Overload Performance of VFD in Hot Rolling Mill Application 11

[Vikram 1971]

Currently using 335 KW 3 Phase Inductor motor (415V 563A 740 RPM) without any VFD in Hot Rolling Mill.
Thinking about investing in VFD to drive the above motor to enjoy saving due to VFD.
The motor operates in 20%-50% load for 50% percent of time. The current peaks up to 800-900 A for 1 seconds or less in every 10 seconds. Motor starting current is controlled via resistance shorting of slip Rings. (Motor runs idle-No load for atleast 20min in an hour)

My Question is -
1.) Whether VFD will be able to help in energy saving??

2.) I have decided a VFD (ABB Manufactured) rated for 600A. Will it be able to momentarily (less than 1 second) supply 800-900A during over loads???? What will happen in overload? will it Stop or Slow the process???

Attached the current reading during Normal operation.
Thanks in advance!!!!!!!!!
This is a big investment, I just want to be sure whether it worth it or just a marketing gimmick.

 

[FacEngrPE]

The rotor energy recovery system would require less maintenance than the LRS, that could be more important then the energy savings. For acceleration only, depending on duty cycle, a bank of grid resistors might have a lower overall cost, than a VFD. It is likely neither will pass a hurdle rate test unless you assume the existing LRS requires replacement.

Variable Frequency Transformer rotary units located at Linden NJ are a rather extreme example of an alternate use of a wound rotor motor. In this case the motor is used as a transformer, and torque applied to the shaft controls the power transfer. The frequencies on the two sides do not need to stay in lock step or even match.

 

[LionelHutz]

This isn't a very big motor. A soft-starter with a resistor and shorting contactor would work just fine.

A slip energy recovery system is more or less a specially programmed regenerative VFD so it'd likely cost more than a normal VFD for the stator. There is just no point in it unless you need speed control, which doesn't seem to be the case.

 

[Gr8blu]

Vikram - sorry for being late to the party but traveling this month has severely curtailed my online capabilities. Here are a couple of things to consider - some of which are covered in other posts in this thread.

Rolling mill duty is - usually - a constant power application that occurs at various speeds. It also - generally - requires some amount of regenerative energy dissipation. This is because the "spike" in current occurs twice: once as the bloom hits the rolls (an "overshoot" condition, which could easily be anywhere from 150-250% of rated current with a VERY fast rise time and a relatively short duration), and once as the bloom exits the rolls (an "undershoot" condition which isn't as severe a magnitude as the overshoot, but is just as rapid a rise time and is also in the regenerative quadrant). The other thing to remember about this process is that these spikes occur FOR EVERY BLOOM entering and exiting the rolls. Given that your process is - probably - a "hot" rolling process, where the bloom is heated to temperature and then processed before it cools, there will be a certain number of blooms through the rolls every so often. For a major hot mill process, this means something like one bloom every 120-150 seconds, with a "touch time" of roughly 7 second (for the first roll) to maybe 28 seconds (for the final roll). This repetitive yet highly-specific and intermittent loading is how the machine can be rated for such a relatively low steady state condition and still achieve the desired short-time overloading. There is nothing that a VFD can do for you in these instances, because the control of the process is through the application of CURRENT-ON-DEMAND.

I suspect the employed LRS assembly limits the starting (inrush) current of the wound rotor design to something like 150-200% of rated current, thereby limiting the development of accelerating torque - and thus lengthening the time it takes to get to operating speed. Here, a "full blown" VFD may be an alternative as long as it can handle the magnitude and duration of the acceleration profile, at the desired frequency of occurrence. You could also approach it as a "starter" type of application, where the drive is only sized to get the unit up to operating speed, rather than run continuously under elevated loads. In either case, you definitely need to talk to the drives people with details on the process cyclic loading profile so that the best solution can be reached. The drive you're looking for will be connected to the rotor winding circuit and have to supply power (and absorb regenerative power) - you're not going to be using the drive as a source for the machine's stator winding. The proper rotor energy recovery system would be a better solution to the rotor situation than a drive, almost certainly.

Bottom line: the VFD approach is not going to save you anything in the big picture because the magnitude and frequency of the load swings are more-or-less independent of speed. The motor already only draws the current required to run the process - all you'd be adding is the additional losses of the drive itself to the equation, which is pretty much going to be a net loss in terms of energy usage. Drives are much better suited to applications where there is truly variable torque loading - such as is found in centrifugal loads like fans and pumps - where affinity laws can be used to moderate energy usage.

Converting energy to motion for more than half a century

 

[Vikram 1971]

Sorry for such a late response.
Thanks again everyone for helping me in clearing doubts regarding VFD.
Jraef, thanks for letting me know about "slip recovery system". But as LionelHutz told - It wont help in my case and slip recovery system requires a huge capital investment as well.

Thanks Gr8blu for clearing so many things regarding rolling mill and current utilization. It really helped me and finally I understood CURRENT-ON-DEMAND process.
Thanks.