Replacing DC motor with induction motor & vfd

[Pratik jain]

If existing DC motor rating is 1900kw, 0-265-1050 rpm, 690v, mill application, then what will be the calculation for replacing this DC motor with induction motor with vfd.
Pls note that DC motor will be having constant torque upto base rpm ie upto 265rpm, which AC motor should also deliver.

 

[edison123]

https://acim.nidec.com/drives/kbele...rive-based-on-load-characteristics.ashx?la=en

pdf link

Muthu

http://www.edison.co.in

 

[FacEngrPE]

Edison's link is spot on.
Addition - if the load is from a variable source - like a crane driving into a stiff wind, size the drive based on peak torque for the highest wind gust expected while operating. If complex controls are present some additional margin may be needed to ensure there is enough torque for the control response.

If the motor needs to deliver high torque at zero or low speed for purposes other than starting (load holding for example) encoder feedback and additional cooling provisions will be needed (Fans, Oversized frame selection, etc) consult with a Motor OEM Engineering Department.

1900kW (2500 HP) is large enough that it will be worthwhile presenting the entire requirement set to a motor OEM engineering department or a systems integrator for a concept proposal for drive and motor.

 

[Gr8blu]

First question should always be - WHAT DRIVES THE REQUIREMENT TO CHANGE?

The given speed range (1050 top, 265 base) is 3.96 to 1. This is achievable with a pretty good drive (operation at 15-60 Hz, for example). However, the ability to maintain precision control at very low speeds (below 6 Hz definitely, or about 105 rpm with the above-mentioned drive) is going to be difficult at best, and may be unachievable, depending on how close a tolerance is required. This is a reflection on the drive topology and control scheme, rather than the induction motor.

Remember that the AC drive will likely be limited to a short-time overload in the 110-150 percent of rated current range for no more than 60 seconds. Ensure that this peak corresponds to the actual peak(s) that the DC machine is currently seeing - AND what the projection might be for the near- to medium-future operations. If the DC machine sees 200% current peaks now, then the drive is going to need to be rated something higher than expected from a "general industrial" kind of application to achieve the same accelerations and transient response.

Also be aware of how often (and for how long) the peak current is held. That may also result in an uprate of the AC solution. There is a difference between "frequent" and "occasional" momentary peaks! "Occasional" means something happens as often as a catastrophic failure or Act of (insert deity here). Anything occurring more often than that is considered "frequent".

If there is extended (i.e. > 50% duty cycle period) operation in the lower end (i.e. < 500 rpm) range, the AC system will likely be LESS energy efficient than the existing DC system.

Last thing - the AC motor is going to be physically larger than the existing DC. At a guess, by at least 20% volume. PLUS it will require separately powered force ventilation (like the DC) - it will not be able to cool itself.

Converting energy to motion for more than half a century

 

[jraef]

However, the ability to maintain precision control at very low speeds (below 6 Hz definitely, or about 105 rpm with the above-mentioned drive) is going to be difficult at best, and may be unachievable, depending on how close a tolerance is required. This is a reflection on the drive topology and control scheme, rather than the induction motor.

That has not been true for 20+ years... Flux Vector Control VFDs have been around for a long time now and are just as capable (if not more so) than DC drives and motors. The latest iteration now is "Encoderless FVC" where we no longer need the shaft encoder to attain that velocity and torque precision. I have (in the distant past now) built demo units where we couple them together to show that there is nothing the DC drive and motor can do that the AC drive and motor cannot do.

Determine the torque output of the DC motor, convert that to an AC motor HP at it's base speed and apply a heavy duty (constant torque) rated VFD to that motor based on it's FLA rating.

So assuming 1900kW @ 1050RPM, that means it is 1900 x 9550 / 1050 = 17280Nm of torque. Assuming a 4 pole 1450RPM 50Hz motor, that is 17280 x 1450/9550 = a 2624kW AC motor. Nearest "standard" size is 2750kW @ 690VAC, so a VFD rated for around 3700A. Very doable in this day and age. You probably want a Line Regen Active Front End drive if it was a 4 quadrant DC drive, but if not, you at least want a Low Harmonic AFE drive.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[Gr8blu]

The base rating for the DC machine (and the process) is 1900 kW at 265 rpm, with the higher speed being reached via a "weak field" condition. Rated torque (for sizing shafts and mechanical fits, etc.) would then be 1900 x 9550 / 265 = 68 471 N.m. Given that it's in a mill application, the frequent peak torque near the bottom of the speed range is likely to be roughly 1.75x that value, or about 120 kN.m. At the top end (1050 rpm) the peak torque will be more like 24 kN.m


Edit: As to whether an AC machine or DC machine is the right one for a given job, consider the following application.
The motor rated shaft power is 2600 kW. The rated speed is 500 rpm. The mechanical design is sized for the peak torque (plus safety factor) - so roughly 300 kN.m impact load. The motor must accelerate from rest to rated speed in 150 degrees of mechanical rotation in less than 0.020 seconds. It must then reverse direction in order to reset for the next operation.
This is a case not all that uncommon in the metal rolling industry.

Converting energy to motion for more than half a century