Balancing and Sizing a 120/240V 3-phase panelboard

[apelk]

I keep coming across this and need some more people to weigh in. Say you are loading up a 120/240V panelboard, balancing the loads in amperes. So on the sides of my panel schedule, I have so many amps running on each line for each load, which of course are calculated based on whether the load is 3-phase, single phase, or 120V single phase. Do I sum total all of the 'A' line currents, 'B' line currents, and 'C' line currents and that is what the panelboard size should be based on? Is that the actual current running on the lines?

After I have the loads balanced as best I can, if I take the total KVA on the panelboard and divide by 240 3-phase, does that number hold any relevance? (This is how I'd typically size a 120/208 or 277/480...) It ends up being less than the calculated individual line currents, so I don't want to undersize any feeders! I'm just not sure if this is the real number that the service sees and the line currents calculated above are only used to balance the panelboard.

Thanks!

 

[GoldDigger]

I keep coming across this and need some more people to weigh in. Say you are loading up a 120/240V panelboard, balancing the loads in amperes. So on the sides of my panel schedule, I have so many amps running on each line for each load, which of course are calculated based on whether the load is 3-phase, single phase, or 120V single phase. Do I sum total all of the 'A' line currents, 'B' line currents, and 'C' line currents and that is what the panelboard size should be based on? Is that the actual current running on the lines?

Why do you have A, B and C lines in a 120/240 panelboard?

The size of the panelboard is the common maximum rating for each individual line conductor not a sum of those ratings. As long as each individual line current is below the rating of the panelboard, you are fine.

 

[jumper]

Why do you have A, B and C lines in a 120/240 panelboard?

I imagine the OP has a 240/120 center tapped high leg Delta.

 

[kwired]

I keep coming across this and need some more people to weigh in. Say you are loading up a 120/240V panelboard, balancing the loads in amperes. So on the sides of my panel schedule, I have so many amps running on each line for each load, which of course are calculated based on whether the load is 3-phase, single phase, or 120V single phase. Do I sum total all of the 'A' line currents, 'B' line currents, and 'C' line currents and that is what the panelboard size should be based on? Is that the actual current running on the lines?

After I have the loads balanced as best I can, if I take the total KVA on the panelboard and divide by 240 3-phase, does that number hold any relevance? (This is how I'd typically size a 120/208 or 277/480...) It ends up being less than the calculated individual line currents, so I don't want to undersize any feeders! I'm just not sure if this is the real number that the service sees and the line currents calculated above are only used to balance the panelboard.

Thanks!

If you have a high leg (which I am assuming you do) then depending on your loads you may not be able to balance your 120 volt loads equally across all three phases without additional transformation anyway.

Highest line current on any one phase is the minimum size panel needed (to some extent) because if you load any phase beyond the main setting it will eventually trip even if both other phases are below the trip setting.

 

[JoeStillman]

With a 4-wire delta, you have to look at each phase leg separately. The currents are inherently unbalanced on that system.

 

[apelk]

Golddigger: When I said "Do I sum total all of the 'A' line currents, 'B' line currents, and 'C' line currents and that is what the panelboard size should be based on?" I meant sum the lines separately, not A + B + C. Yes, high leg system. Sorry I forgot to mention that.

I agree that it will be nearly impossible to perfectly balance the load and that the panelboard/feeders need to be sized for the highest line current. So I guess that sort of answers the question... that the total KVA on the panelboard divided by 240*1.73 is an irrelevant number? It should be correct if your panelboard only consisted of three-phase loads I suppose. I am just trying to understand the math behind the system and why for a wye system, when the loads are pretty much balanced, you can take the total KVA on the A + B + C and divide by the three phase voltage and get the correct number as an average. For delta high leg we cannot do that. That is what I believe, anyway - I'm trying to find a way to convince/explain to my dept head... Does anyone have any references for these calculations? That will definitely hold more ground than "some guys on mike holt forum said so." We've been looking at http://ecmweb.com/nec/understanding-basics-delta-transformer-calculations but dept head keeps going back to wanting to consider a panelboard as just a total KVA load and sizing the feeders based on total KVA divided by 240*1.73. I just don't think that number accurately represents the ACTUAL current running on the lines.

 

[jim dungar]

Golddigger: When I said "Do I sum total all of the 'A' line currents, 'B' line currents, and 'C' line currents and that is what the panelboard size should be based on?" I meant sum the lines separately, not A + B + C. Yes, high leg system. Sorry I forgot to mention that.

I agree that it will be nearly impossible to perfectly balance the load and that the panelboard/feeders need to be sized for the highest line current. So I guess that sort of answers the question... that the total KVA on the panelboard divided by 240*1.73 is an irrelevant number? It should be correct if your panelboard only consisted of three-phase loads I suppose. I am just trying to understand the math behind the system and why for a wye system, when the loads are pretty much balanced, you can take the total KVA on the A + B + C and divide by the three phase voltage and get the correct number as an average. For delta high leg we cannot do that. That is what I believe, anyway - I'm trying to find a way to convince/explain to my dept head...

You are making it hard by trying to make it easy.

You have two systems: one at 240V delta and the other at 120/240 single phase.
Figure out your delta kVA, figure out your single phase kVA, now add them together.

This is a worst case kVA and does not necessarily represent the real world.
The problem is that 240/120V systems are not always created from three equally sized transformers. The most common arrangement is for the center tapped transformer to be larger than any other transformer. Another common arrangement is to use only two transformer in an open delta. The forumlas for determining the size of each of the single phase transformers varies slightly with the connection, but they all still come down to: kVA = (three phase portion) + (single phase portion)

Do not try to directly use wye formulas with delta systems. While there is the same 'magic' 1.732 factor; in wye systems it is associated with the voltage calculations, while in a delta it is involves the currents.

 

[kwired]

With a 4-wire delta, you have to look at each phase leg separately. The currents are inherently unbalanced on that system.

That it dependent on what load is connected to it. If it is entirely balanced three phase loads, then it will be balanced. Problem is there is often at least some 120 volt load connected to such a system and that will throw balance off some without additional transformation to be able to use the high leg, but again could be made up by adding some 240 volt single phase loads to the high leg.

Golddigger: When I said "Do I sum total all of the 'A' line currents, 'B' line currents, and 'C' line currents and that is what the panelboard size should be based on?" I meant sum the lines separately, not A + B + C. Yes, high leg system. Sorry I forgot to mention that.

I agree that it will be nearly impossible to perfectly balance the load and that the panelboard/feeders need to be sized for the highest line current. So I guess that sort of answers the question... that the total KVA on the panelboard divided by 240*1.73 is an irrelevant number? It should be correct if your panelboard only consisted of three-phase loads I suppose. I am just trying to understand the math behind the system and why for a wye system, when the loads are pretty much balanced, you can take the total KVA on the A + B + C and divide by the three phase voltage and get the correct number as an average. For delta high leg we cannot do that. That is what I believe, anyway - I'm trying to find a way to convince/explain to my dept head... Does anyone have any references for these calculations? That will definitely hold more ground than "some guys on mike holt forum said so." We've been looking at http://ecmweb.com/nec/understanding-basics-delta-transformer-calculations but dept head keeps going back to wanting to consider a panelboard as just a total KVA load and sizing the feeders based on total KVA divided by 240*1.73. I just don't think that number accurately represents the ACTUAL current running on the lines.

If you were supplying entirely three phase balanced loads then KVA / volts / 1.73 is your amps per phase. It is only the unbalanced portion that will have to be added (per phase) on top of the balanced portion - just like you would for a wye system that is not balanced.