al borlan
Member
- Location
- Newport, Mi.
Now that LED lights are popular do you think designers of new construction will continue to use 277 lighting circuits?
LED lighting in new buildings
- Thread starter al borlan
- Start date
- Status
- Location
- Massachusetts
I would assume as long as it is available in the building they will.
If you need 480 for HVAC, elevators etc you might as well use 277 for the lighting. Might make them 15 amps and 14 AWG circuits though.
Electric-Light
Senior Member
Why would you think any different? The power requirements for traditional fluorescent and LED solid state fluorescent luminaires are about the same for given output.
Not everyone would agree with that comment.
And LED's just keep getting better and better.
I wouldn't say the same for fluorescent.
Electric-Light
Senior Member
Fluorescent fixtures can get delivered efficacy of around 90-95 LPW taking reflector losses into account.
What about LED systems? For the time being, about the same. You have to look at the maintained lumen.
Some are terrible and rated at 30% degradation @ 50,000 hrs.. and some like N80 nLight uses active LED degradation compensation to hold the same output by raising the power consumption.
no matter how you try to spin it, constant light output is a good thing.
Electric-Light
Senior Member
Linear fluorescent systems don't justify degradation compensation, because the lamp depreciation lumen is small enough (under 10%) to not have dramatic output difference during lifetime. It's possible to apply compensation using dimming ballasts with lighting control preloaded with RE80 degradation curve, but fixture cost will not stay at $8,000/million lumens
anymore.
An LED install that starts at 10kW/million lumens will consume 12.5kW/million lumens when LEDs have degraded to 80% of original output but kept at constant light level with a compensation system. Obviously systems that use active LED degradation compensation must size electrical wiring around the power requirements with LED decay comp. maxed out as the power usage goes up as they accumulate hours.
It is absurd to allow L70 as the threshold, because this represents a depreciation level far worse than F40CW
Last edited:
Electric-Light
Senior Member
Also, many electronic fluorescent and LED ballasts have a percent or so lower power consumption due to reduced losses within the ballast on 277v power than on 120v.
This doesn't include the added loss of running power through 480 to 208/120v transformer. The rule of thumb is always use the power closest to service voltage.
This doesn't include the added loss of running power through 480 to 208/120v transformer. The rule of thumb is always use the power closest to service voltage.
Last edited:
OK, lets look at the numbers.
Here is Lithonia's 2GT 8 fixture. 2 lamp T8. We are looking at the first photometric table:
http://www.acuitybrandslighting.com/library/ll/documents/specsheets/gt8-2x4.pdf
And here is their 2GTL LED fixture:
http://www.acuitybrandslighting.com/library/ll/documents/specsheets/2gtl 2x4.pdf
T8: They are using a lamp average output of 2850 lumens. Let's use the GEB 10IS option for the ballast. That has a 0.88 output factor, and the ballast will use 59 watts.
The fixture efficiency is 81.7%, so total average output is 2850 lumens x 2 lamps x 0.88 X .817 = 4098 lumens. That's 69 lumens per watt average, and for a relatively short life of 24,000 hours.
A new lamp puts out 3000 lumens. The best this fixture will ever do is 3000 X 2 x 0.88 X .817 = 4313 lumens, or 73 lumens per watt.
LED:
This is much easier to calculate. Let's use the 4800 lumen fixture with the 3500K lamp. That's probably the closest we can get to the T8, and I'm being nice because the numbers get better with a 4100K fixture.
The fixture starts at 4815 lumens and uses 47 watts. That's 102 lumens per watt. Let's be nice to T8 again, and assume this is the initial output, and doesn't include LED depreciation.
The LED's have 90% lumen maintenance at 60,000 hours. (With T8, our third lamp change would be half way to its demise). The LED fixture is putting out 4333 lumens, and still using 47 watts, for 92 lumens per watt.
The truth may hurt a little, but the LED is doing better after 60,000 hours than the T8 did on day one. In fact, the LED is doing almost 30% better at the end of its life than T8's best day.
And we didn't even look at T8's worst day - we used it average output, not the lower output that we would expect near the 24000 hour mark.
Electric-Light
Senior Member
That ballast would be in line with specs we had in year 2000.
They didn't use HPT8 in their testing.
http://www.grainger.com/ec/pdf/Philips-ICN2P32N-Spec-Sheet.pdf
0.89 per 56W with 32W lamps or 0.90 per 47W with 2700 lm 28W nominal lamps which is more in line with what we have today.
Now we're at 4,860 lm per 47W.
Which isn't the best available. 90-94% is what you can get on premium fixtures these days. If we use 92%, we're at 95 LPW. This is the biggest factor and it explains most of why two T8 lamps retrofit 4 lamp fixtures in many cases. Many old fixtures have TERRIBLE optical efficiency.
And that fixture has changed substantially in the last few years. The previous version was far worse than T8. So, the currently best available T8 and commercially common LED are about tie.
LED life is extrapolated a huge amount. 10,000 hours of known and 50,000 hours of guessing. T8 performance data is based on measured hard data from years of experience.
You're using standard life lamp based on 3hr/start. Rated life on long life T8 lamps are 42,000 to 75,000 hrs based on 12 hrs/start. The lamp lumen depreciation on HPT8 is very low. <10% over lifetime.
When you're considering a retrofit, it's reasonable to use the latest available options to select from.
Then in the end, LEDs end up far worse in $/million lumens of system output. LED systems that offer warranty conditions comparable to ballast warranty available for fluorescent systems. Those with similar warranty further drives up $/ML.
Last edited:
Fulthrotl
~Autocorrect is My Worst Enema.~
- Occupation
- E
yes. for the same reasons they use them now....
no transformer losses, a bezillion lights on a 20 amp circuit,
and almost every device being made is universal voltage.
The Lithonia SP8 has an 85.5% efficiency, but I don't know of any troffers that are higher rated. Open fixtures like high-bays are probably more efficient, but that would make this an Apples to Oranges comparison.
Using that more efficient fixture, and your lamp, and ballast brings T8 up to 4337 lumens at 48 watts for a new lamp. That's 90 lumens per watt initial, and 88 lumens per watt average.
That's a lot closer, but the LED is still doing a little better at the finish line than T8 is doing out of the gate.
I actually trust the LED depreciation numbers more than the T8 numbers. Many of the LED estimates have turned out to be conservative as they get more data.
And the T8 numbers seem to be really dependent on starting intervals - put a T8 in a toilet room with an occupancy sensor and you either have to use a rapid start ballast which uses more power, or the lamp life will tank.
Dollars per million lumens is a whole different subject.
I do agree 277 volts will still be used, and those are some good reasons. Even most of the sensors and power packs being made dual voltage 120 or 277 volts.
I'd also mention lower voltage drop on 277 Volt circuits.
Electric-Light
Senior Member
That's fair enough if you're limiting to reasonably common troffers. The 90-94% figures were for high-bays with a specular reflector. The lower the output per fixture, the smaller the gap between fluorescent and LED in $/million lumens. The lamp decay rate of fluorescent lamps is well established and highly predictable. It drops a bit in the first few percent of life and from there, it's very stable. Rare earth lamps are very durable. LEDs are less predictable. A major drop in rare earth prices over the last 2-3 years make fluorescent lamps more competitive.
You should confirm if LED fixtures use any sort of active degradation compensation to achieve the L90 rating. It's important, because, you need to know about intentional rise in kW/million lumen consumption.
Now, if we're gonna to play spec sheet war:
28W lamps on instant start ballast. I didn't include programmed start. They're not really in common use.
38,000hr (12hr) 32,000 hr(3hr) 2,725 new 2645 mean LM
http://download.p4c.philips.com/l4bt/3/332936/energy_advantage_t8_332936_ffs_aen.pdf
68,000 hr 60,000 2,650 lm 2,570 mean LM
http://download.p4c.philips.com/l4b/9/927853283601_na/927853283601_na_pss_aenaa.pdf
94% at 90,000 hrs
https://www.platt.com/CutSheets/Philips/P-6049-E 2XL EA T8 28W bulletin_v1_web.pdf
L70 rating for LEDs is inappropriate for indoor lighting, but specification should be as specific and demanding as possible. I think L70 or L80 are acceptable for household bulb or outdoor HID replacement. $/ML goes as you add more terms :happyyes:
"life shall be rated L90 xx hrs".
I think the difference is academic at this point since we're talking about extrapolated values.
Our difference of 90 LPW vs 85 LPW comes out to 11.75kW/ML vs 11.1kW/ML, which is only about 5%. Do LED sales people ever admit to this.... or are they even aware?
68,620kWh/ML/year vs 64,824kWh/ML/year. at 16 hours per day. The biggest savings are from cutting excess lighting level, and turning off when possible. The difference is only about $500-600 per year per ML.
So, selecting fixtures that cost an additional $10K/million lumens over 5% difference doesn't add up. Time value of money is EVERYTHING in real life. The maintenance savings is unknown. If they're going to need cleaning during their lifetime, slipping in group re-lamping is not a huge added cost. Also, LED ballasts are likely going to be more difficult to service as they're not standard stock item.
Stating the obvious but lower the annual energy bill, the less you're able to pay for capital improvements with saved bills. Repeating myself, you could save money either with a 35mpg or a 40mpg to replace your 12mpg guzzler, but there's a very good chance you can't pay off the price premium of 40mpg model with the additional gas savings you'll have over the 35 mpg model.
T8s have tens of thousands of real hours and actual used lamps have been available for testing. LEDs have not existed for the same duration and there is a lot we don't know about how they actually AGE(not ust output, but color change as well) in real clock hours. That's my justification, but to each their own.
The lumen drop is all about running hours. Each starting removes a bit of life from lamps, but energy use and usage hours stop when they're off, so you could still get a reasonable calendar years of life, which is relevant for ROI. If each start takes away one hour of life and you start it 10 times a day, you're still ahead if the total off time is long enough.
Something with a rated life that's 3x higher can not justify three times the upfront cost. We've had 0-10v dimming ballasts for a loooong time. But, they only sell a handful, because, they significantly increase $/ML. LEDs would increase this even more. Although, in competition with LEDs and some Title 24 requirements, step or fully ballasts have come down in prices.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Question, do they make 480 volt indoor LEDs other then Highbays?
Electric-Light
Senior Member
Possibly. There are a few 480v LED ballasts.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Electric-Light
Senior Member
What are you trying to do? wattage lumen application? The selection is quite limited compared to 120-277
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Something like a troffer.
Electric-Light
Senior Member
Couldn't find any. I guess not.
- Status