three phase vs. single phase light load calculations.

[HighPotter]

Had a bit of a "conversation" with an engineer today, regarding his sizing of OCP and wire for a lighting project.

Project involves(2) 100' light poles, with (12) 1000w hps light fixtures on each tower. These are the " Halophane" drop ring style tower and lowering device. The distance is 250' from the 480 3 phase panelboard to the first tower, then 500' more to the last tower. At each tower base, is a backboard with a 60 amp 3 pole 480 v fusible disconnect tapped from the (3) 2/0 copper wire that is run from the main panelboard to the first tower and then on to the 2nd tower.

He calculates the connected load per tower at 25 amps " single phase", because he says that even though the lighting is balanced over 3 phases, the lights themselves are single phase. Or 31.25 at 125%

I calculate the load per tower at roughly 14 amps -3 phase. Becaus the feeder IS 3 phase. Or 17.5 at 125%

As you can see, the difference in these calculations effect everything from wire size, to ocp to disconect size.

Have I been wrong all these years?


HP



HP

 

[alehman]

Assuming the lights are balanced between three phases:
12,000 VA / 480 / sqrt(3) = 14.4A

Realistically, you must include power factor, ballast factor and voltage drop in your calculations. Circuits must be sized for 125% of the continuous load. Since the circuits are long, you should check the voltage drop.

 

[HighPotter]

Thank you.

Yes, I agree that they should be calculated at 125%. What I don't agree with is his 30+ amps calculation. I do not see where he is getting this. This is not a single phase calculation.

HP

 

[alehman]

Are the lights wired line-to-line? It wouldn't affect the result but maybe that's confusing him.

 

[BJC]

We need more info. 30 amps may be fine. Are the indivual fixtures fused? If they are it makes sense to over size the fuse at the disconnect. A use thats too small at the base of the tower is going to put the whole towere out if one fixture goes bad.
What is the temperature range where these things are installed?
As long as the wire up to the fixture ring is protected I don't see a problem. I would rather wait till morning and refuse and or replace one fixture than lower the ring at night in the dark and try to figure out which one shorted.

 

[HighPotter]

Yes, the fixtures are wired 480v, phase to phase, there is no neutral.

No, the individual fixtures are not fused. Each tower has a 3 pole fusible disconnect, contactor, and timer at the base.

The temp range is mild. 30F to 60F in the winter. 60F to 90F in the summer.

HP

 

[BJC]

Are they "internally" fused?

Do you have a catalog number?

 

[EEJaime]

You should be aware that this type of lighting fixture can be obtained with 480v, 3ph ballasts if ordered that way. However, even if the ballasts are 480V, 1ph, the system will see a balanced three phase load. If you are sizing the riser conductors from the base of the pole to the fixture heads, the calculation is single phase, or for a 1000w. fixture about 2.5 amps assuming a 20% ballast loss. The pole connection will be 12 x 1200va = 14.4 kVA or 17.3 Amps at 480v, 3ph. so a 25A., 3-pole fused disconnect is the minimum requirement, (17.3 x 1.25 =21.7A). This, as mentioned by others above will need to be adjusted for other factors, which may affect the conductor loading. Don't forget the load for the fixture lowering device if a motorized device is used, (may require independant connection, transformer, etc..., will not be operated with fixture load on).

Good Luck

 

[HighPotter]

Thanks for everyones response.

BJC...here is the link the the lighting

http://www.holophane.com/product/pdfs/HL-1998.pdf

1000w hps 480v

As it stands right now, the engineer is staying firm. He still contends that the calculations are to be based on 480 single phase, because the light fixtures themselves are single phase.

I have never seen anypne calculate a bank of 480v lighting this way. It baffles me.

HP

 

[stevenal]

Here's you ballasts info:

http://www.holophane.com/product/pdfs/HL-301.pdf

2.3A at 480V. This takes into account the ballast efficiency and power factor alehman mentioned. Looks like a 20A circuit would cover it.

 

[HighPotter]

Thank you.

Now my question is...

Who here feels that taking into account the info I provided.... that the load calculation should be based on 480v single phase, or 480 three phase?

HP

 

[BJC]

480 or 277 volt It doesn't matter. The steady state load will be as calculated ( with allowances for efficiency and power factor). I don't have any ideal what a three phase ballast is, if EEJamie can post the catalog reference to one I shall be grateful for the education.
The Brochure you referenced doesn't tell me if they have internal protection or fuses. We don't know what the cold starting characteristics of the fixtures is and how long it last under volage/temperature conditions. My experience is that you pick a ballast based on conditions and what you want. Fuses are an option you have.
I am curious about your "conversation". Whats your stake in the installation? Was this all the results of an argument of somekind or did he freely share his information.
As stated I think the installation as you have described it is fine.

 

[HighPotter]

My stake in the installation is as follows....

Our firm was awarded the project ( which actually involves several of these installations). The county ( the "owner") is now asking us to " value engineer" the project, to free up capital for some emergency work that needed. Our firm is very well versed in these installation, and feel there is an inherant flaw in the calculations. Which adds unneeded cost to the project. My conversation with the engineer discussed these topics. But we could not get on the same page. He continues to feel that all the lighting calculations should be based on single phase. I disagree completely.

With his calculations, 30 amp contactors become 60 amp. 30 amp disconnects become 60 amp. #2 copper becomes 2/0 copper.

The fact is, I will be able to prove my point at the end of the project, with a simple amp meter.

Thank you for the continuing discussions.

HP

 

[BJC]

I can understand somegetting defensive in a value engineering situation. 18 -20 amps is somewhere in the range of normal steady state load. The exact characteristics of the ballast, variation in voltage, operating temperature etc. can change a lot.
If you don't like the design pay the guy off and hire someone else to do it.

 

[advidana]

he is right. the lights in the bank are factory connected single phase not three phase therefore the two loads are single phases that is split accross the three phase source. one phase will carry the total single phase current the other two will be less. These loads cannot be balanced unless you add one more tower with 12 lights.

 

[alehman]

advidana - you lost me. Are you saying that all 12 lights on one tower are connected line-to-line between just two phases? If so, how do you know that?

 

[advidana]

experience

 

[RalphChristie]

Advidana:

480V 3phase

On a bank:
6 lights between phase A and phase B?
6 lights between phase C and phase B?

Although possible, why doing it in that way? Why not balancing it between the three phases? Maybe a centre-tap transformer, although I was under the impression that it was a balanced three phase source?

 

[BJC]

Highpotter
Your right about the wattage per ballast, 12 per pole 4 between each phase, total about 16 amps on a 3 phase 480 volt ckts.
I have been in on things like your describing and know sometimes more goes on that gets told, especially in a forum when someone is asking for backup for a valuing engineering decision.
A typical process goes like this:
Owners architech - We need an electrical design for permit ASAP, it's an easy project the owner wants to duplicate his installation at location "X" ( or that company XYZ has at podunk), same layout same number of poles, same type etc. Hurry and get the design so we can get the permit.
The decision is which two of good, cheap or fast do you want.
Good ( good to them meaning just like the job they wanted duplicated ) and fast are what they ask for so they get it.
In the mean time the owner finds they don't have as much money as they thought, lighting vendors submit ( and the owner approves ) different fixtures ( and maby a differnet layout ) than were originally requested.
So the owner went cheap and now wants things redesigned, I have been there and done that, did it free once but no more.
My point is we seem to be getting limited information to justfy a value engineering decision when we don't know all the facts. The history of the project may tell us more. Most engineer don't take projects with an understanding that they will continually work on it. If there is better information available now than in the beginning of the project pay the guy to redo it, if he's incompetent hire someone else to do it and sue him.
Things are not always as they appear. For example the 2/0 cable, is it in conduit or direct burried? If it's dirrect burried 2/0 may be a good choice as burring small wire is a waste of time.

 

[EEJaime]

This is a very basic installation. It should not require this much analysis. From the hundreds of installations I have designed, seen installed and observed, I would think that you would have the following:
1)Each mast will have 12-1000w HPS fixtures four will be connected between phases A-B, four between B-C and four between C-A. The SINGLE PHASE conductors from the base of the pole to the fixtures may be individual pairs for each fixture or a single pair of conductors for each group of fixtures on each phase pair. Since the load on four fixtures will be about 10.5 Amps, #12 AWG riser conductors are sufficient.
2)At the base of the pole each set of risers will terminate on terminal blocks, control contacts, or whatever means is selected to make the terminations. These six sets of single phase conductors will end up paired at two per phase to the three phase incoming pole feed. Note that each of the single phase risers will require fusing to protect the conductors which will be smaller than the incoming distribution conductors. You can use either a fusible disconnect, (30A, with 25A fuses) as a main, or a small load center with a 25A main circuit breaker and three 15-20A, 2-pole circuit breakers to feed the riser conductors, if fuses are not desired.
Whomever is telling you to calculate this at single phase is either assuming you are taking your three phase distribution system and feeding an entire pole assembly with a single 2-conductor, single phase, connection, (which will minimize hardware at the poles, but increase feeder conductor sizes), or does not understand the installation.

IF you were to take a three phase distribution feeder on site and from this run only a single phase-2 wire feed to each pole assembly, then you would have 12 fixtures x 1250 VA/fixture = 15kVA at 480V/1ph or 31.25 Amps, which would require a 40A feed with #8 AWG incoming conductors. The system as a whole, (assuming there were enough of these poles), would balance eventually if the phases were alternately used, but it's just another way to skin the same cat. Costs will vary depending on distances, quantities, stock availability, etc..., but all in all it's a very plain and simple cat.