VFD and motor contribution to lineside faults 6

[veritas]

Hi

Not dealt with VFD's very often so need some advice please. I have read that a 4-quadrant VFD allows operation of the motor in all 4 quadrants, i.e. rotation and torque in the same direction as well as rotation and torque in opposite directions. When torque and rotation are opposite to each other then motor is breaking and acts as a generator feeding current back to the VFD. Hope this is correct.

Does this apply when there is a line side fault? Does the motor see the dip due to the fault as a "brake" and thus pump current back to the system? Is this how a motor controlled by a VFD drive can provide motor contribution to lineside fault cuurent? For a typical induction motor, how much contribution (as seen on the lineside of the VFD) can one typically expect from the motor via the VFD for a lineside fault?

Thanks.

 

[jraef]

It kind of depends on the drive. Many of them have line side filters that have capacitors, so those will always contribute a little bit (relative to the drive size).

In general though, the way a drive goes into regeneration is that it simply monitors the DC bus voltage and when that rises above a predetermined threshold, the drive’s Active Front End to fire it’s transistors and feed power back into the line source. For that rise in the DC bus to happen, there must be an “overhauling load”, meaning the load is mechanically spinning faster than the commanded referenced speed (ie frequency command). So unless the commanded speed is lower, the drive will not regenerate.

During a line side fault, where the line voltage goes almost instantly to zero, the DC bus in the drive can’t be maintained. If the motor is running it will deplete the stored energy in the bus caps in a matter of milliseconds, so the bus voltage will drop. In theory IF the drive is programmed to do so, it will override and lower the speed command in an effort to keep the motor operating with the reduced bus voltage. In that effort, it could lower the speed reference below the load speed and the load will pump energy into the DC bus. Then if it builds up to the threshold level, it could conceivably put energy back into the line.

It’s also possible, because for a line regenerative drive to function it must be monitoring the line at all times, that you just program the drive to turn off if the line voltage drops too low. Or you can just put the drive into “Coast” mode so that the motor is no longer fed with flux energy and it therefore can’t become s generator. There are several ways to avoid having it feed a Fault by just being cognizant of the issue when commissioning the drive.

So it’s a misnomer to say a Line Regenerative drive WILL contribute, you can say it MIGHT contribute under the right circumstances if allowed to do so.


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

 

[itsmoked]

You beat me to it Jeff!

yOU HAVE MOST OF IT.. You're missing one point. Actually all VFDs tend to be four quadrant, what they do with the regenerated power is what differs. 99% of the drives stuff the regenerated power back onto the VFD's capacitor bank. This raises it's voltage until it reaches a capacitor challenging limit at which time the energy stored is shunted to a braking resistor. Smaller drives include the breaking resistor. Larger drives require you provide the breaking resistor sized to handle the maximum of what the the installation can feed back. A crane could feed a whole lot back, stopping a machine spindle from speed,, can return a high peak but a very low average, etc. etc.

The key is if the drive is actually a regenerative drive in that case there is an entirely additional bunch of hardware which is essentially a grid-tie inverter that will return the excess power piling up on the capacitor bank as sync'd AC power to your mains. Regenerative drives cost 200% of the other other 99% of VFDs that are not, simply because they are rarer and have twice the electrical hardware inside them. (They're also larger for the same reason.)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[itsmoked]

Rereading both posts I'd add:


1) If it's not a regenerative drive it will not contribute one bit back to the source EVER.

2) If it is a regenerative drive it likely will not ever return more than it could draw as presumably the load it's running is consuming and would continue to consume until it's gone out of control on low voltage at which time the regeneration would trip off or have only a fraction of the power available to feedback.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[veritas]

Wow, some excellent information, had to read it through a few times. Thanks so much. I'll need to read a few more times to properly assimilate it all.

The drive I'm dealing with is an ABB ACS880 feeding a 250kW motor.

One more question. What can I expect will be the inrush current for the drive and for how long? Typically for a DOL drive of the same size, I would normally allow for 6 x rated current for around 10s unless starting current and time provided by manufacturer.

Thanks again.

 

[JuanBC]

Hi, nice information! Thank you veritas for asking such thing.

Can anyone give examples of commercial "regenerative drives"? How can I identify one of them?

1) If it's not a regenerative drive it will not contribute one bit back to the source EVER

I've always thought that "drives" in general does contribute "something" to line-side faults, this shows me that I was wrong

Thanks!

JBC
.......
"The more I read, the more I acquire, the more certain I am that I know nothing"

 

[itsmoked]

veritas; You don't specify the -xx number so we don't know if you have a regenerative drive. It would have to be a -11 or a -17 to be one.

As for inrush on power up a drive as large as yours will have a very large capacitor bank that would typically drop a source to its knees if it were just directly closed on. I would expect the drive to have a pre-charge stage where the capacitor bank is charged thru a resistor via a relay for perhaps 10 or more seconds before the drive will allow any output. So as for inrush I'd imagine it's inconsequential being probably less than what the drive would draw running at full load.

As for the motor it will likely have very little inrush as that's what VFD's do for you. The motor will be accelerated over time, always at a high power factor so there is always a fairly high back-EMF unlike a DOL motor start where the power factor may briefly be horrid, perhaps 0.1. You can completely control the inrush with a VFD down to nothing more than FLA.


JuanBC; VFDs generally contribute nothing to the source side at all. All the power they draw comes in thru one-way rectifiers, it cannot return.

Identification is by the drive part numbers and as stated they're usually larger and definitely more expensive.

ABB Has a whole regenerative drive brochure. ABB Regenerative Drives


Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

JuanBC,
For the most part, unless a drive specifically SAYS it is Line Regenerative, it is not. As mentioned, you are going to pay a LOT more for one, as in 2x the price, because you are in essence buying two drives for one motor. So unless you are of the habit of grossly over spending, you are not likely to "accidentally" end up with a Line Regen drive, you have to want it...

And there are now "Active Front End" drives that are NOT line regenerative as well, referred to as "Low Harmonic" versions. Same drive, different firmware inside. The Active Front End is never configured to allow regeneration, but because it has the necessary line side monitoring, it is used to alter the rectification pattern using what's called "Selective Harmonic Elimination" to accomplish reduction of line side harmonics to IEEE-519 acceptable levels. those however still have the filters and therefore the capacitors on the front end that are on-line all of the time.

Keith,
Good catch earlier. When he said "4 quadrant" in the very beginning and I took that to eliminate having to discuss non-regen drives. But technically a standard drive with dynamic braking resistors could be considered "4 quadrant" in a way.

And pre-charge inrush current limiting is usually no more than 1 second, although the VFD mfrs often have a stability check after that for another second or two on very large drives. Caps charge very fast too...


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

 

[JuanBC]

Thank you both Keith and jraef!

I will like to highlight two nice statements:

As for the motor it will likely have very little inrush as that's what VFD's do for you. The motor will be accelerated over time, always at a high power factor so there is always a fairly high back-EMF unlike a DOL motor start where the power factor may briefly be horrid, perhaps 0.1. You can completely control the inrush with a VFD down to nothing more than FLA.

And there are now "Active Front End" drives that are NOT line regenerative as well, referred to as "Low Harmonic" versions. Same drive, different firmware inside. The Active Front End is never configured to allow regeneration, but because it has the necessary line side monitoring, it is used to alter the rectification pattern using what's called "Selective Harmonic Elimination" to accomplish reduction of line side harmonics to IEEE-519 acceptable levels

Just by chance, in a few days I must parameterize a drive that feeds a 260 kW gasoline pump. The drive is a Schneider Electric Altivar ATV680C25Q4X10 "Low Harmonic" drive and in the Programming manual I've read that the "current limitation" is by default 115% of motor FLA

JBC
.......
"The more I read, the more I acquire, the more certain I am that I know nothing"

 

[edison123]

Come for the motors, stay for the drives.

Thanks guys.

Muthu

http://www.edison.co.in

 

[sibeen]

A follow up question. I've come across a site where the lifts are fed via a regenerative drive. When a lift is descending then there's power being fed back into the grid and all is fine. The issue is when the power fails the site then transfers across to on-site backup generators. Now when a lift descends the regenerated power trips the generators off-line. Any thoughts about solutions?

 

[itsmoked]

Oohhh that's a problem!

A few things can be done about that.

1) The elevator hardware should have an input that disables regen and switches to energy dump braking resistor(s).

2) The elevators get an input that tells them "no net export" and so one never descends unless another is rising. They can force this by just randomly raising an empty one.

3) Stop the elevators during generator operation. (keep stretchers available)

I'm kinda surprised the rest of the building can't easily absorb the exported power. Sounds like the the exported power ends up being a major percentage of the building consumption.

What size building is this? Multiple elevators? Generator kW?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[sibeen]

Yeah, big building Keith, and I've only had a cursory look and kick of the tyres so far. 24 lifts fed from multiple switchboards. They are doing a lift upgrade program. The previous lift system used braking resistors and the new system was put in place for power efficiency. The discovered this issue during commissioning. It will probably only occur whilst the building is lightly loaded. Haven't dug down into building normal loading as yet. We're still in fairly early information gathering mode. Have only monitored one distribution board and was seeing around 400 kW peak regen.

1. Nup, it's not that smart.

2. I've already suggested that the lift logic may be needed to be changed to do as you've suggested.

3. 50 odd stories - may need very strong people

 

[itsmoked]

Wow. That is a big'un.. Much bigger than I was even thinking. That would demand something like a PLC just to run the elevator allowance logic during generator opps. Regen is da bomb on this thing though. I image a couple hundred K$/yr power savings.

With that many elevators it should be fairly seamless to get a low net export coming out of the shafts without many people even noticing any slight anomalies in elevator response.

I hope you can keep us informed on this fascinating problem.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[edison123]

Can't the regen option be disabled in the drive when gen power comes on?

Muthu

http://www.edison.co.in

 

[sibeen]

Muthu, there's no braking resistors on the drives and no room to install any, and that power has to go somewhere. One option we are kicking around is a load bank installed near the generators that can be quickly electronically switched in when required, but as I said this is very early days and that's just something I blurted out in a meeting without much (if any) thought going into it.

 

[davidbeach]

That’s a well known issue with elevators. You need to limit the number of running elevators and then also run other large loads such as the building ventilation system.

Elevators are typically counter balanced at half their rated load. A lightly loaded elevator must be driven down and will regenerate going up. Fully loaded it must be driven up and will regenerate going down.

 

[itsmoked]

Gads, that's right David!

You'd need a lot of planning in that PLC. Load banks for this application would dump a lot of heat which often means, pretty big, and somehow dealing with transporting the heat. Best would be the roof except these days all the roofs on buildings of this magnitude are considered art objects. More fascinating yet.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

It also depends on the nature of the line regen drive. There is a known phenomenon with active front end (AFE) inverters having negative interactions with backup generators because of the drastic difference in the system impedance of the generators (being higher). It can create a resonance effect with the AFE in the drive and the wild swings in voltage will force a race between the drives or the generator to shut down, usually won by the generator protective devices. Some drive mfrs have come up with ways to dampen this in their AFE algorithm and in the way they design their line side L-C-L filters, others have not, choosing instead to just tell you “Don’t use them with backup generators.” Unfortunately that message rarely makes it to their sales force, who don’t like to say no to an order and the end user finds out only AFTER commissioning.

Not only that, but one manufacturer uses what’s called a “Fundamental Front End” which is a lower cost way of attaining a regenerative drive, but has poor performance and makes this problem even WORSE! I’m thinking that’s what you have, because they primarily train their sales force to sell it on “energy savings” based on their FFE having lower losses than other AFE drives, which at less than 0.1% difference (in a laboratory) is basically a crock of bovine excrement. Most other AFE drive manufacturers have investigated and rejected FFE designs because the problems they cause are not worth the perceived benefits even though the cost is lower. The one that proceeded with it though is so big that they can basically get a large number of their loyal customers to buy it anyway.


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

 

[edison123]

This might be off-topic, pre-VFD, how was the regeneration handled in elevators?

Muthu

http://www.edison.co.in