Can the motor hold this duty cycle?

A motor fails thermally when the heat generated in the winding exceeds the heat the assembly can move out to the surrounding metal and air. Copper loss dominates, and copper loss goes with the square of current. Because current in a permanent-magnet DC motor is proportional to torque, squaring and time-averaging the torque over one cycle gives a single equivalent torque that heats the winding the same amount as the varying real load. That is the root-mean-square torque, and it is the number this screen is built on.

The squaring is the reason a short, hard segment matters far more than its duration suggests. A motor that runs at 5 mN·m for nine seconds and 20 mN·m for one second has a mean torque of 6.5 mN·m and an RMS torque of 8.5 mN·m. Sizing to the mean would be wrong by a third, in the direction that burns windings.

RMS averaging has one hard precondition: the cycle has to be short compared with the thermal time constant of the motor. A small DC motor's winding reaches temperature in tens of seconds; the housing and the mounting metal take many minutes. If a cycle runs for ten minutes on and ten minutes off, the winding is at steady state long before the rest period arrives, and the RMS figure is meaningless. In that case the correct screen is the worst continuous segment on its own, which is why this tool reports it and warns when the cycle gets long.

Peak torque is a different question with a different answer. It is not a heating limit but a mechanical one: gear teeth, the output shaft, the bearing support and the coupling. A peak that is inside the momentary rating and a peak that is inside the continuous rating are two separate checks, and passing the RMS screen says nothing about the first.

What the screen omits is as important as what it computes. It has no thermal model. It does not know your mounting: a motor bolted through a face to an aluminum bracket runs far cooler than the same motor clamped in a plastic housing, and the difference can be tens of degrees. It does not know your enclosure or your airflow. It does not include iron loss, brush friction, bearing drag or gearhead drag, all of which produce heat that never appears as output torque. It does not derate for ambient, because a derating curve that has not been measured for the specific assembly is a guess with a graph around it.

So this screens a cycle. It does not qualify one. A cycle that fails the screen is very likely to fail on the bench and should be fixed on paper first. A cycle that passes the screen has earned a thermal test: run the real mechanism, at the real ambient, until the case temperature stops rising, and log current through the whole cycle. That measurement is the evidence. The number on this page is only the reason to go and get it.