Overvoltage during a VFD start 3

[Freshsurfer]

Hi,

We're doing a commissioning procedure over a 150kW (460Vac) conveyor. This motor have a VFD.

During the start (without load)we note an overvoltage on the 460Vac MCC that feeds the VFD.

It could be a problem on the transformer? or some kind of harmonics distortion that is causing this effect?

 

[waross]

Motor voltages allow for line drop.
Standard motor voltages are multiples of 115 Volts, standard system voltages are multiples of 120 Volts.
A 460 Volt rated motor is intended to be operated from a 480 Volt supply.

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

 

[LionelHutz]

Need more details. Type of VFD. Details about this over-voltage.

I doubt it was harmonic distortion.

 

[waross]

Metering errors are common on unloaded wye-wye banks.
Been there, doe that.
I have seen it in the field.
I have seen it in the lab.
I have put the scope on it in the lab.
Metering errors are common on unloaded wye-wye banks.

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

 

[jraef]

Is the VFD tripping on over voltage, or is there a meter on the entire MCC that is indicating an over voltage condition after the VFD starts the motor?

Is it a Line Regenerative VFD?

Do you perhaps have Power Factor Correction Capacitors connected to the MCC?

Details matter...


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

 

[Freshsurfer]

Hi, thanks for your comments!

Here more details:

Motor voltage: 460Vac 50hz
MCC voltage: 480Vac

After discuss with our supervisor, we have more relevant information: the length of the cable from the VFD to the motor is about 580 meters. Due that the drive have a sinusoidal filter in order to minimize the capacitive effect on the equivalent circuit.

Cause this is a brand new area, no other equipment is running.

Is possible that this filter is contributing to the system become a pure capacitive?, cause at this moment there is no other motor (inductive load) running and the system is getting in resonance.

This could be possible?

Raul

 

[davidbeach]

What’s the overvoltage? Also 50Hz seems odd with that voltage.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Freshsurfer]

Hi @davidbeach, I don't have this information yet, but it burns some fans in other equipment below the same MCC, I'm expecting that this could be higher than 30% above the nominal voltage.

 

[waross]

Motor voltage: 460Vac 50hz
MCC voltage: 480Vac

In the NEMA world:
Motor voltage: 460Vac 60hz
MCC voltage: 480Vac
Would be considered exactly right.
Both a 460 Volt motor and a 480 Volt MCC seem unusual for a 50 Hz system.
If that is a 60 Hz motor then 480 Volts is much too high for 50 Hz.
The filter should not affect the line voltage when the motor is not running and probably not when the motor is running.

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

 

[jraef]

Unless it is an Active Front End VFD, meaning one with an inverter on the front end as the converter, instead of a diode bridge converter, then a filter, or ANYTHING on the OUTUT side of the VFD would have no effect on the MCC bus feeding the VFD. There is obviously more to this scenario than you are telling us.

The ONLY thing that a standard diode bridge front-end VFD can do to possibly affect the LINE side voltage is if it is resonating with capacitors on the line side bus, in which case the VFD would also be tripping on Over Voltage of the DC bus, probably first. If you have a capacitor bank on the 480V MCC bus, that is your problem. If it has a contactor to disconnect it from the bus, open that contactor and see if your problem goes away. If there is no contactor controlling your capacitor bank, that is the problem you must fix.


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

 

[Freshsurfer]

HI @waross, yes it is an 460Vac 50Hz. The site is a copper plant in Chile.

Hi @jraef, the VFD is an active front end (ACS880-37). Based on your explanation there is no way that the sinusoidal output filter could being causing this overvoltage, right?

I'm in Brazil bringing support to our guys at site in Chile. I will ask them for more information including the confirmation of a capacitor bank on the MCC bus.

Thanks!

 

[LionelHutz]

There's the drive model, details like that certainly help. When I looked up ACS880-37 the info I found said it's not line regenerative meaning it's not an active front end.

What burned exactly?

Resonating with capacitors is a possibility, but the VFD should have tripped on over-voltage if the voltage was reaching damaging levels.

 

[Freshsurfer]

Hi @LionelHutz, the equipment where the VFD is applied is a conveyor totally flat non regenerative.

At the same area we have a MV drive (ACS2000) with some fans in 480Vac that are feed by the same 480Vac bus, they have burned due this overvoltage. Also we have a crane that trips by overvoltage when the conveyor was running.

 

[jraef]

The U37 variant of the ACS880 is not line regenerative, but it is an AFE drive, the AFE is used for harmonic mitigation. So it cannot feed voltage back into the bus, but AFE drives CAN have resonance with OTHER NON-AFE DRIVES on the same bus or AFE drives from a different mfr, especially if the other drives do not have line reactors. If I’m not mistaken, ABB may actually recommend that their AFE drive have its own isolation transformer if there are other drives on the same bus. I strongly suggest getting them involved before you damage something. Resonant situations like that rarely end well.

It’s one of the risks of enthusiastic salesmen selling a complex technology that really should have a qualified engineer looking at the entire situation. AFE drives are great, but WAY more complex than standard 6 pulse inverters.


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

 

[Freshsurfer]

Thanks @jraef!

Yes we have some non-AFE drives on the same 480Vac bus. They are tripping due overvoltage when the ACS880-37 is running, also the main transformer is vibrating.

All this effect could decrease when the rest of the equipment begin to operate? There are a large number of inductive loads that are being commissioned.

 

[jraef]

It may not be alleviated by increasing the load, in fact it may get worse. The main culprit in these new AFE drives is the exitance of an LCL filter on the FRONT END of the drive to deal with the high order harmonics created by the line side inverter. In some designs, that LCL filter is connected to the line all of the time, hence the resonance issue with other VFDs.

Again, I stress the need for you to involve ABB in this discussion. Everyone that makes AFE drives has a "secret sauce" in their AFE algorithms and LCL filter designs so they all behave slightly different to specific situations. The best people to address yours are those that understand their own products.

I have been involved with several projects using Rockwell's PowerFlex 755T AFE drives and have had success simply adding Line Reactors to the other drives (which I felt should have been there anyway). But just because that worked on that brand is not a guarantee it is the right solution for the ABB drives. I put a Rockwell AFE drive into a pump station where there was an existing ABB ACS800-U37 AFE drive (the predecessor to the ACS880) and as soon as the Rockwell drive was powered on, the LCL filter in the ABB drive over heated and caught on fire!


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

 

[Drivesrock]

With a true active front end drive there should be a large input inductor, of around 10% impedance (Z) and there may be a PWM filter in there too - parts of the "secret sauce". The active front end control must 'know' (be programmed with) the correct value of that inductor else the voltage boost to be performed across it may be divided between the inductor and the supply inductance (perhaps the MCC supply transformer) and in this case you would see the MCC volts rise.
And for the input harmonics to be very low, the DC link voltage may be around 110% x the normal dc link voltage of a diode rectified 480Vac supply (eg. 1.35 x 480V =648V) so around 720Vcc.

Another aspect you perhaps need to be concerned about is your motor. If you do have a boosted dc link voltage the PWM voltage will be higher and with the long cable run the motor insulation will be under greater stress.

 

[Freshsurfer]

Thanks @jraef, ABB guys are working on that. My concern is about the other electrical equipment plugged at the same 480Vac bus that would suffer this overvoltage. They said that could be a grounding issue and they will check if the common point of the secondary part of the transformer (neutral) is hardwired to the VFD frame.

Thanks @Drivesrock!, we are using a 1k class cable for the motor, but I didn't understood your comment, is possible that the drive could being supplying a higher voltage to the motor?

 

[LionelHutz]

The output RMS voltage is not higher, but the peak voltage is higher. You have a sinewave output filter so that will filter the output to be fairly close to the RMS output voltage.