The move to brushless motors in small devices is usually described as a technology shift, as though brushless motors became better. They did not. A brushless motor has always been more efficient and longer lived than a brushed one of the same size. What changed is the cost of the electronics that make it run, and once that cost fell far enough, the crossover point moved down into products where a brushless motor would previously have been unthinkable.

The Change Was in the Driver, Not the Motor

A brushed motor needs nothing between it and the supply: two wires, a voltage, and it turns. A brushless motor cannot turn at all without a controller that knows the rotor position and switches three phases in the right order, which means MOSFETs, a microcontroller, position feedback, and firmware. Twenty years ago that hardware cost more than the motor it drove, which confined brushless designs to applications where the performance was worth the price. Integrated driver ICs, cheap microcontrollers and back-EMF sensing have changed that. The sensorless technique in particular removed the Hall sensors and their extra wires, so a small brushless motor today is a motor with three leads and a small board, and the arithmetic works in products where it did not before.

What the Brush Interface Actually Costs

Three failure mechanisms exist in a brushed motor that do not exist in a brushless one. The brush wears, and at 0.01 to 0.03 mm per hour of running the brushes become the component that decides the life of the motor. The commutator surface wears with them, which raises contact resistance progressively rather than suddenly. And the arcing at the brush edge radiates broadband electrical noise, which matters to any device with a radio, a sensor, or an audio path inside it.

The efficiency difference is measurable and usually quoted at around ten percentage points on small frames, with the gap traceable to the brush contact, which dissipates roughly one to three percent of the input power on its own, and to friction at the same interface. Two further differences are less often mentioned but matter to a designer. Brush wear produces carbon dust inside the housing, which is unwelcome in optical and medical equipment. And the heat in a brushed motor is generated in the rotor, which is the hardest place to cool; in a brushless motor the windings sit in the stator, where the housing itself becomes the heatsink.

Where the Crossover Point Now Sits

The devices that have moved first are the ones with the most to gain from a long life and a small battery. Battery-powered appliances where run time per charge is part of the specification, handheld tools that should outlast their batteries, medical and laboratory instruments where carbon dust is unacceptable, and equipment with a radio inside it that cannot tolerate brush arcing. In each of these, the brushless premium is paid back by something the customer can see: run time, service life, or the absence of a failure mode.

The pattern is that the crossover follows the life target rather than the power level. A device expected to run a hundred hours in its life has nothing to gain from brushless, because the brushes will outlive the product. A device expected to run five thousand hours has nothing to gain from brushed, because the brushes will not. Devices sit between the two, and the driver cost is what decides which side of the line they land on.

Where Brushed Motors Still Win

Brushed motors remain the right answer more often than the trend articles suggest. They stall predictably with maximum torque at zero speed, where a sensorless brushless driver struggles to start at all. They need no firmware, no current-sense tuning, and no EMC work on a switching device. They are cheaper as a system, not just as a motor. And where the application stalls the rotor regularly and deliberately, or runs from a rough supply, the simplicity is worth more than the efficiency.

There is also one limit that catches designers out. On a brushless motor with an integrated driver, the electronics set the maximum ambient temperature, and it is usually well below what a brushed motor with the same insulation class will survive. A hot enclosure is a reason to stay brushed.

What Switching Actually Changes

Four things change in the design, and they are engineering tasks rather than purchasing decisions.

The motor is no longer the whole system. The driver becomes part of the thermal and electrical budget, the peak current the supply must deliver changes, and the voltage ceiling is now set by the electronics rather than by the winding.

The EMI signature changes rather than disappearing. Brush arcing produces broadband noise across a wide frequency range; a brushless driver produces narrowband noise at the switching frequency and its harmonics, superimposed on fast switching edges. That is often easier to filter, but it is not free, and conducted emissions still have to be dealt with.

Thermal design moves. Heat now enters the stator, so the mounting and the interface to the enclosure become part of the cooling path, and an insulated mounting that was harmless before becomes a thermal problem.

And the starting behaviour has to be specified. A sensorless driver cannot know the rotor position at standstill, so it has limited starting torque at low speed. Where the load is heavy at startup or the motor has to hold a position from rest, Hall sensors or a sensored driver are the correct choice and should be specified at the outset rather than discovered during commissioning.

What to Send Us

The duty cycle and the expected total running hours, the supply voltage and its tolerance, the torque required at startup and at running speed, the ambient temperature inside the enclosure, and whether the device contains a radio or a sensitive sensor. Those five answers decide whether brushless earns its premium on this application, and which of the six families is the right starting point. The engineering team can quote both a brushed and a brushless option for the same load so the comparison is made on total cost rather than on motor price.