VFD switching during regenerative braking

[bill318]

Hello All,

I have been researching VFDs and regenerative braking on 3-phase induction motors and thus far have been able to find only basic information on how a VFD and motor interact to provide regenerative braking.

From what I have been able to find so far is that the VFD drives at a lower frequency than the free-run frequency of the motor causing a reverse current flow in the VFD and back to the DC bus.

After a search on this forum, I located the following thread, but I am looking for more detailed information...

http://www.eng-tips.com/viewthread.cfm?qid=17968

Is the VFD just supplying the magnetizing current while capturing the torque current supplied by the motor during regen? From what I understand, during normal drive conditions the magnetizing current lags the voltage by 90 degrees and the torque current is in phase with the voltage. Under regeneration I'm not sure what the phase relationships are.

Is an external speed sensor required to keep the VFD frequency below the motor frequency or can a VFD subtract the drive signal from the motor current waveform to determine the motor speed?

Are there any good sites or information available that show the phase relationships between the VFD and the motor currents/voltages that would explain what is actually taking place during regeneration?

-Bill-

 

[sreid]

Normally we think of an induction motor as supplying shaft power but it can easilly absorb shaft power. This commonly happens, for example, when a crane is lowering a load; the motor is turning one direction but it has to create torque in the opposite direction. For this condition the motor rotor turns faster than the stator frequency and power is returned to the supply. This is called an over-running condition.

When a VFD is decelerating an induction motor the same condition applies and the rotational energy in the motor rotor/load is returned to the power supply. This energy pumps up the power supply capacitors which is usually dumped with a regen clamp.

 

[jraef]

There have been a lot of recent advances in regen AC drives and I am not up on the latest. From my last exposure, regenerative VFDs work(ed) fairly simply. When you wanted to stop, the VFD output freq. was driven to a very low level, say 10-15Hz. At that level the excitation was still present, but virtually any continued rotation was now exceeding the "synchronous" frequency applied to the motor. The load inertia becomes the prime-mover, and with excitation the AC motor rotating over base speed is an induction generator. The faster it is turning, the more negative torque it converts to power. That energy is converted by the VFD back-end transistors to DC on the bus. Then a separate set of transistors on the front-end of the VFD reconverted that DC into a fixed frequency voltage applied back to the supply source. Once the motor speed goes below that threshold output frequency, the VFD switched over to DC injection braking to finish the job because there was usually very little energy left in the load. You don't need to know the rotor speed for absolute sure, but I believe newer versions with Open Loop Vector control do it anyway (since they already can) for better control.

Some VFDs are set up to allow their DC link to be tied to the DC links of other VFDs so that the regen power from one motor can be used as motoring power for another.

This company makes aftermarket regen modules to apply to existing VFDs. Nearly identical idea except they do not track anything, they just monitor the VFD DC bus voltage, which means you must manually program the VFD to go to a low freq. output (instead of Off), and DC inject separately from the Regen module. They still work fine though because once you establish the desired braking time, you simply set your Decel rate a little faster than that. It almost always means the commanded frequency is lower than rotational speed. They used to have some good application information, but it appears to be undergoing some changes and I couldn't access it.

http://www.bonitron.com/drive_site/line_regen.htm

"Venditori de oleum-vipera non vigere excordis populi"

 

[bill318]

Thank-you for the replys. The application I am interested in is for an electric vehicle. I understand how regen for a DC motor and controller works, but the 3-phase AC motor regen is another story. I basically wanted to understand what was going on and "how it was done" inside the controller and if feedback signals other than current were required from the motor.

I know that Siemens makes a 3-phase motor controller with regenerative braking for EVs, but they do not go into any detail on how it works.

Being able to dynamically change the braking rate and perhaps capturing the brake energy in a capacitor bank to use when the vehicle accelerates again. The capacitor bank should be able to absorb large currents much better than the battery bank alone.

The VFD was the closest thing I could find that had a similar application.

-Bill-

 

[mc5w]

If you look closely at the PWM bridge that drives the motor there are inverse connected diodes that permit regenerated energy to flow back into the DC bus. Anytime that a PWM drive is powering a reactive load such as a motor the drive needs these even though the net power flow is into the motor.

When regenerating, the PWM transistors and the motor inductance acts like a a boost converter that pushes most of the power flow through the diodes. This is not altogether differnet than when a PWM drive for a DC motor is reconfigured to act as a boost converter to boost the voltage of the DC motor to match the DC traction power mains during regeneration.

Mike Cole, mc5w@earthlink.net

 

[busbar]

 
Another search term for regenerative VFDs is "four-quadrant" drives. On feature needed for power to reverse are SCRs and gating replacing the input-diode bridge.

 

[ruggedscot]

You have to really watch this believe me....

If you are using high power drives then there is a very good chance that you will cause trouble with the power co by putting power back to the utility. Power Co's dont like this and if you have an HV feed through Power Co protection you can find that it will reverse power trip the breaker. This could cause a lot of trouble as you would find that the motor you are trying to slow down would then go freewheel.

Also you can find that if you try to drop speed too quickly you can raise the dc bus voltage above the protection level of the drive and same thing happens it cuts out.

You can get round this by using some form of power absorber like a high power resistor to try and sink the power.

Rugged

 

[jraef]

ruggedscot,
You know, I never thought about the reverse feed sensing capability of some breakers (and Feeder Protection Relays too for that matter) and the fact that they would not allow regen. This is the hot topic in a lot of industries because in high duty cycle or high inertia applications people do not want the resistor bank that would be necessary for effectiveness. But if they can't regen back to the utility, they would not have a choice!

"Venditori de oleum-vipera non vigere excordis populi"