modification of 3 phase panel

[55classicride]

Can a 3 phase 120/208V distribution panel be modified and fed with 120VAC single phase. This would require somehow jumpering the 3 phases together within the panel. Does anyone know if this has been done before? Seems to me that this type of modification would void the UL listing of the panel. Others have told me that's it a piece of cake, no problem.
Would it be better to run 3 conductors all the same phase from the 120VAC source to the panel, so not to modify the panel?

 

[charlie b]

I don?t think either option is code compliant. To modify the panel interior would, in my opinion, void the UL listing. But more importantly, both options would use a panel in a manner that is not consistent with the manufacturer?s instructions. Take a look at the second sentence of 408.30. The manufacturer must mark the panel with the voltage and phase configuration for which the panel is rated. Then take a look at 110.3(B). We must install and use equipment in accordance with the instructions that came with the manufacturer?s listing or labeling.

 

[jim dungar]

Can a 3 phase 120/208V distribution panel be modified and fed with 120VAC single phase. This would require somehow jumpering the 3 phases together within the panel. Does anyone know if this has been done before? Seems to me that this type of modification would void the UL listing of the panel. Others have told me that's it a piece of cake, no problem.
Would it be better to run 3 conductors all the same phase from the 120VAC source to the panel, so not to modify the panel?

Its not the number of conductors you run but rather the system that feeds the panel. Most manufacturers list their equipment with the maximum line-line voltage, the number of phases, and the number of wires. If your panel is listed as 208Y/120 3 phase 4 wire you need to run all of these different conductors.

Anything you do to modify a product in the field (including punching your own knockouts) can potentially invalidate the UL listing. But, because listing is not required for most panelboards it would be up to your AHJ to approve an installation where you are not exceeding the manufacturers maximum line-line and line-ground voltages.

 

[yanici]

Anything you do to modify a product in the field (including punching your own knockouts) can potentially invalidate the UL listing.

I never considered that.

 

[don_resqcapt19]

While not exactly the same, the NEC does permit single phase 120 to feed a 120/240 volt 3 wire panel in Article 690.

690.10(C) Single 120-Volt Supply The inverter output of a stand-alone solar photovoltaic system shall be permitted to supply 120 volts to single-phase, 3-wire, 120/240-volt service equipment or distribution panels where there are no 240-volt outlets and where there are no multiwire branch circuits. In all installations, the rating of the overcurrent device connected to the output of the inverter shall be less than the rating of the neutral bus in the service equipment. This equipment shall be marked with the following words or equivalent:
WARNING
SINGLE 120-VOLT SUPPLY. DO NOT CONNECT
MULTIWIRE BRANCH CIRCUITS!

Don

 

[stevebrown]

More info on the problem

More info on the problem

Full disclosure - I am working with Mr. 55 Classic Ride on this project.

Charlie B - I understand your issue of the panel being used in accordance with manufacturer's instructions. However, this installation is in an electrical generating station owned by the utility. As such, it is not under the auspices of the NEC. What I need is some engineering analysis of this installation.

In our design, we are replacing a three-phase inverter with a single phase unit. Apparently, there were once some 208 Volt three-phase loads fed from this inverter, but not any more. Since all loads are now single phase, we wish to replace the inverter with a single-phase unit. The sticking point is that these loads feed a number of computer systems throughout the plant. Operations is very concerned about how long these systems will be out of service during the installation of the new inverter. Of course, replacing the panels will lengthen the time that these systems are out of service, and makes the problem worse. We will have a temporary feed to supply these loads during construction, but this is still seen as a higher-risk power supply than a battery-backed inverter. In other words, during the time of construction, there will be ahigher risk of losing these loads due to a power supply failure than if the inverter were in service. This is where the push is coming from for re-using these three phase panelboards in a single phase application.

When this idea was first presented to me, I thought it was silly, and probably dangerous. After looking at it, I'm starting to agree with the utility that it is a workable solution. We were already going to add a warning placard to alert personnel that the panel has been converted to single phase, very similar to the one that Don_Resqcapt19 pointed out in his post.

Here are the key points - 1) the panel voltage will be within manufacturer's ratings (120 Volt single phase), 2) the inverter limits fault current to a value well within the panel short circuit rating, and 3) no modifications are being done to the panel internals. The plan is to utilize the existing 4 wire power conductors to supply the panel. The major difference is that these three conductors are now carrying 120 Volt power that is in phase with each other instead of being 120 degrees out. Line-to-line voltage is now zero.

Can you come up with any reasons why this would be a bad idea? If there is a solid engineering reason for not doing this, we will dig our heels in and tell the utility "no". However, after spending some time thinking about it, I'm not finding any reason why it won't work. Anybody have any thoughts?

 

[charlie b]

Can you come up with any reasons why this would be a bad idea? If there is a solid engineering reason for not doing this, we will dig our heels in and tell the utility "no". However, after spending some time thinking about it, I'm not finding any reason why it won't work. Anybody have any thoughts?

I see two possible issues, and I think they are worth investigating. Both might prove to be non-problems, and the installation might turn out to be safe. But I would not sign and seal a set of construction documents that issued the design you propose, without first investigating these two issues. I can think of no other engineering reasons why it would not be safe.

Issue 1:
Reference the code article posted by Don. I do not know how to find out the rating of the neutral bar. Normally, in a 3-phase panel, the neutral bar would carry no significant current. The maximum current it could be called upon to carry would be the rated current of one of the phase bars. That is, if the panel is carrying its rated load, and if all the load is on Phases A and B, then the neutral bar will be carrying the same current as Phase A and as Phase B.

However, consider your situation. Suppose that the three phase bars (A, B, and C), each of which is getting the same phase as its power source, is carrying full rated current. That would mean that the neutral bar would be carrying three times the maximum amount of current that it would carry, if the panel had been wired for a three phase supply.

I am inclined to think that this is a deal breaker. But again, I do not know how to tell the maximum rating of the neutral bar.

Issue 2:
You have three vertical bus bars running down the center of the panel. Under the usual three-phase installation, the magnetic field generated by current flowing in any one of the bus bars would be essentially cancelled out by the current flowing in the other two. But your proposal would have the three magnetic fields in phase with each other. The total field, as seen by the immediate vicinity, would triple.

As viewed from far away (let?s say 10 feet from the panel), the magnetic field of the three phase bars might be cancelled by the magnetic field from current flowing in the neutral bar. But as viewed from closer in, let us say as viewed from the surface of the panel enclosure, there will be a high magnitude, localized magnetic field. My concern is that this could generate a current within the enclosure itself. What I mean is this: Suppose you ran a strip of black tape, starting at the top, center of the front of the enclosure, running the tape down the front of the panel, across the bottom, up the top of the back side, and along the top edge to your starting point. I suspect that the high magnetic field might cause a current to flow along that path. The front of the enclosure itself might be overheated by this current, and the current itself might make the panel enclosure unsafe to touch.

I am certain that this current would flow. Of course, I have no idea whether the current would be so small as to be undetectable, or whether a person could be seriously burned, or even electrocuted, by touching the panel.

 

[stevebrown]

Good points

Good points

Charlie - Issue number one, as you say, could be a deal breaker. The neutral bus may be oversized in this panel, or it may not, I don't know. However, it is probably not sized to carry triple the original rating. This is a really good observation, and one that we'll follow up on. Thanks for the insight!

I'll have to think a little more about issue number two. Intuitively, it seems to me that the induced current will be of a very small magnitude. Let me puzzle on this a little more.

Again, thanks!

 

[charlie b]

I'll have to think a little more about issue number two. Intuitively, it seems to me that the induced current will be of a very small magnitude. Let me puzzle on this a little more.

My intuition says the same thing. After all, this sort of thing is allowable, under the circumstances that Don pointed out, for a 120/240 panel (i.e., with two vertical bars). I wouldn't think that one more would turn "safe" into "dangerous."

 

[winnie]

Additionally, since this is an existing installation, I would want to double check that there are in fact no multiwire branch circuits supplied by this panel.

-Jon

 

[jim dungar]

However, this installation is in an electrical generating station owned by the utility. As such, it is not under the auspices of the NEC.

You say this is for computers so it probably is exempt, but I like to remind others, that there is no true blanket exceptions for all utility owned structures.

According to NEC 90.2(A)(4) "Installations used by the electric utility...that are not an integral part of a generating plant, substation, or control center" are covered.

I know most AHJs exempt all utility installations (and many do the same for industrials) but that is their perogative not an NEC requirement.

 

[kingpb]

I have extensive experience in power generating facilities. Although they do not have to comply with the NEC, they typically try too. I agree completely with Charlie on this one. I understand they are trying to find a work around, and that keeping operations running is paramount. But this usually means they are willing to spend more $, not less. Believe me, most utilities, want to do it right. The last thing they want is a shutdown. I would look for the right way to do the job. Perhaps contact a custom panel builder, or even the inverter supplier, to get you a single phase panel, while using the one you have temporarily. At most, the existing panel would have a 200% neutral bus, but in no way a 300%. You may have to install a single phase, 200% neutral panel next to it, temporarily and switch over some loads.

The long and short of it, do not cut corners. A single shutdown can cost millions of dollars, a few thousand for a new panel is nothing in this case. Look for the right solution, not the cheapest one.

 

[scott moran]

You say that because it is a Utility it is exempt from the NEC. I believe that they installations for the delivery of electricity to consumers is exempt. However, their facilities should not be as a "blanket" statement.

 

[petersonra]

I think I ran across this once before. Seems to me at least one of the panelboard manufacturers has some kind of kit to convert a MLO panel to use just 120V instead of 120/240. In the murky recesses of my mind...

 

[kingpb]

You say that because it is a Utility it is exempt from the NEC. I believe that they installations for the delivery of electricity to consumers is exempt. However, their facilities should not be as a "blanket" statement.

Do not confuse Utility with IPP (Independent Power Producer) They both own power generating facilities, but different rules apply.

Certainly, a building downtown that houses the utility accounting, HR and so forth certainly is built to NEC. The plant, or generating station and all buildings on it, are not required to follow the NEC, albeit they pretty much do. Not the older one's so much, but any of the newer ones will/do.

 

[dbuckley]

As mentioned earlier, it is the neutral rating that is the key to this working. With the assumption that this is a standard panel (and not an IT rated panel with 200% rated neutrals for IT use) the only safe assumption to make is to derate the panel to one-third of it's original rating. If that doesn't fit, then I would say you are outside the rated capacity of the panel.

The other thing I'll mention but which I'm sure you've already thought of is the UPS; is it single phase in, single phase out, or three phase in, single phase out? In the latter case you need to think about bypass arrangements, as under bypass conditions the entire load will come from one supply phase. This phase will also be providing the fault current to clear faults, as UPSs cant do that, so they drop to bypass for a couple of seconds so utility power can provide the fault current.

Edit: the OP actually used the word "inverter" rather than "UPS", so I may have read more into this aspect than is the case.

 

[benaround]

Seems like a nice new 3ph inverter would be so much more adaptable, What

makes the single phase unit worth all this trouble?

 

[stevebrown]

Why change?

Why change?

Seems like a nice new 3ph inverter would be so much more adaptable, What

makes the single phase unit worth all this trouble?

Well, there were two factors that drove us towards replacing the three phase unit with a single-phase unit. Remember, all of the original three-phase loads have been removed over the years, so there is no need for a three phase unit.

The first factor was price of the unit. Due to the reduction of load, the original 40KVA unit was grossly oversized. Although we have not completed our analysis, it looks like a 20 KVA single-phase unit is sufficient for this application.

The second factor is standarization of equipment. The station has a number of 20KVA units already installed. This was the only 40KVA three phase unit installed at the plant. By switching to a unit that is already in use at the plant, they save on spare parts storage and training for Operations and Maintenance personnel.

 

[earshavewalls]

Just a thought, but have you checked with the manufacturer of the panelboards? You may be able to simply replace the bussing within the panel (the 'guts'), let the back boxes remain, and easily meet code with a minimum of down time.
If the manufacturer has multiple internal configurations for each panelboard enclosure, the U.L. listing would not be violated, the configuration would fit the use, and everybody could be happy.
Of course, if this is not the case, never mind.......

 

[LarryFine]

I've noticed that, while the panel's neutral bus has been mentioned over and over, nobody has addressed the neutral conductor in the feeder. 20Kva is 166a @ 120v. I'll bet the original neutral was not sized for this.