240 to 480 step up transformer questions

Status
Not open for further replies.
Hi, I've just moved my business into an older building and the electrical service is 240 volt delta. I have 2 CNC machine I need to hook up with a total load of 45 amps @ 480 volts. Both have VFDs driving their spindles. The electrician is proposing reverse feeding a 480 delta x 240 delta stepdown transformer. I asked him about corner grounding and he no, so he's proposing running it as an ungrounded delta system.

This sounds kind of dangerous, since in the event of a ground fault it becomes an ad-hoc corner grounded system and the breaker doesn't trip. Is there a simple way to add ground fault protection?

I've also considered intentionally running it as a corner grounded delta system. The problem there is that I'm finding differing opinions on whether the VFDs can tolerate it. Some manufacturers recommend disconnecting the MOVs, some drives don't provide for that, and some people even say that if the component spacing is right in the drive it's no big deal.

The third option would be ordering a custom transformer which will convert the 240D into 480Y. This seems like a much better solution, but it would cost an extra thousand dollars. What do you guys think?
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
...The third option would be ordering a custom transformer which will convert the 240D into 480Y. This seems like a much better solution, but it would cost an extra thousand dollars. What do you guys think?

Worth every penny.

The other option you didn't mention is to ask the utility for a 480V service and get yourself upgraded right now if you business is going to grow. It won't cost any less next year.
 

drcampbell

Senior Member
Location
The Motor City, Michigan USA
Occupation
Registered Professional Engineer
... an ungrounded delta system ... sounds kind of dangerous ...
Trust your instinct.
At very least, ground-fault detection should be incorporated, which would erode the cost saving vs. doing it right.

As for the VFDs tolerating corner-grounded delta, there are different opinions because there are different equipment designs. Ask the manufacturer whether these specific VFDs are suited for CGD.
 
There is nothing wrong with a corner grounded delta system. Just make sure your phase that you ground is marked gray and not an ungrounded color and "over label" everything so there is no question what has been done when you have a different electrician come out to look at something 10 years later. But if an ungrounded system can be avoided, you should avoid it. If your electrician is arguing that it is against code, look into 250.26(4). Many electricians (my last employer included) do not understand corner grounded systems and think they are illegal. If you are one day going to utilize a neutral tap off of this transformer, spring the extra money for the delta-wye.
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
There is nothing wrong with a corner grounded delta system. Just make sure your phase that you ground is marked gray and not an ungrounded color and "over label" everything so there is no question what has been done when you have a different electrician come out to look at something 10 years later. But if an ungrounded system can be avoided, you should avoid it. If your electrician is arguing that it is against code, look into 250.26(4). Many electricians (my last employer included) do not understand corner grounded systems and think they are illegal. If you are one day going to utilize a neutral tap off of this transformer, spring the extra money for the delta-wye.
Electrically there is nothing wrong with a corner ground other than the befuddlement they cause some electricians, that's true, but that is not the issue. There are lots of vfd's and cnc machines that do not like them and the trend for the last twenty five years has been for everything to be a grounded wye so increasingly machines that are able to cope with a corner ground or a Delta system are less and less.
 
Electrically there is nothing wrong with a corner ground other than the befuddlement they cause some electricians, that's true, but that is not the issue. There are lots of vfd's and cnc machines that do not like them and the trend for the last twenty five years has been for everything to be a grounded wye so increasingly machines that are able to cope with a corner ground or a Delta system are less and less.
I did not know that. Does the VFD just see it as a phase loss because there is no voltage to ground on that leg?
 

kwired

Electron manager
Location
NE Nebraska
I did not know that. Does the VFD just see it as a phase loss because there is no voltage to ground on that leg?
NO phase loss because there is still 120 degree phase differential and 480 volts between each leg. Problem is majority of off the shelf drives usually have MOV surge protection in the incoming lines, they are connected each phase to ground, but only rated for 277 volts nominal. 480 volts to ground will burn them out in very short time. Ungrounded system, they might be fine until a line becomes grounded, then you have 480 to ground on the other two phases. Otherwise they don't care what is grounded or if anything is grounded to function.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
What does the VFD then use as a ground reference for the pulsed DC output?
The DC bus voltage will not be symmetrically referenced to ground in the same way it would with a wye grounded input.
Troubleshooting the run between the VFD and the motor sounds a little iffy.

If the input is ungrounded delta then the VFD output will also be ungrounded, with possible problems with voltage withstand in the motor, no?
 

mike_kilroy

Senior Member
Location
United States
What does the VFD then use as a ground reference for the pulsed DC output?
The DC bus voltage will not be symmetrically referenced to ground in the same way it would with a wye grounded input.
Troubleshooting the run between the VFD and the motor sounds a little iffy.

If the input is ungrounded delta then the VFD output will also be ungrounded, with possible problems with voltage withstand in the motor, no?
No. The output has no need, desire, or want, to be referenced to ground.

Sent from my SM-G900V using Tapatalk
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
No. The output has no need, desire, or want, to be referenced to ground.

Sent from my SM-G900V using Tapatalk
But as a matter of simple physics the motor insulation is rated based on the voltage to ground that it can withstand as well as the voltage difference between the ends of the windings.
The magnetic forces that cause the rotor to spin could care less about ground, or indeed anything but the voltage imposed from terminal to terminal on the winding.
But given that the stress imposed on the insulation (particularly in the form of dV/dt) is higher for a VFD driven motor than for a 60Hz driven motor, I do question whether the voltage to ground can be safely ignored.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
What does the VFD then use as a ground reference for the pulsed DC output?
The DC bus voltage will not be symmetrically referenced to ground in the same way it would with a wye grounded input.
Troubleshooting the run between the VFD and the motor sounds a little iffy.

If the input is ungrounded delta then the VFD output will also be ungrounded, with possible problems with voltage withstand in the motor, no?

No. The output has no need, desire, or want, to be referenced to ground.

Sent from my SM-G900V using Tapatalk
But, many drives have Common Mode noise filter capacitors on the DC bus that ARE supposed to be referenced to ground, and when fed by a Delta system, that gets messed up. Corner grounding the Delta can risk losing the transistors that are connected to that DC bus. A-B drives provide you with easy to access jumpers for the MOVs and for the CM caps that removes the ground reference points for both systems so that the components are not damaged. But in removing the jumpers, you are voiding the UL listing of the drives and you are removing the protections those references provided.
PowerFlex Jumpers.jpg

Delta, whether corner grounded or not, is NOT a good choice for feeding an VFD. North America is the only place where Delta power systems are used, so the rest of the world is expecting to see a Wye system, so components are selected based upon that, meaning the maximum voltage reference to ground is EXPECTED to never be more than 58% of the Line to Line voltage. In a Delta system, the potential to ground could be the same as Line to Line. In fact, the "CE" conformance they all need in other countries totally depends on a Wye system serving the VFD, so for them, this is a non-issue because "delta never happens". Most Japanese based drives just have a disclaimer somewhere in their literature telling you that the drives are suitable for ONLY connecting to a solidly grounded Wye system. Some European based drives at least provide US customers with optional instructions for removing the ground reference points, but on some (Siemens for example) it's a royal PITA; you have to disassemble the front-end of the drive to get to them. The few remaining North American based mfrs, like A-B, make it easier to remove the reference points but like i said, it voids the UL listing because the UL listing was done with the jumpers installed.

Bottom line, pay the extra $$ for the Wye secondary. Corner grounded delta is fine if there were NO drives involved, but it's terrible if there are. Just say no to Deltas.
 

topgone

Senior Member
But, many drives have Common Mode noise filter capacitors on the DC bus that ARE supposed to be referenced to ground, and when fed by a Delta system, that gets messed up. Corner grounding the Delta can risk losing the transistors that are connected to that DC bus. A-B drives provide you with easy to access jumpers for the MOVs and for the CM caps that removes the ground reference points for both systems so that the components are not damaged. But in removing the jumpers, you are voiding the UL listing of the drives and you are removing the protections those references provided.
View attachment 17540

Delta, whether corner grounded or not, is NOT a good choice for feeding an VFD. North America is the only place where Delta power systems are used, so the rest of the world is expecting to see a Wye system, so components are selected based upon that, meaning the maximum voltage reference to ground is EXPECTED to never be more than 58% of the Line to Line voltage. In a Delta system, the potential to ground could be the same as Line to Line. In fact, the "CE" conformance they all need in other countries totally depends on a Wye system serving the VFD, so for them, this is a non-issue because "delta never happens". Most Japanese based drives just have a disclaimer somewhere in their literature telling you that the drives are suitable for ONLY connecting to a solidly grounded Wye system. Some European based drives at least provide US customers with optional instructions for removing the ground reference points, but on some (Siemens for example) it's a royal PITA; you have to disassemble the front-end of the drive to get to them. The few remaining North American based mfrs, like A-B, make it easier to remove the reference points but like i said, it voids the UL listing because the UL listing was done with the jumpers installed.

Bottom line, pay the extra $$ for the Wye secondary. Corner grounded delta is fine if there were NO drives involved, but it's terrible if there are. Just say no to Deltas.

All those, I agree.
 
Wow, you guys were busy while I was gone! Lots of good info here. I was hoping somebody would suggest a clever way to add ground fault protection to the delta/delta transformer for a hundred bucks or so. Oh well.

If I go with a proper step-up transformer will the inrush current be significantly less? Enough to make a difference?

Also, does the transformer have to be bonded to building steel even if it's only 5 feet from the panel and hooked up with GRC? There's a nice little nook for the transformer close to the panel, but the nearest beam is about 25 feet away.
 

mike_kilroy

Senior Member
Location
United States
But, many drives have Common Mode noise filter...

I agree with this reply as long as the "But" is removed. Of course everyone is on same page about MOV and filter caps linked to ungrounded AC input, and repeating the reasoning is always good reinforcement. But that was not the question I replied to.

Since we are on why to have grounded input again, one not mentioned is that unreferenced means when lightning hits nearby and causes huge common mode voltage spike on all 3 input lines, there is no low Z path to ground to help reduce them. The result is lots of repair business for vfd repair houses. Most pcb trace spacing and components inside the vfd or drive do not like 2-3-4-5,000V between or across them and fail. Passing this thought on to the motor, I can tell you that although we repair lots of vfds due to lightning strikes, in 44 years of doing this I cannot point to a single motor that failed at the same time - the vfd fails and the spike mostly does not get thru to the motor to damage it.

So again, in answer to the comment about the motor needing the vfd output referenced to ground, it simply does not.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
...
Also, does the transformer have to be bonded to building steel even if it's only 5 feet from the panel and hooked up with GRC? There's a nice little nook for the transformer close to the panel, but the nearest beam is about 25 feet away.
It is a separately derived system and must be connected to a grounding electrode if you want a grounded system.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I agree with this reply as long as the "But" is removed. Of course everyone is on same page about MOV and filter caps linked to ungrounded AC input, and repeating the reasoning is always good reinforcement. But that was not the question I replied to. ...
Point taken, my "but" shouldn't have been there. I think I initially misinterpreted GoldDigger's statement to mean the DC bus reference to ground and reacted without fully seeing it. Bad habit...
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Some VFDs will interpret a corner-grounded source as a phase loss or a phase imbalance, set a fault code and refuse to start.
True. Some VFDs derive their control power from 2 phases of the AC side with an SMPS and to protect it from phase loss, build in a required voltage reference to ground on that SMPS input, so if one side is already at ground potential it trips off line. Those are also the drives that cannot be used for single phase input (or are specific as to which terminals must be used if using a North American single phase input where it's not L-N)

Many (if not most) VFDs however derive their control power from the DC bus using a DC/DC chopper power supply so that the input voltage tolerance is wider, which also makes corner grounding on the input possible (from that standpoint).
 
Status
Not open for further replies.
Top