VFD Regenerative Braking

[JRoss 84]

I ran across a customer interested in using regenerative braking to reclaim electricity and produce a greener operation. Here's the scenario. They're running a carriage with a VFD, using airbrakes to stop the rather heavy load. They want to switch to using regenerative braking instead of airbrakes. Then, rather than just burning the energy produced in a resistor bank, they want a way to save it or return it to the grid. That way they can end up using less electricity, saving money. Has anybody ever heard of doing this, and/or have recommendations of resources for this type of endeavor? The only thing I have off the top of my head is a Design News article from some years ago where this was done by Allen-Bradley on a military vehicle as a special project.

[Leitmotif 0]

ASSUMING you have "standard" VFD where incoming 3 phase is rectified to obtain DC for DC bus: 1. You cannot pass power back thru the rectifiers to the AC line 2. You CAN tie the DC buses of two VFDs together such that when one is regen and other motoring the regen will generate power to the motoring VFD DC bus. 3. With some tinkering and calculation you could in theory use regen braking resistor to provide heat to something else ie office air or hot water or (??) Dan Bentler

[paulengr 44]

Most VFD's these days automatically do regenerative braking. Sometimes you just have to turn it on. Inside the VFD you will typically see 6 IGBT's or IGCT's or equivalent (or sometimes more) on the inverter side of things. This converts power from DC to AC to drive the motor. On the front end (the rectifier part), you will see either 6 SCR's or 6 IGBT's/IGCT's. On an SCR front end, it is non-regenerative and can't drive power "backwards". On an IGBT/IGCT front end however, it can do just that. On common DC bus systems, the inverter (or equivalent for DC drives) and the front end are two separate pieces. An "IGBT" or "IGCT" front end is usually called an "active" front end. Frequently these systems are useful where you have multiple motors and some motors are in "regen" while others are in "motoring". The common DC bus is more efficient than individual drives because there are no losses in converting from DC back to AC and then AC to DC again from one motor to another. At one time, the cost of an IGBT or an equivalent component (such as a GTO with the associated driver) was very high relative to the cost of simple SCR's (SCR's are slightly more costly than diodes). A regenerative drive was usually roughly twice the cost of a non-regenerative drive. These days, that cost differential has dropped to almost nonexistant. Most of the time with a current model drive, it can ONLY be ordered in the regenerative style. Now, the downside. Mechanical braking is bar none the fastest way to decelerate. Second best is dynamic braking with a resistor bank. Third best is DC braking on AC motors (putting a DC current on the fields)....warning, this puts a lot of heat stress on a motor. Fourth best is regenerative braking. Fifth best (worst) is coasting to a stop. Fortunately, you can do regenerative braking AND dynamic braking simultaneously. The drive can first try to stop purely with regenerative braking. If it can't stop fast enough, then it can go ahead and dump the extra power out a resistor grid while simultaneously using regenerative braking. Typically if you implement this and you have a load that can move without power (such as an elevator or a crane), then you still need mechanical brakes. But if implemented correctly, the drive first fluxes up the motor, then releases the brakes, moves the motor, and then engages the brakes while the motor is still holding the load. If implemented this way, the brakes virtually never wear out. Furthermore if it has a "torque proving" mode (I know at least the AB Powerflex 755 has this), it can actually go into a fail safe mode where it gradually lowers the load in the event that the mechanical brake system fails. Whether you can do away with the mechanical brake entirely depends on two factors: 1. How fast your deceleration requirements are. If you need sudden, machine-destroying instant "jerk to stop" braking, then regenerative braking won't help much. 2. Whether the load can still move under gravity or an equivalent force when the drive shuts off (and whether this amount of motion matters). Normally, the end user's requirements are vague enough that I can get away with doing braking mostly with regenerative mode. I usually buy a relatively small resistor bank anyways because the cost is minimal and it gives me options to stop quicker. I normally try to buy Stearns brakes because they are incredibly simple and reliable and then drive them off an output from the PLC that only triggers when the motor drops out of "run" mode. I normally just put the drive into flux vector mode (better control than volts/hertz) unless positioning or stopping are critical, in which case I buy a very rugged encoder and a brake (if needed), and operate the drive with full vector control. This is the same as a servo drive (other than using a general duty motor), but you only go down that road if you need full blown motion control.

[JRoss 84]

Thanks Paul, very informative and helpful. After I posted, I was doing some web searching and found some drives with the regen option. But this help me flesh it out a little bit. In this case, the load is a rig that travels horizontally along a track. I'll have to find out about the stopping requirements, but I like the idea of combining regen and dynamic.

[dellae 0]

Regen (4 Quadrant control) - Look at ABB, they have being doing it for a number of years. They have an ACS800-17 (WWW.ABB.com/drives) range designed to do exactly what you are talking about, also used on crane/hoist systems. I have done it on a timber carriage (150kW), currently involved in a crane upgrade in a steel mill using 90kW. ABB have inbuilt firmware for all the common applications so setup is relatively straight forward. It is much more efficient than mechanical braking. The brake limit is the power of the motor connected to the load. Whatever power your motor can output (say 100kW) is the same as it can absorb.