3 phase vfd only pulling current on two phases 1

[jsftech]

I have a 50 horsepower heavy duty drive (AB powerflex 700) connected to a 50 horsepower 1185 RPM, 460V 59.2 FLA motor. It spins a 78.75-inch diameter, 3100-pound cage through a 2.0-1 right angle gearbox.
There is also a 250 HP fan pulling air and ground silica through the cage. This airflow will spin the cage without the cage motor being energized. The system is used to classify ground silica particles.
We run the drive around 37 hertz for one product and 21 hertz for another product. We have been having trouble deceling the drive from 37 Hz. to 21 Hz. (decel inhibit fault). I have checked into a dynamic brake and a regen unit. I have managed to eliminate the faults for now by changing the Feedback select parameter from slip comp to open loop and enabling Flux Braking. We only decrease the speed 3 to 4 times a week. The drive shows about 22 amps at 37 Hz. and 17 amps at 21 Hz.
While troubleshooting I decided to check input amps with a clampon and noticed there was no amps on one phase. I switched that input phase with another one and I still didn't have any current on the same input phase of the drive.
I contacted Allen Bradley and they said there was a problem with the drive. They shipped us a new one. I installed it today and it has no input current on the same phase. I contacted our local Allen Bradley distributor and they have never heard of a drive doing this.
Has anyone ever seen this before? Do I have a problem? The drive is in service. Allen Bradley had some concern about the overhauling load while trying to troubleshoot the decel inhibit fault. I don't think the drive is under sized. The drive is only showing 22 amps on the output. It is a 77-amp drive. Does the drive sustain a low amp output by only pulling power from two phases? Allen Bradley doesn't seem to think so. When the drive is accelerating it is pulling amps on all three phases. Thanks.

 

[waross]

Start checking voltages including voltage drops.

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

 

[jsftech]

Input phase to phase voltages are 468, 468, 478.

 

[waross]

Is that at the panel or at the VFD?
Is that loaded or unloaded?
Are the two lines that measure 478 Volts the two lines that are drawing the current?

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

 

[jsftech]

Thats at the panel when most all loads are running. I will get more specific data tomorrow. Is there something specific I need to pay attention to? I've had some concern with power quality. The local utility has a Dranetz power quality analyzer that has been recording all week. It is connected at the incoming connections of our MCC. They will be in tomorrow to retrive it. Is there something I should pay attention to when we analyze that data? Thanks for your input.

 

[itsmoked]

jsftech; How does the 478V line figure in your phase swapping test?



Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottyUK]

The input to that drive is almost certainly a dumb three-phase 6-pulse rectifier, so the most likely cause of no current draw on one phase would be a sick rectifier or a bad / incorrect connection.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Marke]

Because the input to the drive is essentially a three phase rectifier into a capacitor bank, the current is going to flow on the crest of the waveform. If one phase is lower in voltage than the other two, much lower current, or even no current will flow in that phase.
You have quite a range between the maximum and minimum phases and I suspect that that is where the problem is.
If you add a three phase AC line reactor in the input, the current will be more balanced and in this case, you may have the current in all three phases.

Best regards,

Mark Empson
L M Photonics Ltd

 

[jraef]

Might the effect also be exacerbated by the fact that at his operating speed, the VFD is so lightly loaded that the voltage imbalance CAN allow all the current to be drawn from only 2 phases? If you think in terms of running a VFD on a single phase input on purpose, what do we tell people? We tell them that the capacity of the VFD is limited to about 1/2 of the rated capacity. He is running at 1/2 speed, so the HP is reduced by the same amount. The imbalance causes the effect Marke mentions, but nothing is affected because it isn't a problem for the VFD to draw from only 2 phases in that scenario.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

Yep, load it fully and the DC buss drops, allowing the lower voltage phase to conduct and contribute. Presently the VFD drawing from the higher phases is going to help reduce the difference in the phase voltages.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

To address the first issue. If you only have a problem with regeneration during the deceleration period then a braking resistor would work. If this cage will actually be spinning faster than the 21hz speed then you should look for a regen unit to use with the drive.

If the voltages were as follows;
line 1 - line 2 = 468
line 2 - line 3 = 468
line 3 - line 1 = 478

If you swapped line 1 and 2 then you could still have the same voltage readings and the drive would behave the same way.

Under low loads the drive will be able to keep the caps charged with current drawn from 2 phases. If you loaded the motor further the ripple voltage on the buss will increase and it will begin to draw current on all 3 phases.

Since you are indicating the air from the fan will spin this cage I'd have to think you have very little load on this motor. The drive will only draw the real power the motor uses + losses from the line. If you ran the motor uncoupled you would see something like 1A on the input side and 20A on the output. This is normal. If there is a voltage imbalance it will only draw the power from the highest phase to phase lines.

 

[jsftech]

Thanks for the info so far. When I swapped two phases on the input I swapped B & C. C phase had no current before or after swapping the phases.

Here is some data I collected yesterday.
With the drive running at 22 Hz., 17 amps, & 653 BVDC

Input (before line reactor)

Voltage Current
A-B 479 A 5.0
B-C 470 B 6.2
C-A 471 C 0.2

Output

Voltage Current
A-B 164 A 16.9
B-C 164 B 17.1
C-A 164 C 16.9

Here is some data I collected this morning with all loads in the plant on.

Drive is running 32 Hz., 22 amps, 630 BVDC

Input (before line reactor)
Voltage Current
A-B 472 A 13
B-C 461 B 13
C-A 460 C 2

Input (after line reactor)

Voltage
Output
A-B 471
B-C 461
C-A 460

Voltage Current

A-B 240 A 21
B-C 240 B 21
C-A 240 C 21

Two amps is the highest I've seen C phase since I noticed the issue. Perhaps it is because our input voltage is a little low this morning.

I hope I've answered all the questions you guys had. Remember this drive was brand new yesterday; it was replaced because the old one was doing the same thing. When I installed the new drive I also installed the line reactor. Thanks.

 

[DickDV]

In line with Marke's and jraef's answers, I don't think that the drive is bad especially when current tends to balance out when seeing heavier loads as in accelerating.

As to the decel issue, you will need some form of braking. The type is largely an economic one. If you are seeing a relatively small or intermittent braking event, then use a snubber resistor package and waste the energy as heat. If, on the other hand, if the braking energy is continuous at the slower speed or if your calculations show that recovering the braking energy will pay for a regen module for the drive, then choose that more expensive option. It usually requires nearly continuous braking to justify the regen option with a payback period of 18 months or less.

 

[waross]

Have you checked the transformer taps on the main transformers? One tap set wrong in a three phase transformer bank will do this.
Another issue. With the high voltage on one phase the drive is acting as if it was fed from single phase. It should not be operated in excess of its single phase rating until the voltage issue is resolved.
One solution may be to search for some transformers rated 277V:480V connect the primaries in wye with the neutrals floating. Connect the secondaries in delta. The primary neutral will displace slightly and the delta will produce balanced three phase voltages. Use surge protection on the secondary as this connection is prone to switching transients.
This is not a very good solution. It results in an ungrounded system which may cause problems for the VFD unless the delta system is grounded somehow.
Can anyone suggest whether a three phase delta:wye transformer will give the same secondary balance as the three single phase transformers?
Another issue, this will not work if the voltage difference is the result of a displaced neutral on the primary. (Displaced neutral: The result of neutral currents in the neutral conductor and the resulting voltage drop on the neutral.)

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

 

[jsftech]

Thanks for the info guys; I'm really getting some good insight. At first I suspected the incoming might have been an issue.
Upon doing research I discovered NEMA standard MG 1-14.34 which defines voltage unbalance. I also found out a voltage imbalance of 1% can cause a current imbalance of up to 10%. We can see a current imbalance in most all our motors up to around 7%.
When calculating our voltage unbalance I found it to always be less than 2%. Everything I have read about VFDs says they can handle a voltage unbalance up to 2%. So I started leaning away from incoming being the cause of our issue. But possibly I need to reconsider.
Our distribution transformer was replaced back in November of last year. We replaced 3 single phase 500KVA transformers with one three phase pad mounted 2500KVA transformer. Both were feed 12,470 volts Delta on the high side and stepped down to 480 volts wye. We are only pulling 3 phase loads and a few 480 volt lights from this bank. We have no neutral ran from the wye connection. The old bank we had wasn't even grounded. When they installed the new bank they grounded the wye connection.
The old transformers also produced about 10 volts high on one phase. We have had a voltage and a current imbalance for a long time. I didn't think we had any issues until this drive started acting erratic. I have also recently noticed on our 600 HP motors the current between phases have exceeded 120 amps sometimes this equates out to only a 7% current imbalance. That’s when I called the local utility. I'm hoping to get them to redistribute some of their loads on the incoming line to bring our voltages closer together, if I can convince them that this is a significant issue although they are delivering us power within spec of what they are required to. Maybe that will correct our issues. Do you think I'm on the right track? Thanks.

 

[TurbineGen]

What are your voltages without the load connected? Could it be that phase C has more significant loading upstream and the voltage is not able to overcome the voltage of phase A and B? I know you mentioned that all loads are 3 phase except for some lights. How many lights? Other than that, concur with Bill that your transformer could be off tap.

I think you could be on the right track. Would it be possible to trip some of the other loads (specifically the single phase lights) and see how the VFD behaves?

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[jsftech]

A few weeks ago I checked the voltages at our MCC Before we fired the plant up. The only load was a 480v 3 phase well pump and a 45 KVA 480v to 208v transformer. I turned off all the lights and the voltages were still about 10v high on one phase. Somewhere around 502,492,492. I’m not sure if the 3-phase transformer would have a different tap for each phase, plus our old bank showed the same voltage deviation. I'll check with the utility next week. They are supposed to come out and check our voltage coming in. We'll find out if the voltage coming in is even.

 

[ozmosis]

I'm surprised the PF700 hasn't tripped. I'm quite sure this drive does not have CT's on each phase but measures the DC link ripple and determines phase imbalance. In this way it should trip but if the load is light it might be holding out.

 

[DickDV]

jsftech, I just don't see where you have any problem here except possibly the need to add a little snubber brake capacity.

Flux braking rarely exceeds 10% of the motor rated hp so, if that helped as much as you say it did, then an inexpensive snubber brake rated about half of the drive rating would be plenty.

Drives run on imbalanced input power all the time and, while not ideal, it does not usually cause any trouble.

 

[mccla5214]

Questions:

Have you checked the voltage to ground on the inputs?
Are two inputs to ground higher than the third?
Is the low voltage leg to ground the one with no current?
Have you removed one of the legs that have current to see if the one without current starts working?

With 2 legs higher than the third and lightly loaded, semiconductors want to draw more current through the device with the highest voltage differenct across the device. Since the input to a VFD is 6 semiconductors you can see where a low voltage drop could occur.

I would suggest that you disconnect one of the legs that are pulling current and see if the dead leg rises from the dead.