Power Supplies and Drives for Motor Control

A variable-speed machine has two power problems sharing one cabinet, and they are usually specified by different people. The drive converts incoming power into the variable-frequency supply the motor needs. A separate low-voltage supply keeps the PLC, the HMI, the safety relays and the drive's own control logic running.

The second problem is the one that gets underestimated. The drive is a large, abrupt, non-linear load sitting on the same incoming supply as the electronics that control it, and the moments when the machine is working hardest are exactly the moments when the control rail is under most stress.

Variable frequency drive modules with a DC input

The VFD category holds drive modules rather than cabinet drives. The VFD series spans roughly 150 W to 750 W in two input flavours — a 48 V DC input version and a 230 V AC version — and the modules are built to IEC/EN 61800-5-1, the safety standard for adjustable speed electrical power drive systems.

The DC-input variants matter more than their power rating suggests. A drive that accepts 48 V DC directly can be fed from a battery bank or a solar array without first inverting to AC and rectifying back again, which removes a conversion stage and the losses that go with it. That makes them a natural fit for off-grid pumping and ventilation, where the alternative is a full inverter sized for the motor's starting surge.

Size a drive from the motor's rated power and current, then check the starting torque the load actually demands. A centrifugal pump or fan needs little torque at low speed and is easy on the drive. A loaded conveyor, a positive-displacement pump or anything with significant static friction needs full torque from standstill, and a drive chosen on running power alone will trip on overcurrent every time it starts.

Why the control supply sees the worst of the machine

Starting a motor pulls a large current for a short time, and that current produces a voltage dip across the impedance of the incoming supply. On a strong industrial feed the dip is small. On a long rural feed, a generator, or a site where several machines start together, it is not.

The control supply sits on the far side of that dip. If its input drops below the point where it can hold regulation, the 24 V rail sags, and the controller that just commanded the motor to start resets — which stops the motor, which ends the dip, which lets the controller boot and try again. Machines have been known to cycle like this for hours before anyone identifies the cause.

Two specifications prevent it. Hold-up time is the interval the supply maintains its output after its input disappears, and a supply with a longer hold-up rides through a deeper dip. A wide input range gives margin before hold-up is called on at all: a unit specified across the full universal range keeps regulating at input voltages where a narrow-range unit has already given up. Where the feed is genuinely weak, feeding the control supply from a separate circuit ahead of the drive's own isolator is worth the extra cable.

Regenerated energy has to go somewhere

When a drive decelerates a load faster than friction would, the motor acts as a generator and pushes energy back into the drive's DC bus. Bus voltage rises. If it rises far enough, the drive trips on overvoltage — or, if the protection is inadequate, the bus capacitors take the damage.

How much energy comes back depends on the inertia of the load and how quickly it is being stopped. A fan impeller or a loaded flywheel holds a great deal; a small pump holds almost none. High-inertia loads and repeated fast stops are the combination to watch.

There are three usual answers: lengthen the deceleration ramp so the energy dissipates as losses over a longer interval, fit a braking resistor that turns it into heat, or — where the cycle is frequent enough to justify the cost — use a drive that can return the energy to the supply. Decide which before the cabinet is designed, because a braking resistor needs space, airflow and a thermal cut-out that nobody planned for.

EMC: in this cabinet, the drive is the aggressor

A drive switches hundreds of volts in tens of nanoseconds, many thousands of times a second. That is an efficient way to control a motor and an equally efficient way to generate conducted and radiated interference, and the motor cable behaves as the antenna.

The measures that work are physical rather than electrical, and all of them are cheaper at the design stage than after a failed test:

• Use screened motor cable and terminate the screen with a 360° gland at both ends. A screen taken to a terminal by a short tail — a pigtail — loses most of its effect at the frequencies that matter.
• Keep motor cables physically separated from signal, sensor and communication wiring, and cross them at right angles where they must meet.
• Fit the mains filter the drive manufacturer specifies, directly at the drive, with a short low-impedance bond to the backplate.
• Give the cabinet a single well-defined earthing arrangement rather than several competing ones.

The control supply is a victim here rather than a source, but it still needs to tolerate what the drive puts on the incoming mains. Immunity is specified by test level, so if the installation is electrically harsh the supply's immunity ratings are worth reading rather than assuming.

Choosing the control supply itself

Once the hold-up and input-range questions are settled, the control supply is specified like any other cabinet supply. Rail-mounted units from the DIN rail category sit next to the controller and are the usual choice where the panel is maintained in the field. Panel-mounted units from the enclosed category — LRS, RS, HRP and the higher-power UHP and MSP families — give more watts per unit of cost and per unit of width.

One thermal point is specific to this application. The drive dissipates a percentage of its throughput as heat inside the same enclosure, and on a 2 kW drive that is a substantial number of watts. The supply's derating curve has to be read against the cabinet temperature with the drive working, not against the temperature measured on a cold morning with the machine idle.