Regen Questions


Thread Starter

Michael Little

Have a couple questions regarding Regen:
1) First of all, when does it occur? It seems that a motor will go into regen only when the torque of the motor is less than the what is
required to over come the inertia of the system (such as when moving something downward and trying to stop).

2) Is regen an issue when dealing with steppers, especially when the motor is speced with large torque margin?

3) Will a system that uses an acme self-locking screw ever go into regen?

Kaufman, George

In general, a drive system may operate in all four quadrants of speed and torque. In two quadrants, the motor is actually generating or putting power back into the drive. For a more
complete discussion of regeneration, please visit and download the Handbook of AC Servo Systems (see page 35).

Best regards,
George Kaufman
Point to point comments:
1) Regen occurs when motor becomes a generator (negative power) whether by command or by overhauling load. Power = Torque x Speed. There are two torque/speed quadrants out of the four where motor becomes a generator:
A)During decel: speed = pos and torque = neg (reversed), B)An unwinder, for example: speed = neg (reversed by load) and torque = pos.
2) Torque margin not very meaningful re. regen. Whether or not a drive gives regen fault depends on several factors: dc bus trip point, dc bus level, inertia/regen energy (1/2 J w*2 neglecting friction)and dc bus capacitor(s) available energy storage (1/2 C V*2). If a drive trips on regen, dynamic braking (typically an option)needs to be added or increased.
3) Not familiar with this device.
1) You are almost on the right track. The motor is not in regen if it is putting energy into the system. It is in regen if it is removing energy
from the system. It is concievable to decelerate a load without regen.

2) Steppers motors do not regen.

3) An Acme screw will not regen, as a general rule. But it does not stop instantly either! During the short stopping time it will regen. This will drive a servo system nuts!, because it looks like a very non-linear load. Generally, when using an acme screw, don't stop faster than the load would stop by itself. This keeps you clear of the regen region.
Worm gear boxes fall into the same situation as the acme screw.

David Kane - Kane Engineering Group Inc.
CMCS (Certified Motion Control Specialist)
[email protected]
Step motors can certainly regen. When I worked at Superior Electric (Warner Electric), we either designed drives with built-in DB capability or offered it as an option. For example, during decel you are actually ahead of the next (target)equilibrium position which means negative torque as expected, and positive speed. Depending on inertia, decel rate and other factors, you may trip the drive on regen.
This is news to me! But not common, to the best of my knowledge. (Admittedly I work mostly with electric servos and servo hydraulics) Could you please provide more information? Is there additional feedback added to the motor? What kind of improvement does this provide to the system?

Thanks for the input!!

David Kane
Certified Motion Control Specialist,
Association of International Motion Engineeers AIME
> This is news to me! But not common, to the best of my knowledge

Sure it's common. With a stepper drive, just as a servo drive, if the torque and the velocity are opposite signs, power transfer is negative (as Tom Gianni pointed out above) which is to say, from mechanical power to electrical power -- this is generation.

You get into this situation any time you ask the system to decelerate faster than losses would cause it to coast to a stop.

The feedback loop in an "open-loop" stepper system is in the electromagnetic structure of the motor. When you are motoring, the field angle created by your currents is "ahead" of your mechanical angle, which creates a torque with the same sign sense as your velocity, hence motoring power transfer.

When you are generating, this field angle is "behind" your mechanical angle, which creates a torque with the opposite sign sense from your velocity, hence generating power transfer. You create this situation automatically just by requesting a deceleration.

In a "closed-loop" servo system, the feedback loop involves electronics and/or software, but the fundamental power-transfer principle is the same. Because of its external feedback, a servo system can keep the relative angle pretty closely 90 degrees ahead for motoring or 90 degrees behind for more efficient current use.

Any time you are generating with a transistor-based drive -- servo or stepper -- current is flowing through the flyback diodes, causing an increase in voltage on the DC bus and charging up the bus capacitors. This electrical energy can go several places:

1. It can be turned back in to mechanical energy the next time you start motoring again.
2. It can be dissipated in a shunt resistor.
3. It can be shoved back onto the AC supply lines with the appropriate circuitry.
4. It can blow up the drive if it is not sent anywhere and the drive is not properly protected with overvoltage fault detection. ;-)

Curt Wilson
Delta Tau Data Systems