Motor starting currents. Utility input power considerations when 3

[VLFit]

Using a Motor Test Set with many reduced voltage taps. Utility worried about inrush on a 5000 HP, 12470V motor. Since we will be starting at lower voltages and working up, there will be no full v online starting currents. When will my max current occur? Need they worry?

 

[crshears]

What % reduction from nominal voltage will you be applying when starting from rest?

Will you be testing quite often and repeatedly, or is this a one-off?

Can you incorporate soft start to minimize inrush?

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[VLFit]

The percent reduction will be the minimum needed to break over the torque to get it spinning, probably 60 - 65% of nominal.

The testing will be occasional as we rewind motors, but not everyday.

The soft start is the reduced voltage starting.

Thanks

 

[waross]

Your peak current will depend on the motor speed at which you step up the taps, or increase the voltage.
A 1760 RPM motor has a slip of 40 RPM.
At a slip of 80 RPM the current will be about 200% of FLA.
So if you go to 100% voltage at 1720 RPM on a 1760 RPM rated motor, the current will be about 200% of FLA.
At 1700 RPM, expect about 250%.
Slower than that the relationship becomes non-linear.
This is for a 1760 RPM rated motor.
At other speed ratings the current is roughly proportional to the slip up to about 250% slip.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[bacon4life]

It would worry me. Even without considering the starting currents, just the excitation losses from the motor would be a noticeable reactive load on a 12.47 kV circuit. It would drop the voltage by at least a couple of percent, and with larger impacts the farther your shop is from the substation.

Loads that operate occasionally can be extremely difficult for the utility to appropriately plan for. This would be particularly true if the motor testing happened to coincide with the utility system being in an abnormal configuration.

I typically relay on the billing system for figuring out appropriate loads to model for each customer. I fear that this load is weird enough that it would require manual intervention each time the utility runs load-flow simulations on the circuit.

 

[waross]

I heard of a major natural gas pumping station where the utility restricted starting of the very large compressor motor to something like between 2:00 am and 4:00 am.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[VLFit]

Thanks all. But one question still remaining is: if the motor under test is brought up to speed slowly, and never slammed on line, why do we care about starting currents, locked rotor currents, no load or full load currents in terms of the effect on the utility of the service is designed for the full load capability of the motor test set, in this case 1500 kva.

 

[VLFit]

Thanks Mutha. Good to hear from you

 

[bacon4life]

As a first order approximation, reactive power correlates to voltage magnitude and real power correlates to voltage phase angle. For typical 3 phase distribution circuit the voltage drop of 1500 kVAR is significant whereas the voltage impact of 1500 kW is minor.

Is it a fair assumption tat your motor would be running no load at approximately 1500 kVA @ 0.1 pf? And that reduced voltage starting would have a spick of roughly 200%?

I would care about the no-load testing because of the possible impacts to voltage. Adding 1500 kVA @ 0.1 pf is almost as much voltage change as I would expect from turning off two switched pole mounted capacitor banks. Even if the utility had installed 2 spare switch capacitors close to you, it would be challenging to align the operation of utility owned capacitors with the timing of your motor testing. For certain loads with low power factor I have required customers to install switched capacitors so that the Point of Connection is maintained at a reasonable power factor.

In many places on my 12.5 kV system just turning off the motor would cause flicker above the visibility curve in the ancient GE flicker curve. For nearly all locations in my system, turning off the motor would cause visible flicker when when being served from an alternate substation. Although outdated, the flicker curve is a baseline for starting a more in depth analysis.

For controlling the substation transformer LTC or substation voltage regulator, the designer needs to consider the power factor of the load in order to determine voltage regulator setpoints (band center, R, and X) that will work optimally. Large swings in load power factor make it more challenging to optimize voltage regulator settings.

 

[edison123]

Here are some numbers

Motor tested was 11 KV, 9.6 MW, 590 A FLC, 100 A NLC which stayed the same throughout the ramping up to speed of 1500 RPM.

Supply source 600 KVA DG via MG set at 405 V, 229 A, 147 KW, 149 KVA, 0.98 PF at the DG output side. Zero inrush amps.

Muthu

http://www.edison.co.in

 

[VLFit]

Thanks Muthu for the info. I'll try to write up today an explanation of the motor test set's function with its multiple voltage taps for gradual motor starting. They are worried about the effects on their system and need assurances.

Thanks to all for your response.

This is a great forum for us when we need help from others in our field of engineering.

Mike
HVI

 

[Gr8blu]

Ballpark estimate for full voltage line start would be that inrush is around 6.5 per unit (6.5 x full load amps) for a squirrel cage induction motor, and around 4.5 pu for a synchronous motor. You might be lucky and have a low-inrush design which would drop the currents to around 4.5 and 3.5 pu respectively.

At any given (lower) voltage tap, the inrush current will be more-or-less linear to the per unit change in voltage: this means that at 80% volts you draw 80% of the full voltage inrush current. The other factor is that the developed torque is proportional to the square of the voltage change - which means that at the 80% tap, you only get (0.8*0.8 = 0.64) pu torque. This isn't so bad once you have accelerated to some appreciable speed (50+% or so) - but can be a show-stopper at very low (< 10%) speeds.

When using an auto-transformer approach (discrete taps that result in fixed percentage of rated voltage at output), the idea is to accelerate as quickly as possible (usually), which means stepping up the output voltage at some speed point(s). Every time the tap is switched, the motor will slow down very slightly, and the switch in voltage will result in a step (upward) in inrush current.

A solid state soft starter (i.e. switched resistor bank) has far more "steps" available compared to a traditional auto-transformer, which means the inrush is more controllable and what the utility sees is more consistent. Using a VFD as the soft starter allows even more control, since it can be programmed to regulate on current - and the setpoint becomes the maximum current draw seen by the utility.

Time to accelerate will increase with reduced inrush current: there is a limit (at any current above the nameplate rating) at which the motor can safely run with some thermal damage to the cage winding.

Converting energy to motion for more than half a century

 

[edison123]

Reducing the voltage alone without corresponding frequency reduction results in

1. Torque reduction proportional square of the voltage and hence might require significant voltage to start the motor defeating the multiple tapped voltages.

2. The inrush current will still be 2 to 3 times FLC at the minimum till the motor reaches 90 to 95% of rated RPM.

A static VFD of that size will cost significantly but will keep the inrush current to NLC.

You could hitch a pony motor with a clutch / fluid coupling to bring the rotor up to speed and then close the supply breaker to the main motor and disengage the pony motor. Again, this is a pita to set up for each and every motor.

I have looked at the above options and settled on the DG-MG set route of completely electrical VFD.

One time cost, complete ownership of the system and most importantly, permanent freedom from utility insanity.





Muthu

http://www.edison.co.in

 

[Pandit1]

The starting of an induction motor at reduced voltage causes the motor to decrease its starting current and starting torque. If the driven load permits the lower starting torque, the motor can be started using a reduced voltage starting method.

When the motor is started at reduced voltage, the back emf develops and the starting current remains within limits. The best method is V/f control, which reduces the starting current while maintaining the same torque delivery capacity.

 

[crshears]

"Using a Motor Test Set with many reduced voltage taps." "The testing will be occasional as we rewind motors, but not everyday."

Based upon the foregoing from the OP, the submission

If the driven load permits the lower starting torque, the motor can be started using a reduced voltage starting method

although true, is not applicable in this instance, since the implication is that the motors tested are being run up without applied load, given that the motor in question is in their shop and not re-installed at the customer's location.

Just saying.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]