Dear friends,

I'm trying to imagine what would happen if a centrifugal pump that's rotating reversely, at high speed, due to a leaking nrv, starts? Would the torque exerted be greater than the normal torque of the shaft?

Can we quantify this force?

MAN.

Yes, the torque exerted will be greater than normal and the motor would trip on thermal overload or instantaneous current high protection.
It can be quantified . Pete, where are you for the mathematical calculation?
Regards
Irshad

It depends on the inertia of the load. For instance, water backflowing through a pump head will have less inertia than a reverse spinning AHU fan. Also consider the diameter of the turning mass compared to the rotor.

Centrifugal pump cavities have some amount of flow-by where small leaks may not even turn the impeller. Unless starter settings are set very close to actual start-up current, they usually accomodate some amount of longer than normal starts. AHU's however, are notorious for tripping overloads when they are backspinning. In cases where engineering solutions won't stop backspinning, I've seen VFDs installed because certain models have the capability to sense backspin and gradually bring the load to a stop, and then start it up in a forward direction.

Good comments. Here are my thoughts:

The max electromagnetic torque from the motor during start is fixed by the motor torque-speed curve. It actually occurs at somehwere around 80% speed. As we decrease speed below that thru zero and into backwards rotation, the electromagnetic torque gets lower and lower. The electromagnetic torque for zero speed is higher than for reverse rotation.

In analysing the torque, there is complication of oscillation which may occur for example due to torsional resonance. But excitation of torsional resonance due to sudden application of torque shouldn't be any different for starting from standstill or reverse rotation.

If we average over the period of any torque oscillations (to remove the oscillations), then the remaining average accelerating torque should be given as:

Tshaft = Telectromagnetic * Jpump/(Jpump+Jmotor)
where J is rotating inertia (mass times square of radius).
This neglects load torque assocaited with pumping which is generally small especially during early parts of the acceleration.

Again the max Telectromagnetic is no different. So the bottom line, there is no reason to expect higher shaft torque during start from reverse rotation than during start from standstill.

If the motor trips, it would most likely be from overload / time overcurrent. The reason is not that the torque is higher, but that the time to come up to full speed (where current starts ddropping) is longer. Instantaneous trip is technically possible since current is slightly higher in negative speed region than at zero speed, but it's not a big difference (slope of current vs speed is relatively flat in that region) so I don't think it's likely.

This message has been edited. Last edited by: electricpete,

To summarize: I think the max shaft torque seen when starting a reverse-rotating pump would be about the same as seen during a normal start (just lasts for a little longer time). My reasoning is above - let me know if questions about it.

quote:

Originally posted by MAN:
Dear friends,

I'm trying to imagine what would happen if a centrifugal pump that's rotating reversely, at high speed, due to a leaking nrv, starts? Would the torque exerted be greater than the normal torque of the shaft?

Can we quantify this force?

Man, I'm not sure how to measure the torque you describe, but it is usually enough to break the shaft on pumps with a large enclosed impeller or on muti-impeller pumps. And they don't have to be back spinning all that fast. The electric motor will stop the rotation instantly and the rotating mass will snap the shaft at the first inboard bearing of the pump.