Power quality vs. VFD

Status
Not open for further replies.

nolabama

Senior Member
Location
new orleans la
I will try to be brief.

Are Allen Bradley Powerflex 700 drives able to be tuned to accept variable input powers. POCO is giving me single phase VAC between 495-501. After rectifiers VFD buss VDC is 802. Drives go into over voltage fault. Switch to standby power and input power is 471VAC. Drives function as advertised. This seems to be a rather low tolerance. I dont know if it matters or not, but the drives convert the single phase power into three phase. POCO is 255-260ish to ground, generator is 240ish to ground.

I need the drives to function on POCO power, not generator power, Duh.
I do have an open ticket with Allen Bradley - but I have not called them yet; this is a recurring issue with the POCO.

The result of the drives failing is a drawbridge stuck in the partially open position, ie closed to rail and water traffic. I dont need to be out bucking a transformer to solve this problem.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Do you have a line reactor ahead of the drive? If not, start there, it's cheap insurance and it will drop the line voltage a bit. Be that as it may, any 480V drive is (should be) designed for +10%, so right there it should tolerate 528V. But the PF700 is actually designed to take higher, the line voltage trip threshold is 570VAC and the DC bus voltage trip threshold is 810VDC. So in reality the PF700 has MORE voltage tolerance than most other VFDs. If you are tripping, it's likely something else other than what you are seeing from those numbers.

I would suspect one of the following issues:
1) that the added ripple from having 1 phase input is behind the problem. If you have not doubled the size of the VFD in order to provide for enough capacitance to absorb that extra ripple, you could be in trouble. When on the generator there is (comparatively) high impedance so the non-linear nature of the way the VFD draws current will cause a voltage drop in the generator, what's called a "flat top" on the sine wave. This has the effect of lowering the DC bus voltage as well, which might explain why you see the problem go away when running on the genset. But when you are on the utility source, the low impedance means you get the full effect of the peaks, and with insufficient capacitance in the drive, you might be ending up with significant resonance that builds up long enough to trip the drive, but is too brief to be showing up on the display.

2) is it tripping just sitting there, or when operating? You say it's a drawbridge, what are you doing with the regen energy that is coming back from the motor when the bridge is lowering? Do you have a dynamic brake package for it? Was it properly sized to handle the amount of regen coming off of that motor for long enough to finish the cycle? If not, and the drive is limiting the regen energy into the resistor to keep from burning it up, then the excess energy has nowhere to go and the drive trips off.

I'm happy to help, but you'll have to provide more info. By the way in full disclosure, I work for Rockwell, and albeit not in the VFD group, this is something I wave worked on for most of my career.
 

nolabama

Senior Member
Location
new orleans la
Line reactor is installed. Drive is 50 horse. Motor is 25 horse. It takes 12 horse to move. The problem occurs when in operation, but last fault has occured when we started the drives up on utility power. The bridge is a swing span so regen energy can occur in high wind situations. It has not been windy, and we have a large resistor bank installed. The braking package is a motor brake. No machinery brakes.
 

nolabama

Senior Member
Location
new orleans la
Solution? We will be running a computer model of our primary side transmission to see what is going on. We take power from the POCO and run 9 miles underground, then step down at bridge site. I still think the problem lies within the drive converting single phase to three phase, but getting three phase out to the location is not gonna happen. I think we are about 25 miles from the nearest source of three phase power. POCO proposes a voltage regulator on the primary side at our end, sorta like a mini substation at our switchgear. The other option would be a single phase to three phase convertor. I dunno which would be more reliable to handle the problem, I would like both. Cost will probably not allow that to happen.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Solution? We will be running a computer model of our primary side transmission to see what is going on. We take power from the POCO and run 9 miles underground, then step down at bridge site. I still think the problem lies within the drive converting single phase to three phase, but getting three phase out to the location is not gonna happen. I think we are about 25 miles from the nearest source of three phase power. POCO proposes a voltage regulator on the primary side at our end, sorta like a mini substation at our switchgear. The other option would be a single phase to three phase convertor. I dunno which would be more reliable to handle the problem, I would like both. Cost will probably not allow that to happen.
So 9 miles of HV single phase power run out to a transformer feeding this VFD? You may have some additional harmonics in that feed as a result of that distance and it is passing through the transformer into the VFD, showing up as extra ripple, kind of like a tank circuit. What is the impedance value of your line reactor? If you didn't specify, A-B would have given you 3%, if so see if you can borrow a 5% to try out, or change taps on the transformer if you can. You also might want to consult with someone like TCI on a harmonic filter ahead of the drive instead of just the reactor, tuned to the specifics of your installation. Did you get a response on your service ticket yet?
 

nolabama

Senior Member
Location
new orleans la
I have not called the ticket in yet. POCO pointing to harmonics also - and reminding me that they are well within the 5% limits they need to be in. Not sure of the of the line reactors size. Taps are set to the lowest settings on the secondary transformors.
 

Besoeker

Senior Member
Location
UK
You also might want to consult with someone like TCI on a harmonic filter ahead of the drive instead of just the reactor, tuned to the specifics of your installation.
My two cents worth.
You will have seen from that other thread (if you didn't already know) that we make variable speed drive systems and sometimes have to instal harmonic filters to mitigate potential supply voltage distortion. We have done filters for some large UPS systems, usually a few hundred kVA so not single-phase.

But we have a lighting division and now and again I get called in to look at some power quality issues. These are, of single-phase non-linear loads. Harmonic filters for this can be rather more difficult than for three-phase. You have to deal with lower orders and and that gets you closer to the fundamental power frequency and that makes the design a bit challenging. But doable.
A few pictures of voltages and currents from site might be quite interesting to look at........
 

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
Working on it from the "resolve the symptom" issue rather than really attack the problem, have you considered a small "buck" transformer to drop the voltage perhaps 5%? I'm not sure there is a standard product, but it shouldn't be difficult to get one wound.
 

Besoeker

Senior Member
Location
UK
Working on it from the "resolve the symptom" issue rather than really attack the problem, have you considered a small "buck" transformer to drop the voltage perhaps 5%? I'm not sure there is a standard product, but it shouldn't be difficult to get one wound.
Sensible and should not be hugely expensive.
 

nolabama

Senior Member
Location
new orleans la
We have harmonic filters installed. I do not know anything about them. Bucking a transformer is not a good option considering the remoteness of this bridge - its a 30 minute boat ride if the bridge is in a fault. Rail travel to it is also not so good - I have been waiting over an hour for my turn to check on this thing today. I have requested power quality monitoring but I dont think I will get it. If it helps - the overvolt faults cleared yesterday evening and returned mid morning today. I would like to think that the drives could handle some voltage fluctuations. If they are indeed rated for 528vac they should not be giving me a problem to begin with. Unless the three phase conversion is that big of a deal.
 
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Solution? We will be running a computer model of our primary side transmission to see what is going on. We take power from the POCO and run 9 miles underground, then step down at bridge site. I still think the problem lies within the drive converting single phase to three phase, but getting three phase out to the location is not gonna happen. I think we are about 25 miles from the nearest source of three phase power. POCO proposes a voltage regulator on the primary side at our end, sorta like a mini substation at our switchgear. The other option would be a single phase to three phase convertor. I dunno which would be more reliable to handle the problem, I would like both. Cost will probably not allow that to happen.

One of the speakers at a recent State IAEI meeting was a rep for a company that makes a rotary converter that works with VFDs. He was a salesman of course but it is certainly an option if the BB doesn't work.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
...If they are indeed rated for 528vac they should not be giving me a problem to begin with. Unless the three phase conversion is that big of a deal.
As I said above, the OV trip threshold on THAT drive is 570VAC, 810V on the DC bus. If you are getting 570VAC on the input side, you have a much more serious problem. It's far more likely that you are getting a high DC bus situation that is somewhat transient because it is based on other issues like harmonics.

You now say you HAVE a harmonics filter though? What kind? It's possible that if that was incorrectly designed, that could be a contributing factor. The simpletest would be to leave it disconnected for a while and see if the OV faults go away.
 
Last edited:

Besoeker

Senior Member
Location
UK
Bucking a transformer is not a good option considering the remoteness of this bridge
I don't understand the connection you are making between having a buck transformer and the remoteness of the location.
From your original post the fault that trips the drives is over voltage.

POCO is giving me single phase VAC between 495-501. After rectifiers VFD buss VDC is 802. Drives go into over voltage fault.

I should have looked at this a bit more closely and I apologise for not having done so.
The 802Vdc should have rung alarm bells before.

If you rectify AC the resulting Vdc should not exceed Vac*sqrt(2) which is the peak of an AC sine wave.
But, working back from the 802Vdc, that would give you 567Vac.

I can think of a few possible explanations - and please don't take this as a reflection on you:
  • The voltage measurements are not correct. Could be either AC or DC.
  • The supply has a lot of rubbish on it that is being collected by the bucket capacitors that follow the rectifier.
  • The supply voltage is so distorted that the ratio of peak to RMS is 1.6 rather than 1.4.
  • Or any combination of the above.

A picture of the supply voltage waveform might be quite telling.
 

Besoeker

Senior Member
Location
UK
Unless the three phase conversion is that big of a deal.
Ought to have addresssed this as well.

The VSD is roughly in three parts.
Fixed frequency AC to DC conversion.
The DC link section mainly bucket capacitors but also containing a DC link choke on larger/higher spec. units
DC to variable frequency AC conversion. Usually a six device IGBT bridge.

That last part is a common configuration pretty much regardless of how the DC is produced. So, I don't think your problem lies there. The 802V, in my opinion, is.
 

nolabama

Senior Member
Location
new orleans la
my bad on saying that it was rated for 528vac

my bad on saying we have harmonic filters - we do not

As to the voltage measurements not being correct. You would be right. They are not the measurement at time of fault, mearly what is present when I show up several hours later. We are not going to get a picture of the waveform, I think that would be the one piece of data that would clear everything up.

As to the buck boost transformer not being a viable option. The secondary side power has been adjusted up in voltage several months ago, now I am having a problem in the other direction. I have set all transformers I have to the lowest voltages.


The 9 miles of HV is #2 copper, and at the very least is in rmc where it comes up to make junctions. I have no idea if they ran the entire length in rmc.

The voltage measurements are not correct. Could be either AC or DC.
The supply has a lot of rubbish on it that is being collected by the bucket capacitors that follow the rectifier.
The supply voltage is so distorted that the ratio of peak to RMS is 1.6 rather than 1.4.
Or any combination of the above.


Ok then - would a buck boost transformer help the rubbish issue in the capacitors?
Would the buck boost help the distorted waveform? What would?

And finally the POCO is telling me that with such a light load attached to a very long primary we are experiencing a capacitance of sorts with the line itself. When they adjusted the transformers several months ago I was told they shut power off at the POCO and had to "discharge" the lines with a grounding device.... I was not there and have no other knowledge of this.
 

nolabama

Senior Member
Location
new orleans la
btw - thanks for the help guys - My brain hurts from the amount of real thinking I have had to do in the past week on this bridge. I dont know if I would say I am over my head, but I am learning a ton.
 

Besoeker

Senior Member
Location
UK
As to the voltage measurements not being correct. You would be right. They are not the measurement at time of fault, mearly what is present when I show up several hours later.
Yes, but if you measure about 500Vac and over 800Vdc even if it is several hours after the fault, that still indicates a significant problem that needs to investigated and correcting. Unless the 800V is from the drive historical log and not measured at the same time as Vac. You mentioned that there are braking resistors. The ones I have used with inverters have a brake chopper (IGBT) that gets triggered if the Vdc goes above a preset limit. This could happen during regenerative operation where the motor is being driven by the load and acts as an electrical generator.
We are not going to get a picture of the waveform, I think that would be the one piece of data that would clear everything up.
It would certainly help.

Ok then - would a buck boost transformer help the rubbish issue in the capacitors?
It would lower the magnitude of the fundamental voltage and consequently any rubbish, if there is any.
Would the buck boost help the distorted waveform?.
It wouldn't reduce the distortion as a percentage of the fundamental.
What would?
Difficult to know without more information......that waveform.........

And finally the POCO is telling me that with such a light load attached to a very long primary we are experiencing a capacitance of sorts with the line itself. When they adjusted the transformers several months ago I was told they shut power off at the POCO and had to "discharge" the lines with a grounding device.... I was not there and have no other knowledge of this.
I don't have a great deal of experience with transmission level voltages - some at 3.3kV and some 11kV.
Cables have capacitance and as part of our procedure (and standard practice I was told) to ensure safety was to discharge the conductors. We have an eathing (grounding) stick for that purpose. It discharges the conductor in a controlled manner. You then fit earthing leads from the conductors to ground to ensure that they they remain discharged.
My point is that there is probably nothing unusual in having to discharge the cable.
 

nolabama

Senior Member
Location
new orleans la
Difficult to know without more information......that waveform.........

About that waveform.... It turns out all of our communications guys have a ocilliscope on the truck. I am arranging to borrow one next week. That being said, I have never used one before:dunce:. What and where will I be reading. If it does not have a function to download data will a photo of the screen do? I assume we want load and no load waveforms. I assume we want the waveforms at the first transformer(which we will not get- No way I am gonna get stuck on the fixed span and this thing open without a boat- Not gonna get fired for this) and also right before the VFD's. Anything else ?
 
Last edited:

Besoeker

Senior Member
Location
UK
About that waveform.... It turns out all of our communications guys have a ocilliscope on the truck. I am arranging to borrow one next week. That being said, I have never used one before:dunce:. What and where will I be reading. If it does not have a function to download data will a photo of the screen do?
A screen shot would be helpful. Voltage particularly.
But take care. One lead of the scope is likely to be connected to ground unless it's a battery powered unit.

I assume we want load and no load waveforms. I assume we want the waveforms at the first transformer(which we will not get- No way I am gonna get stuck on the fixed span and this thing open without a boat- Not gonna get fired for this) and also right before the VFD's. Anything else ?
The voltage at the VSD input on and off load.
But please be careful about what you are connecting to and that the scope probes are rated for the voltage being measured.
 
Status
Not open for further replies.
Top