Does VFD motor cable need to be shielded?

Yes, in practice always. The drive switches thousands of times a second with fast edges, and an unshielded motor cable radiates that across the panel and the plant. The result is corrupted fieldbus, noisy analogue inputs and occasionally damaged bearings. The shield must be terminated at both ends, over its full circumference.

When do I need a braking resistor?

When the load has significant inertia and must decelerate faster than it would coast, or when it can drive the motor, as on a hoist or a downhill conveyor. Regenerated energy raises the DC bus and the drive trips on overvoltage. The resistor turns it into heat.

Why does a VFD cause bearing damage?

Fast switching produces a common mode voltage that finds a path to earth through the motor bearings, and the resulting current pits the races. Proper shield termination, a shorter cable, an output filter, or an insulated non drive end bearing all reduce it. It shows as fluting on the bearing race and a motor that fails early for no obvious reason.

← Resources

Drives & Motion·7 min read·August 17, 2026

Six ways a drive commissioning goes wrong

Cable length, shielding, ramp times, the missing brake resistor, motor rotation, and the autotune nobody ran.

Short answer

The recurring six are: motor cable too long or unshielded, so the drive causes network faults or bearing damage; ramps left at defaults, causing overcurrent or overvoltage trips; no braking resistor on a high inertia load; motor rotation not checked before coupling; the autotune skipped; and the drive set for network control while somebody is trying to start it from the terminals.

A drive that runs on the bench and misbehaves on the machine is nearly always one of six things, and five of them are decided before anybody opens the parameter list.

1. The motor cable

The most common and the least suspected.

A drive switches its output at several kilohertz with rise times measured in tens of nanoseconds. That is a radio transmitter, and the motor cable is its antenna.

Use screened cable rated for drive output. Terminate the screen at both ends over its full circumference with a proper gland or clamp, not by twisting the braid into a pigtail. A pigtail is an inductor and undoes most of the benefit.

Length matters too. Beyond the drive's rated cable length, reflections at the motor terminals can nearly double the voltage at the windings. The manual states the limit; exceeding it needs an output reactor or a dV/dt filter.

Symptoms of getting this wrong rarely point at the drive. Corrupted fieldbus on an unrelated network, analogue inputs that drift when the drive runs, an encoder that loses counts. All of it is the cable.

2. Ramps left at the default

Ten seconds up, ten seconds down, out of the box.

On a high inertia load the acceleration ramp is too short and it trips on overcurrent. On the same load the deceleration ramp is far too short and it trips on DC bus overvoltage, because the decelerating inertia is now driving the motor and the motor is a generator.

3. No braking resistor

Following directly from the last point. If the load genuinely must stop faster than it coasts, the energy has to go somewhere.

Without a resistor the drive either trips or, on drives with a stall prevention function, quietly extends the ramp and stops when it feels like it, which is worse because it looks like it works.

4. Rotation not checked before coupling

Two minutes at commissioning, and it saves a mangled gearbox.

Uncouple, jog, confirm direction, then couple. On a pump, running backwards is often not obvious from the drive and very obvious from the process a week later.

5. The autotune

Skipped because the motor turns without it, and it does, badly. Poor torque below about ten hertz, sluggish response to load changes, and current higher than it needs to be at every operating point.

It takes a minute. Run it.

6. Command source

The one that wastes an afternoon. The drive is set for fieldbus control, somebody is pressing the terminal start input, and nothing happens.

There is usually a status word or a keypad display showing where the drive is currently taking its command from. Read that first, before the wiring is disturbed.

The commissioning order that avoids most of this

Nameplate data in, autotune run.
V/f or open loop first, uncoupled, jog and check rotation.
Couple, run slowly, watch the current.
Set the ramps against the real load, watching for trips at each end.
Switch to vector or closed loop if the application needs it, and retune.
Only then connect the network and hand control over.

Common questions

Does VFD motor cable need to be shielded?: Yes, in practice always. The drive switches thousands of times a second with fast edges, and an unshielded motor cable radiates that across the panel and the plant. The result is corrupted fieldbus, noisy analogue inputs and occasionally damaged bearings. The shield must be terminated at both ends, over its full circumference.
When do I need a braking resistor?: When the load has significant inertia and must decelerate faster than it would coast, or when it can drive the motor, as on a hoist or a downhill conveyor. Regenerated energy raises the DC bus and the drive trips on overvoltage. The resistor turns it into heat.
Why does a VFD cause bearing damage?: Fast switching produces a common mode voltage that finds a path to earth through the motor bearings, and the resulting current pits the races. Proper shield termination, a shorter cable, an output filter, or an insulated non drive end bearing all reduce it. It shows as fluting on the bearing race and a motor that fails early for no obvious reason.