Out with the offensive fluorescent lamps.

Status
Not open for further replies.
iwire

iwire

Moderator
Staff member
Location
Massachusetts
And in with glorious LEDs. :D

Removing about 75 of these 10 lamp (2x5) T5 HO fixtures.

These are 50 pounds each with the bracket that holds them together. At some point someone added motion sensors.

Topofoldhighbay1_zps9b62f87e.jpg


By my rough estimate a 614 watt fixture is being replaced with a 360 watt fixture. The workers say the new fixtures are substantially brighter.

IMG_20141229_090640_083_zpscc5e1f20.jpg


We are also replacing site lights and office lights. We have to package up all the hazardous waste for disposal. ;)

Lamps_zpsfe9dc4ad.jpg


Kidding aside it does get broken down for recycling.

Steel

Steel_zps8f48fde0.jpg


Copper

CU1_zpsb6151963.jpg


A disassembly area

Table_zpse1d7555b.jpg
 
150104-1536 EST

The wattage reduction is 614 - 360 = 254 W or 0.254 kW. Assume on for 3000 hours/year. Then energy saving per year is 3000 * 0.254 = 762 kWh per year. At $ 0.10 per kWh the cost saving per year per fixture is $ 76. Add to this the reduction in air-conditioning costs. And subtract the increase in heating costs.

.
 
gar said:
150104-1536 EST

The wattage reduction is 614 - 360 = 254 W or 0.254 kW. Assume on for 3000 hours/year. Then energy saving per year is 3000 * 0.254 = 762 kWh per year. At $ 0.10 per kWh the cost saving per year per fixture is $ 76. Add to this the reduction in air-conditioning costs. And subtract the increase in heating costs.

.
Click to expand...


I'm going with 254w savings as you did, but, i'm going with 5000 hours/year and i'm going with 0.13 per kWh because I know where he is so here is what I get. 1270 kWhr per year and $ 165 per fixture per year. Also, since you reduced it 1270 kWH per fixture x 75 thats 107,950 kWH direct reduction on HVAC also.

I'd guess those new LED ones are around $ 175 per so x 75 = $ 14,875 material and i'll guess labor out the door at $125 per fixture x 75 = $ 9,375

Total Cost $ 24,260

Savings per year $ 17,025

ROI = 1.42 year payback

If you add in rebates i'm thinking around $ 75 per fixture x 75 so $ 5,700.00 more

ROI after Rebate 1.06 years

Just a rough guess though
 
How long did the copper sit there? That wouldn't last one day on my job. :roll:

CU1_zpsb6151963.jpg
 
Nice, brings back some memories of a few large lighting upgrades I did. Both we stores maybe 60X100 ish. I started really getting speed down. After we mounted the new fixtures (grid surface mounted T8's) I taped the requisite conductors onto a fish stick and had someone push the scaffold down the line with me on top feeding the conductors through the whole line as fast as I could, then back and pull out a loop at each ballast, drive wirenuts with the cordless, repeat.

What are you doing/getting for the ballasts? We were sometimes getting "electric motor scrap" for them even though not all were magnetic - I think someone screwed up! Then other trips it would just be "tin" category which is lowest value mixed steel (even though tin is more expensive than copper, funny how the word has such a cheap connotation to it).
 
More site details need to be known before we can even come close on payback. A suggested 3000 hours a year means about 8.2 hours a day - that doesn't even consider weekends/holidays etc. where there may be reduced run times in at least some locations.

Then another suggested 5000 hours a year that would be 13.7 hours a day, again with no consideration for weekends/holidays.

Not saying these numbers are out of line, just that they can vary greatly depending on actual conditions at the site, some places could be less hours making payback longer, some could even be more hours accelerating the payback.
 
iwire said:
By my rough estimate a 614 watt fixture is being replaced with a 360 watt fixture. The workers say the new fixtures are substantially brighter.
Click to expand...

Light Emitting Decorations are almost always rated in INITIAL lumens even though many of them depreciate a ton like metal halide. Wording like "brighter" is possibly just an excuse for inadequate performance. High kelvin temps will appear "brighter" at lower illumination levels (like in night light or street lighting levels) whether they're LEDs or 6500K fluorescent. Glaring surface brightness also increases the perception of bright if the fixtures are within line of sight. A 1600 lm clear 100W light bulb will look much brighter than a five inch globe bulb putting out the same number of lumens when you look at the bulb.

You'll notice that HID and fluorescent systems are always designed using the MEAN lumens. It's a huge point of contention as fluorescent lamps have a much lower lamp lumen depreciation than LEDs, yet LED lighting is often done in "brand new out of the box lumens". Out with the offensive assumptions and awaiting glorious objective data. You didn't provide measured values from properly calibrated light meter or a drawing of uniformity of lighting. You said 5 lamps x 2? that's quite uncommon. You sure they were not 4 x 2 = 8 lamps?

The mirror reflector fixtures are usually pushing around 90% downward delivery. http://www.visual-3d.com/tools/photometricViewer/default.aspx?id=18603

This is a 2007 publication so we have a good portrayal of what was available eight years ago.
http://www.gelighting.com/LightingWeb/ind/images/GE_lfl_lamps_t5_wm_ho_sell_sheet_tcm288-37367.pdf

Lithonia FGB164 4 54T5HO T1X20 + eight General Electric 51W F54T5/841/WM/ECO gets delivered output of 36,000 lumens down the bottom new and 33,000 maintained using 432W input to drive 8 lamps. 93 LPW x 90% fixture x 90% LLD gets you 83.7LPW start and 75.3 LPW maintained. Many T5 ballasts officially permit partial lamp operation.

L70 for LED was chosen as the life for the argument "30% drop is hardly noticeable to people". If you step dim to 50% measured output, it will appear 71% as bright. The much more preferable solution is to produce light to the needed illumination level using a source with minimal depreciation so it maintains this level without the need for over illumination or a high level of power boost compensation.

If brand new LEDs are only matching existing fluorescent lamps, they're falling behind unless the rated life is based on L90 rather than the unreasonable L70.
 
Last edited:
Electric-Light said:
Light Emitting Decorations are almost always rated in INITIAL lumens even though many of them depreciate a ton like metal halide. Wording like "brighter" is possibly just an excuse for inadequate performance. High kelvin temps will appear "brighter" at lower illumination levels (like in night light or street lighting levels) whether they're LEDs or 6500K fluorescent. Glaring surface brightness also increases the perception of bright if the fixtures are within line of sight. A 1600 lm clear 100W light bulb will look much brighter than a five inch globe bulb putting out the same number of lumens when you look at the bulb.

You'll notice that HID and fluorescent systems are always designed using the MEAN lumens. It's a huge point of contention as fluorescent lamps have a much lower lamp lumen depreciation than LEDs, yet LED lighting is often done in "brand new out of the box lumens". Out with the offensive assumptions and awaiting glorious objective data. You didn't provide measured values from properly calibrated light meter or a drawing of uniformity of lighting. You said 5 lamps x 2? that's quite uncommon. You sure they were not 4 x 2 = 8 lamps?

The mirror reflector fixtures are usually pushing around 90% downward delivery. http://www.visual-3d.com/tools/photometricViewer/default.aspx?id=18603

This is a 2007 publication so we have a good portrayal of what was available eight years ago.
http://www.gelighting.com/LightingWeb/ind/images/GE_lfl_lamps_t5_wm_ho_sell_sheet_tcm288-37367.pdf

Lithonia FGB164 4 54T5HO T1X20 + eight General Electric 51W F54T5/841/WM/ECO gets delivered output of 36,000 lumens down the bottom new and 33,000 maintained using 432W input to drive 8 lamps. 93 LPW x 90% fixture x 90% LLD gets you 83.7LPW start and 75.3 LPW maintained. Many T5 ballasts officially permit partial lamp operation.

L70 for LED was chosen as the life for the argument "30% drop is hardly noticeable to people". If you step dim to 50% measured output, it will appear 71% as bright. The much more preferable solution is to produce light to the needed illumination level using a source with minimal depreciation so it maintains this level without the need for over illumination or a high level of power boost compensation.

If brand new LEDs are only matching existing fluorescent lamps, they're falling behind unless the rated life is based on L90 rather than the unreasonable L70.
Click to expand...

Blah blah blah.

Customer is happy, we made money. Nuff said. :D
 
iwire said:
Blah blah blah.

Customer is happy, we made money. Nuff said. :D
Click to expand...
I think this comment actually supports my position. There's no substantiated performance merits in using LED indoor lighting. They're like luxury chandeliers and that's that and if they want to buy it, that's their choice, however utilities, energy co-ops etc have got to be much more wary about handing out subsidies on arguably useless retrofits that could have been done without unnecessary high cost procedures.

A reduction in light output by 30% only drops perceived output by about 15%. Many T5 ballasts operate support delamping. In other words, same result was likely by shooting out 30% of lamps with a BB gun without the outrageous expenses LEDs that can only hold its own weight through sales pitch. :lol:

kwired said:
No pictures of finished product?
Click to expand...
What good would that do? Automatic exposure is a standard feature in all digital cameras. Deficiency in illumination level will get adjusted out.
 
Last edited:
Electric-Light said:
What good would that do? Automatic exposure is a standard feature in all digital cameras. Deficiency in illumination level will get adjusted out.
Click to expand...
Wasn't really why I asked, just wanted to see finished install. Kind of no different then those that post a picture of a panel with perfectly routed conductors in it though that doesn't really improve performance of those conductors to any significance.
 
At it again!

At it again!

So back to outside work, this one was a small lot with five old 400 Watt MH flood lights.

Pulled them down and replaced them with five LED floods, have not heard how they look yet. I am expecting them to be a bit darker but maybe I will be surprised.

Old

SOMFLold1_zps52513256.jpg


New

SOMFLNewFar_zps2471221c.jpg


New up close

SOMFLNewClose_zpsbe0402a8.jpg


Overall, the pole and base will be replaced in the spring.

SOMoverall_zps0eb060b8.jpg
 
Status
Not open for further replies.
Top