while the strip is visibly slipping between the rolls, the problem is grip and the motor is not the answer.
What the Torque Figure Cannot Tell You
A torque margin is a necessary number, not a sufficient one, and there are four places where it misleads.
It ignores the control loop. A drive with ample torque and a slow position loop will still lose length. Gain, filter and encoder resolution are not in the arithmetic.
It assumes constant friction. Plucking force changes with die wear, lubrication and pilot clearance. A margin of 2.1 to 1 measured on a new die can fall below 1.5 after 200,000 strokes.
It says nothing about heat. A drive running at 80 percent of continuous rating inside a closed cabinet in a 40 degree shop derates, and the derating is not in the catalogue torque figure.
It is a single-point answer. The worst case is usually the first stroke after a coil change, when the loop is short and back tension spikes. Size for that stroke, not for the average.
Any supplier who quotes a torque number without naming the feed length, the strokes per minute and the back tension has given you a figure you cannot check.
Sizing Checks Before You Sign
Five checks turn the calculation into a specification you can hold a supplier to.
- Ask for peak torque at the roll, not motor shaft torque. Gear ratio and efficiency belong in the answer.
- State the shortest feed length you will ever run. Short feeds at high SPM are the hardest case, not long ones.
- Confirm the back tension figure the drive was sized against, and who owns the loop control.
- Check the cabinet cooling. A derated drive is a smaller motor with the same nameplate.
- Run the acceptance test at the top speed you intend to use, with your strip, for 200 consecutive strokes.
Our engineers size NC servo feeder drives against exactly these inputs, and the figure quoted is peak torque at the roll at the customer's stated SPM.
That is why the published accuracy figure of ±0.05 mm is written against a speed and a stroke count rather than standing alone.
Drive and roll data for each build sits on the coil line product index.
The wider error budget — how drive error, slip, back tension and pilot timing add up across a whole line — is set out in the coil line guide library.
What torque margin should a servo feeder have at 300 SPM?
At least 2 to 1 on peak demand, measured at the roll. Below 1.5 to 1 the drive runs into current limit on part of every stroke and feed length starts to wander.
Does a 5 kW motor always outfeed a 3.5 kW motor?
Only if the gearbox and roll diameter are the same. A 5 kW drive behind a smaller roll can deliver less strip force than a 3.5 kW unit with a larger radius.
How much back tension is typical on a 300 mm strip?
2 to 5 kN on mild steel at 2 mm. Above 6 kN the strip starts to stretch between the loop and the rolls, and length error follows.
At what feed length does inertia stop mattering?
Above roughly 400 mm at 200 SPM the acceleration term falls below 10 percent of peak torque. Below 100 mm it is often the largest single term.
How many strokes should an acceptance test run?
200 minimum, at production speed, with length measured every 20 strokes. A 20-stroke test hides the thermal and loop effects that appear after a minute of running.
Can a drive be oversized to fix a slipping feed?
No. Slip is a friction problem at the roll surface. Oversizing the motor increases the slip and the marking, because the same grip now transmits more torque.
Where to go next
Torque margin sets what the drive can do; the error budget sets what the line actually holds. The next step is to see where the remaining error comes from between the coil and the die.
Read the feed accuracy error budget
