Overload Protection of Inverter Driven Motor-Comments 1

[ThePunisher]

Hi all,

I have one VFD controlling 2 motors (motor are connected to cascaded pumps which operates on the same speed and on the same pipeline).

The motor is rated 2,500 HP, 4,000 V, 3 phase, 43 Hertz, 12 pole, SF = 1.0, making it rated for 430 RPM synchronous speed. The VFD output voltage is 4,160 V. The actual pump speed requirement is 350 RPM and process operators may tend to vary the pump speed depending on the process "through-put" requirements which varies on the liquid density on the pipes. The VFD output section have isolation switch for each motor feeder, with phase CTs connected to GE 339 relays. Each motor RTDs are wired to each GE 339 relay.

I do understand that when inverter duty motors are driven below synchronous speed, the volts/Hz at the motor must be maintained to keep torque capacity. This, theoretically, would require a reduction of VFD output voltage, which in turn can increase the running motor current (if the same work is maintained at the motor) However, the pump actual mechanical torque imposed on the motor should be less than 2,500 HP, that the current may still be less than the motor FLA.

Therefore, whether the motor (or VFD or the motor power cable) will get overloaded (current-wise) or not, will all depend on the amount of work (Pump Torque) imposed to it at a particular speed level. This makes the VFD and process operators given the capability to operate at any varying speed and "process output" requirements. Since the VFD will be set to the standard motor nameplate voltage and frequency...that is V/Hz setting will be 4,000/43 = 93.023, this will be kept by the VFD in relation to its internal protection and control.

So now, if the VFD's individual GE 339s protection settings were to be set, this is how I perceive it philosophically:

1. Individual GE 339 overload setting will still going to be set NOT TO EXCEED 115% motor rated FLA, regardless of VFD operation
2. Individual GE 339 overload setting will still be set to the left of the motor hot and cold damage curves, regardless of VFD operation
3. On the protection and coordination Time-Current-Curves (TCCs), the current overload curve will be set, AS IF THE MOTOR WAS STARTED DIRECT-ON-LINE
4. The motor FLA is still as per nameplate regardless how the VFD will operate and should not be exceeded to maintain rated motor life expectancy

For the VFD, it only sees the summation of motor load currents by the two motors running. Hence, the manufacturer will set the VFD based on the motors' rated V/Hz and its internal protection will be set to protect itself based on this current summation. The VFD has an inherent overload capacity at 150% for minutes of operation only and will shutdown after such time has elapsed. If the sum of the motor current drawn at 350 RPM (35 Hz), depending on pump loading, hits the VFD's internal overload region, then the motors will be stopped at a certain time delay.

If the motor manufacturer would confirm that their motor design can operate at a percentage, a bit beyond motor nameplate FLA, then motors may be driven (by process operators) at a point where the their summation hits the VFD overload region. In such case, the VFD overload capability may have to be reduced from 150% setting to a less than 150% (say 125% and the 25% taken out will be added as continuous capability). At the end of the day, the VFD have a maximum current capacity rating at 0% overload capability. On an extreme case, the VFD overload capacity will be set to 0% and the VFD can output its maximum current rating (which is significantly larger than the sum of motor nameplate FLA), allowing the process operators to more flexibility on the VFD control....and the motors' GE 339s current O/L will be set to protect the motor individually so as not to hit their damage curves.

I will appreciate any comments from experienced VFD folks here in terms on my statements on the GE 339 overload settings and the last couple of paragraphs. Of course, I will be coordinating with both VFD and motor manufacturer.

Thank you in advance.

 

[waross]

Be guided by the motor FLA.
The enemy of motors and the result of overloading is heat.
Heat is the result of I²R.
If you exceed the rated Amps the heat generated in the motor and the motor temperature you may damage the motor.
With VFDs there is an added concern, cooling.
As the motor speed is reduced, shaft mounted cooling fans turn slower and less heat is removed.
A motor running at full rated current at less than rated frequency will have less cooling from the slower turning fans and may overheat.
Technical remedies:
Protect the motor for full rated current as if it was DOL motor.
Reduce the current settings if the motor will be operated for long periods at less than rated frequency.
Or
Add auxiliary cooling fans.

If the motor manufacturer would confirm that their motor design can operate at a percentage, a bit beyond motor nameplate FLA, then motors may be driven (by process operators) at a point where the their summation hits the VFD overload region

This is called the Service Factor. If the motor is suitable for operation above rated HP, the nameplate will show SF 1.1, SF 1.15 or SF 1.25, for allowable overloads of 110%, 115% or 125%
On the CYA (Cover Your Assets)side:

On an extreme case, the VFD overload capacity will be set to 0% and the VFD can output its maximum current rating (which is significantly larger than the sum of motor nameplate FLA), allowing the process operators to more flexibility on the VFD control

If the operators want settings that will allow intentional overloads, I suggest that you ask for a signed request from the operators or their supervisor to that effect.
Then have the request approved and countersigned by your supervisor.
Centrifugal pumps drop HP and torque demands quickly at less than rated speed.
Centrifugal pumps increase HP and torque demands quickly at more than rated speed.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

MV motors are rarely built with a Service Factor, they are generally purpose built. But other than that, what Bill said is right, current is the thing to be concerned with. What a VFD does is change the speed while MAINTAINING torque, or at least torque capability. Torque and current generally follow each other, so unless your load somehow demands MORE torque as speed is lowered, motors are usually fine (cooling issues aside).

If your loads are centrifugal, as dpeed is reduced the need for torque is reduced at the cube of the speed reduction and therefore the likelihood of an overload condition drops precipitously. So even though the motor cooling efficiently may drop off, the load on the motor would be dropping faster.


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

 

[RRaghunath]

jraef,
Thanks for great explanation.
With centrifugal pumps the current drops-off faster than the speed - this is what I have come across.
Since motor cooling is affected at lower speeds, can we say the RTDs are more dependable as overload protection than the current based overload protection??
As a corollary, is it the standard practice to specify RTDs when it comes to VFD fed motors, even at Low Voltage level??

Rompicherla Raghunath

 

[edison123]

All MV motors come with embedded winding RTD's (mostly PT 100 types) as a standard. Don't VFD's also have temperature protection?

Muthu

http://www.edison.co.in

 

[jraef]

MV motors (here anyway) only have embedded RTDs if they are ordered that way. But that said, MOST people will order them that way because MV motors are typically applied to critical assets where unscheduled down time is extremely costly so the money saved by not adding RTDs is lost the first time there is a problem that they could have prevented.

Many VFDs have the ability to accept external supplemental thermal protection, such as thermistors or RTDs, but in your case you would have the same problem, the RTD inputs of the VFD, even if provided, would be useless for two motors. But you indicated that you have Multilingual 369 relays for them, so if you want the BEST protection for those motors, add the RTD input option for those 369s. What the RTD input does in a Motor Protection Relay is to bias the OL protection curve based on real temperature measurements.


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

 

[LionelHutz]

Have you checked the relay is suitable? Know the Multilin relays, I'd expect that relay is rated to work down to 40Hz.

Besides the relay, did you make sure the current transformers and voltage transformers will work both with a PWM waveform and at the frequencies you want to operate?

Even if the motor has a 1.15 service factor, don't set the overload at 115% of the motor FLA. Use FLA only. This compensates for the extra motor heating the PWM waveform causes.

Some protection relays have a VFD function where you can define a shift in the overload curve as the speed is lowered. This compensates for the loss of cooling as you slow the motor and the motor fan doesn't move as much air.

I do understand that when inverter duty motors are driven below synchronous speed, the volts/Hz at the motor must be maintained to keep torque capacity. This, theoretically, would require a reduction of VFD output voltage, which in turn can increase the running motor current (if the same work is maintained at the motor) However, the pump actual mechanical torque imposed on the motor should be less than 2,500 HP, that the current may still be less than the motor FLA.

This constant power thinking is wrong when using a VFD. The motor HP is not constant when you vary the speed of a motor with a VFD. The motor HP becomes proportional to the frequency percentage. In other words, if you run at 75% of rated frequency then the motor is now 75% of rated HP, best case.

 

[edison123]

Jeff - Given that winding insulation is basically temperature based and temperature dependent, it's odd that one had to even spec for embedded RTD's. I have never seen MV motors and generators of various OEM's including old GE, old Westinghouse, Ideal Electric etc. without winding RTD's.

Muthu

http://www.edison.co.in