Distributing Load evenly

Status
Not open for further replies.
A

arnettda

Senior Member
I did a load calculation on a house and came up with 385 amps. I am going to use a 600 amp service due to possible future expansions. I will end up with three 200 amp panels in the house. Is there a good way to distribute load evenly between the three panels? A third of the sq footage on each panel, and then just divide up loads to spread them out. I have never done anything like this so I am looking for a little advice from the more experienced. Do I need to worry about distributing it evenly between the three or just so I do not exceed one panels ratting?
 
For a house that needs three panels, I would stragetically place sub panels in different locations. One for the top floor, some

place on the main floor. You can cut down on the length of home runs. Size your feeders accordingly
 
John120/240 said:
For a house that needs three panels, I would stragetically place sub panels in different locations. One for the top floor, some

place on the main floor. You can cut down on the length of home runs. Size your feeders accordingly
Click to expand...

All panels will be in main mechanical room in basement. As per home owner.
 
121220-0830 EST

arnettda:

John120's suggestion is good.

Why a 385 A load in a home? Of what does this load consist? I am going to guess resistive heating, resistive hot water tank, maybe one or more instant-on water heaters, hot-tubs, and electric cooking stuff.

I would think the type of loads and their physical locations would a major factor in determining how the loads were distributed. How big is this house? 4000, 8000, 12000, or 20000 sq-ft?

In my own home I have a 200 A main panel in the basement, a basement sub-panel, a first floor sub in the pantry, another in the laundry room, a second floor sub, and a garage sub. I also ran a single circuit from the main panel to an outlet near each sub-panel, and this circuit also supplies the power for several low voltage power supplies for my GE RR relays. These panel locations are for convenience, and the load distribution is what it is. The biggest loads all come from the main panel. The most used breakers for on-off purposes are the garbage disposal, and two circuits in the garage for plug-in heaters. When the garbage disposal needs a hand in the disposal both the wall switch and the breaker are turned off. Having a panel in the pantry makes this convenient.

.
 
arnettda said:
I did a load calculation on a house and came up with 385 amps. I am going to use a 600 amp service due to possible future expansions. I will end up with three 200 amp panels in the house. Is there a good way to distribute load evenly between the three panels? A third of the sq footage on each panel, and then just divide up loads to spread them out. I have never done anything like this so I am looking for a little advice from the more experienced. Do I need to worry about distributing it evenly between the three or just so I do not exceed one panels ratting?
Click to expand...

My brain is racing to about 20 different scenarios. Not knowing how long your feeders are, or the loads that require so much amperage, or how you intend to feed the panels makes one "fits all" answer difficult. As pointed out, if the high loading is due to a couple insta-hots and resistive heating, I would have one panel with a separate feeder from the sercice that feeds these loads and is sized accordingly. It is unlikely that this load will change in the future. so say that is a 40 ckt 400A feed. I would bring in a second 400A feed to the other panel and do feed through lugs to a second section. At 600A I am assuming you will have to have a CT meter situation from the Utility? Regarding "balancing" the panels, as a Master Electrician and a designer of the electrical, it is my opinion that you are responsible to do load calculations on the entire service, but you are also required to do load calculations on the individual panels as well. From there you should be able to use common sense to decide where you want to leave your available expansion amperages.
 
Strathead said:
My brain is racing to about 20 different scenarios. Not knowing how long your feeders are, or the loads that require so much amperage, or how you intend to feed the panels makes one "fits all" answer difficult. As pointed out, if the high loading is due to a couple insta-hots and resistive heating, I would have one panel with a separate feeder from the sercice that feeds these loads and is sized accordingly. It is unlikely that this load will change in the future. so say that is a 40 ckt 400A feed. I would bring in a second 400A feed to the other panel and do feed through lugs to a second section. At 600A I am assuming you will have to have a CT meter situation from the Utility? Regarding "balancing" the panels, as a Master Electrician and a designer of the electrical, it is my opinion that you are responsible to do load calculations on the entire service, but you are also required to do load calculations on the individual panels as well. From there you should be able to use common sense to decide where you want to leave your available expansion amperages.
Click to expand...


The best way then is to Just sit down with pen and paper and devide the loads up.
 
gar said:
121220-0830 EST

arnettda:

John120's suggestion is good.

Why a 385 A load in a home? Of what does this load consist? I am going to guess resistive heating, resistive hot water tank, maybe one or more instant-on water heaters, hot-tubs, and electric cooking stuff.

I would think the type of loads and their physical locations would a major factor in determining how the loads were distributed. How big is this house? 4000, 8000, 12000, or 20000 sq-ft?

In my own home I have a 200 A main panel in the basement, a basement sub-panel, a first floor sub in the pantry, another in the laundry room, a second floor sub, and a garage sub. I also ran a single circuit from the main panel to an outlet near each sub-panel, and this circuit also supplies the power for several low voltage power supplies for my GE RR relays. These panel locations are for convenience, and the load distribution is what it is. The biggest loads all come from the main panel. The most used breakers for on-off purposes are the garbage disposal, and two circuits in the garage for plug-in heaters. When the garbage disposal needs a hand in the disposal both the wall switch and the breaker are turned off. Having a panel in the pantry makes this convenient.

.
Click to expand...

8720 sq ft, 4 washing machines and electric dryers, double wall over, sauna, 2-2ton- 1-3ton,1-4ton ac unit, 1 icemaker, 3 dishwashers, disposal, dubwaiter, 5 hp well pump,warming drawer, exhaust hood, 2 wine fridges, 3 -48 inch fridges, 2-36 inch fridges,wine cellar cooler pretty much sums it up. 10kw space heater that I did not count.
 
121220-2341 EST

arnettda:

You have stated all panels are in the same location. Thus, load location is of no importance.

First I would separate the loads as follows:

1. Put all the refrigeration equipment in one category. This does not mean on the same panel.

From your list 2 wine fridges, 3 -48 inch fridges, 2-36 inch fridges, 2 wine fridges, and wine cellar cooler. These are what I define as a "continuous cycling load". The on time might be 50% with the total cycle period about 1 hr. But they won't be in synchronization or have the same cycle time. These as a group probably have an average load about equal to the sum of the average load of each unit.


2. All air-conditioners in another category. These are also a "continuous cycling load" but may operate only half of the year. However, the half year use information may be of no value.


3. The 4 washing machines and electric dryers, double wall over, sauna, 1 ice-maker, 3 dishwashers, disposal, dumbwaiter, (what about cooking equipment) are what I define as "non-continuous cycling loads". These cycle when on, but are on for only limited times.


4. The space heater is probably a "continuous cycling load" part of the year. What other heating equipment is there?


5. You did not mention lighting. If possible the lighting should be on one panel that has the least common impedance with other panels, and if its panel has other loads, then these should have the least inrush current and large cycling current.

From the way the loads operate you need to distribute them to produce the least heating in the panels, and least interference with lighting.

Conceptually separating the loads in this manner may help with your distribution. Obvious you need to consider the NEC criteria, but you want to minimize light flickering. You pick the category for the pump. There are a lot of motors cycling at different times that will cause inrush voltage dips.

.
 
121221-1200 EST

arnettda:

It might be useful to add one additional panel, just for lights. It would be fed directly from the meter with its own wires. This would provide the least common impedance between the large cycling loads and the lights. Then the only common impedance is from the meter toward the power company generator.

Put the biggest cycling loads with high inrush current on the panel with the longest wiring from the meter to the panel. This might very slightly reduce inrush current.

.
 
Only 9 tons of A/C? In Arizona we would expect closer to 20 tons.

The biggest issue I encounter on projects like this is series rating the panels. In my area the available fault current at the service would be around 36KA. I generally have to split the A/C loads between two or three panels to keep the motor load below 100A on each panel. I also find that three panels is not always enough for this size residence. My reading of the NEC is that each panel has to have a calculated load at or below the ampacity of the feeder, and that the "40% of non-continuous over 10KVA" calculation applies only at the service.
 
hmspe said:
Only 9 tons of A/C? In Arizona we would expect closer to 20 tons.

The biggest issue I encounter on projects like this is series rating the panels. In my area the available fault current at the service would be around 36KA.
Click to expand...

What size transformer would the POCO be providing? Many places they would only use a 50 or 75 KVA if possible, but even if they used a 100 KVA the available fault current at the transformer terminals should only be about 22KA.
 
kwired said:
What size transformer would the POCO be providing? Many places they would only use a 50 or 75 KVA if possible, but even if they used a 100 KVA the available fault current at the transformer terminals should only be about 22KA.
Click to expand...

True, but where I live the POCO tables say 167KVA for a single phase 120/240V service. Not that I ever see a 167 in real life in residential areas, but we are required to use the POCO tables for transformer size and for the available fault current at the service. We are not allowed to calculate fault reduction in the service drop or service lateral.
 
gar said:
Why a 385 A load in a home?
Click to expand...

I would bet because that is what the required calculations in NEC Article 220 come out to.

Keep in mind this is just a calculated load and has very little to do with the real load the service will see. The power company often figures about 50% of the NECs calculation and the power company has very good records of what buildings will really use.

You could do a load calculation for your own home and compare it to real time readings.
 
I am glad most of our clients group the panels in just a few locations and if I built my own large home the panels would all be in one spot. There is almost no such thing as a long run in a home.

We have a newer client with very large stores and they drop panels all over the place along with bus ducts on the ceiling set up as 42 circuit branch circuit supplies.


Makes servicing a nightmare.
 
hmspe said:
True, but where I live the POCO tables say 167KVA for a single phase 120/240V service. Not that I ever see a 167 in real life in residential areas, but we are required to use the POCO tables for transformer size and for the available fault current at the service. We are not allowed to calculate fault reduction in the service drop or service lateral.
Click to expand...

You did not mention the load, surely they don't just supply 167 to all single phase services? I have seen many more 15, 25, and 37.5 kVA transformers than 167's. The few 167's I do see actually have pretty significant load. Many 600 amp single phase services I do see are only supplied with 75 or 100 kVA units even though 100 kVA is only 415 amps. They will risk overloading it for certain amount of time too, as long as it has time to cool before being overloaded again.
 
kwired said:
You did not mention the load, surely they don't just supply 167 to all single phase services? I have seen many more 15, 25, and 37.5 kVA transformers than 167's. The few 167's I do see actually have pretty significant load. Many 600 amp single phase services I do see are only supplied with 75 or 100 kVA units even though 100 kVA is only 415 amps. They will risk overloading it for certain amount of time too, as long as it has time to cool before being overloaded again.
Click to expand...

I looked up requirements rather than going by memory. Most transformers I actually see are 50 or 75. Most supply multiple residences. For a 600A single phase service one utility says 42K mains and 30,405A for arc flach calcs with a 100KVA transformer. The other says a 167KVA transformer with 31,094 Isc. I don't agree with the approach, but in this case the POCOs makes the rules. It is possible to get a an actual design from the POCO but it generally takes 6 to 8 weeks, and someone has to sign up to cover any costs. With a standard turn-around time for a residential design of under a week the only real option in most cases is to use the tables.
 
hmspe said:
I looked up requirements rather than going by memory. Most transformers I actually see are 50 or 75. Most supply multiple residences. For a 600A single phase service one utility says 42K mains and 30,405A for arc flach calcs with a 100KVA transformer. The other says a 167KVA transformer with 31,094 Isc. I don't agree with the approach, but in this case the POCOs makes the rules. It is possible to get a an actual design from the POCO but it generally takes 6 to 8 weeks, and someone has to sign up to cover any costs. With a standard turn-around time for a residential design of under a week the only real option in most cases is to use the tables.
Click to expand...

A big factor in the short circuit current department that they apparently are overlooking is the conductor size, type and length. You could have a smaller transformer closer to the load and have a higher fault current level.
 
Status
Not open for further replies.
Top