Are there appreciable efficiency gains from 480 V vs 277 V 1

[wuddog]

I am not an electrical engineer.
Now that I got that out of the way, I have been asked if there is an efficiency advantage between 480V vs. 277V. Intuitively I said yes due to the lower amperage of the 480V, but I was then asked if I could prove it with numbers. At this point I am stumped.
The issue is that we have lights in a parking lot running at 277V. We are going to change them to LEDs that can run at 480V or 277V. The impetus for this change is obviously energy savings, but if we can run at an appreciable more efficient level at 480V we may want to do that. Keep in mind that we will be using the existing wiring which was for 277V so the wires will be oversized I would think. Is there a way to calculate or otherwise show numerically how the efficiency will differ?
I welcome any and all feedback.

Thanks

 

[davidbeach]

You'd probably have to run new wire. At 277V the white/gray wire is shared among multiple circuits and doesn't need a circuit breaker. At 480V it would need to be some other color, would have to non-shared, and would need a pole on two-pole breaker along with the other wire. Stay at 277V and your life will be much easier. Change to 480V and maybe your improved efficiency will pay for the voltage change by the time your grand children retire.

 

[cranky108]

The fact that you are changing to LEDs should also reduce the current.

In new instalations you could justify changeing to 480, but if everything is in place there just would not be enough savings to pay for the labor to make the change.

Besides, if you have 277 already, I would suspect somewhere you also have 480 available, just not wired to the parking lot.

 

[Mbrooke]

Higher voltage without a doubt is much more efficient even with the same current. Being approximate: A 240 volt line at 16 amps can only go half the distance as apposed to a line at 480 volts 16 amps. Now if you halve the current for the same gauge (mm2) at 480 volts (8 amps) the distance now quadruples for the same voltage drop.

480 volts also has the advantage of dispensing the neutral which eliminates the hazard of an open neutral condition. An open neutral can destroy at least 1/3 of the equipment after a break point and all of it under the right conditions.

Around here the defacto standard 99.99% of the time for parking lot and road way lighting is 480 volts 3 phase 3 wire. The only disadvantage is a live screw shell with HID CWA ballasts, but its assumed that you never re-lamp live. That problem is eliminated with LED and CWI / 3 coil magnetic HID ballasts.

This is a good read and has a calculator worth trying out:

http://www.electrician2.com/calculators/vd_calculatoradv.htm

 

[Mbrooke]

* Also, I forgot to add: When a MWBC is run (neutral shared between phases) and the load is evenly divided among phases the voltage drop is equivalent to a 3 wire delta circuit of the same voltage. Because 480 volt ballasts cost about the same as 277 volt ballasts (at least my experience with medium and high wattage HID)not running a neutral ends up being cheaper. Ditto should a splice fail in a Jbox.

 

[wuddog]

Thanks folks. Our main is 480 then it transforms to 277. We would bypass the transformer to achieve 480 throughout the system. Sounds like we may have an argument for going 480 based on your feedback. I appreciate the help.

 

[Mbrooke]

Welcome!

Is the main 277/480 or only 480?

 

[waross]

The wiring is in place for 277 Volts. 277 Volt circuits sometimes run with a one pole breaker. 480 Volt circuits need at least a two pole breaker.
277 Volt circuits must use an "identified wire" for the neutral conductor. The most common identifier is white colored insulation. The identified wire is not allowedd to be used for hot wires, it must only be used as the neutral.
480 Volts generally does not have a neutral, so all the white wires will have to be changed to colored wires.

If you have four wire 277 Volt circuits, there will be little difference in efficiency compared to a three wire 480 Volt circuit.
The load current per phase will be the same for a four wire wye connection as for a three wire 480 Volt connection.
There will be some slight added losses due to harmonics on the neutral conductor for the 277 Volt circuit.
If there are lockable single phase disconnect switches for Lock Out/Tag Out (LOTO) they may be single pole for 277 Volts but must be two pole for 480 Volts.

If the 277 Volt system has a lot of two wire circuits, all the white wires will have to be replaced with colored wires. That may be half of the installed wiring.
Also if single pole breakers are used for the 277 Volt circuits, they will have to be replaced with two pole breakers. There may not be enough room in the panel for the added breaker poles.

The losses are due to resistive heating and are related to the voltage drop in the conductors. The code limits the maximum voltage drop. Designers generally use a lower voltage drop figure than the code allows. So the voltage drop and the losses in the longest circuits is probably around 2% and less on the shorter circuits.
So we are probably at or above 98% efficiency in the supply wiring. Changing that may increase the efficiency to 99%. How much labor do you want to pay for to improve 99% energy.
But, without looking at the plans of your installation and doing the math on losses and savings, "It depends".

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Mbrooke]

The un-emended NEC has no voltage drop requirements. It has recommendations, but they are not mandated:

http://ecmweb.com/site-files/ecmweb.com/files/archive/ecmweb.com/mag/406ecm06fig2.jpg

Technically strictly following the code one does have to change Grey or white to another color if not part of a listed cable assembly, but I see no reason why the grounded conductor could not be re-identified with orange, yellow, ect tape. I know its not following code word for word but it is something done here all the time with no issue as the conductor is rated 600 volts to start with.

 

[davidbeach]

The reidentification is the easy part, the need to go from single pole breakers to two pole breakers is a much greater challenge. It won't pay to make the change.

 

[waross]

"The reidentification is the easy part,"
Not always.
We had several installations where a ground conductor could be bare or green colored. The clearances were adequate but tight. We had bare wire on hand, but did not want to buy a roll of wire when we only needed about 40 feet total.
I had the bare conductors taped green for some added safety. The AHJ would not accept the tape but would accept the bare wires.
If the AHJ agrees ahead of time that taping is acceptable then no problem, but if the wires are taped and the AHJ does not accept the tape, then your costs just took another big jump as ours did.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[cranky108]

Consiter where there are several 277 circuits in a conduit, assuming they are from different phases, you would only need one neutral. This neutral would only need to be sized for a single circuit current return, and if closely balanced would carry very little current.

On a 480 circuit, there would be no current carrying neutral, so you would need additional wires, and have added losses because of the return wire impedance.

So while a 480 circuit has less current, it also suffers with return losses that a 277 circuit could have avoided. Also you system is sized for the current at 277, and removing those wires, and pulling new wires will cost something in labor. Where the sale of the larger wire would net very little. So the cost savings from using a smaller wire is a negitive in an existing instalation.

Also the added losses from the return current on the 480 would reduce much of the savings vs the current 277 system.

In conclusion, the savings for a higher voltage at the 600 volt class, only applies for new systems, and rarely is seen in changing existing systems.

 

[jraef]

Davidbeach and waross are right. BECAUSE you are already set up and wired for 277V, it would be expensive to re-configure the wiring and switchboard for supplying 277V, and to "watt" end? The load is expressed in watts and at any voltage, even if the current changes, watts = watts = watts. Changing the voltage does NOT change the watts consumed by a load, look at your fixture data sheets and you will see.

Increasing the supply voltage can ONLY possibly reduce some LOSSES in the transmission system. So in your case the only ENERGYyou might save is in slightly reduced copper resistance (I[sup]2[/sup]R) losses inside the conductors themselves, and even then, only if you have a somewhat severe voltage drop problem now. I[sup]2[/sup]R losses are also, by definition, based on the CURRENT (I) flowing in the circuit as well. Right now you have current flowing that serves 277V ballasts for (presumably) HID lighting, probably High Pressure Sodium (that slighly yellowish light color). Those are typically low wattage to start, lets pick a common parking lot fixture of 70W per lamp. The input power draw on an HPS fixture for that would be about 85W, and for an equivalent LED driver and lamp setup to give you the same lumen (light intensity) output, the input watts will be about 50W. That is likely why you are doing this, so bravo. But IF you had so many of the HPS fixtures on a circuit that the current draw caused a significant voltage drop, you are now going to reduce that current draw by 47% just by changing the fixtures. I would be HIGHLY surprised if this alone didn't cure any VD problems you may have been having, but it also means the LOSSES you might have had in the existing system will be so negligible now that they are almost not worth worrying about. Sure, overkilling that concept by using 480V will technically reduce it even more, but it just means going from negligible to a little extra negligible. Not enough gain to be worth having to strip out the existing distribution equipment and replace it, then trying to pull new conductors into existing conduits under ground (most likely). One mistake that causes the crew to have to get out the backhoe and tear up the asphalt, and any possible energy pennies you saved for the next 20 years go right out the window...

Also, being that you are not an EE, I doubt that you truly understand what you have for a supply system. Since we start off discussing 480V, you are likely in North America somewhere, 80% chance the U.S. In that case it would be highly unlikely that you have a 480V delta service from your utility, because as a general rule that was only given to industrial manufacturing facilities who wanted continuity in the face of a fault, which would likely mean you would have an EE on staff and would not need our help. So it's far far far more likely that you have a standard 3 phase 4 wire 480/277V service drop, and that your 277 for the lighting is derived from having a 1 pole circuit breaker feeding one leg of the service plus the grounded neutral conductor. So there would be no efficiency gains from "removing the transformer".




"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[Mbrooke]

That's what makes me wonder, what transformer is the OP talking about? And how many breakers feed the lighting? Jraef is correct, unless industrial or a very, very old building (that will most likely be industrial) a 480 volt delta service is not likely. The only exception I could think of is a panel that does not have a grounded conductor run to it which would force a transformer for anything outside of straight 480 volts.


@Cranky108:

Consiter where there are several 277 circuits in a conduit, assuming they are from different phases, you would only need one neutral. This neutral would only need to be sized for a single circuit current return, and if closely balanced would carry very little current.

On a 480 circuit, there would be no current carrying neutral, so you would need additional wires, and have added losses because of the return wire impedance

Why would a 480 volt system need individual returns (I am assuming 6 wires) under the NEC?

It would be much cheaper and still code complaint (as long as the breaker is 3 pole with internal common trip) to combine those 6 wires into 3 wires that are slightly larger, ie normal current x 1.73. Lights would just connect in delta A-B, B-C, C-A.

 

[Sharpie8]

from an operating perspective (neglecting change costs) switching to 480V has higher efficiency due to the removal of the intermediate transformer and reduced I2R losses.

 

[davidbeach]

There's no intermediate transformer to remove. 277V is line to neutral, 480V is line to line, both from the same transformer.

 

[waross]

Hi David;
Look at wuddog's second post. I doubt that there is a second transformer but the OP's second post adds some confusion.
wuddog seems to think that a transformer may be eliminated????

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

I also find it hard to believe the site would have a 480V delta system and a step-down transformer to 277V (this would be single phase or an custom 3-phase??) when the lighting could have been powered from the 480V. A 480V/277V wye connected secondary makes much more sense.

Chasing I^2R losses in the wiring could be a waste of time. You need to define,
- length of wire
- size of wire
- average current in the wire
and then you can calculate the present losses, losses with the LED change and losses with the LED and 480V change. Once you know that, using the $/kWh and hours per day of operation can give you the saving per year in the wiring with each change. I think you'll find it's not worth the expense of pulling wires and adding more panel boards to accommodate the new two pole breakers (effectively doubling the current number of breakers).

Do you have a site plan that give the conduit runs or light locations that you could extract some useful wiring information from?

 

[waross]

Is this a legacy 440 Volt installation with a 440:480/277 Volt transformer installed at a later date to supply a lighting upgrade?
When delta services were common, the standard voltage was 440 Volts. I believe that the change in standard voltage was made over a 20 year period, starting in the 40's and running into the 60"s.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[cranky108]

Because we don't know, we can only guess. But many 480 V systems would have a transformer for feeding 208/120 volt loads, and this was typically a delta to wye.
I wounder if there is some confusion here with that.

277 V is commonly used in building lighting circuits because a single pols switch can be used to control the circuits. But if 480 is used it would require a two or three pole switch which is much less common.
Also a 277 circuit can only have a single phase to ground fault, which would only remove a third of the lighting. Three phase lighting would require any fault to remove 100% of the lighting.