LED commercial lighting retrofit , always keep 30% depreciation in mind

Status
Not open for further replies.

Electric-Light

Senior Member
CREE, a well respected manufacturer of high power LEDs has entered lighting fixture market and they're trying to gain market over fluorescent in commercial lighting. A few years ago, they only made LED chips. Now they make complete LED luminaires in their own name.

CREE is trying to gain name recognition from those who embrace technology through participating in social network sites, such as Facebook, Twitter and posting YouTube videos.

They have generally poor attitude about existing dominant technologies. While its okay that they promote their product, I feel that they're misrepresenting existing technologies.

CREE troffers allow skipping of one group re-lamping. Maintained efficacy is not quite as good as fluorescent. Upfront cost is about 3x as much. Quality of light is comparable to best available T8.

Example 1: agenda against fluorescent lighting
http://www.youtube.com/watch?v=nbMyY59x8Cs
This video portrays fluorescent lighting as flickery, nasty cold colored, poor color rendering light. Something that is stereotypical to older cool white lamps using halophosphate phosphor blend and magnetic ballast operation that hasn't changed much in half a century.

T12 lamps are still available in energy saver version in common commercial lighting sizes. 4100K 62CRI, 34W 48", 60W 96" and 95W 96" HO. These are getting banned in July 2012 and newer fluorescent lamps are substantially better, so its hardly fair to compare against the old tech. Magnetic ballast fixtures are already banned.

New fluorescent lamps have much better color rendition at about 85. They are available in quite a few selection of color temperatures. Warmest being soft white/2700K and "coolest" being daylight 6500K and 3000K, 3500K, 4100K and 5000K to go in between the two extremes.

Lithonia ES8 fluorescent is around $110/ea online, and offers complete luminaire efficacy of 86 lumens per watt.

http://www.acuitybrandslighting.com/library/LL/documents/SpecSheets/2ES8P-2x4.pdf

CREE says the pricing is set by distributor and claims their troffers are priced competitively with "architectural grade" fluorescent fixtures. Some fancy 2x4 fixtures that look similar to the CREE troffer cost over $400. I really doubt the CREE troffer is $110.

KEY POINTS:
Fluorescent 2 lamp 2 x 4: $130 or so with lamps. More for decorative models.
Lamp life: 30,000 to 40,000 hours. Fluorescent lamps lose about 5% output over lifetime
86 lumens per watt. CRI: 85 Brand new, its not quite there with CREE LED's specs, but it does not degrade as much.

CREE LED troffer:
Probably $350+. 3500 or 4000K only, 90 CRI.
Efficacy is 90-110LPW on the day its commissioned, depending on the model chosen. The same models become 63-77LPW after 50,000 hours.

Also, 110LPW model is expected to cost more than 90LPW model.

fixture life, estimated by extrapolation to 70% output to 50,000 hours. Power consumption remains the same. LED fixtures are rated to lose about 30% over their lifetime, more if temperature is higher. THIS REALLY IS *A LOT* Even F40/CW from 40 years ago didn't lose that much, according to 1972 edition of IESNA Lighting Handbook

Both fluorescent and LED system suffer from LDD from dust accumulation in optics and require cleaning to maintain maximum fixture efficiency.

Example 2: LED is super cool running deception
http://www.youtube.com/watch?v=LyVU470TQBk
Incandescent lamps shed most of heat through infrared energy.
Many MR-16 lamps are available with reflector that lets the infrared leave through the back to reduce heat damage to the items being lit.

While CREE LED lights do not shed much energy as infrared, they're far from super efficient. They dump a substantial portion of input power through conduction through massive heatsinks.

PAYBACK CALCULATOR is messed up. Payback happens sooner when you over-estimate the maintenance cost of competing technology and assume that bank lends you money at zero interest.

http://crseries.creeledlighting.com/#payback

Conveniently omitted...
When each fixture is estimated to cost $220 more than a common high efficiency fluorescent fixture, how does this affect the overall project cost? Where does the cost difference come from and what about the opportunity cost or financing cost related to the higher upfront material cost?

Reality is not based on simple-payback-method calculation. If you have the money, there is opportunity cost to this spending. If you borrow it, the lender will charge interest.

The CREE Payback calculator assumes too short of lamp replacement interval for fluorescent benchmark and doesn't address the issue of financing cost associated with higher upfront cost.

over estimation of maintenance cost of fluorescent system used for comparison
Old magnetic ballast system usually lost two lamps for every failed lamp as they are series wired. This reduces group re-lamping interval. Lamp rated life is around 40,000 hours for high-end lamps when they're used in typical >12hrs/day commercial use. The relamp interval is around 70% of rated life for parallel wired instant start, so that is 28,000 hours. CREE payback calculator is downplaying this by about 45% using approximately 15,500 hrs for interval.
 
Last edited:

mivey

Senior Member
Thanks for the information.

I guess to make LEDs more economical in the long run the government will have to subsidize the costs like they have done with solar power. It would be interesting to see who is really behind the wheel of some of these technology pushes.
 

mivey

Senior Member
Hmmm. How about some disclosure then EL?


Add: I do think the LED stuff is over-hyped in many applications.
 

Electric-Light

Senior Member
Thanks for the information.

I guess to make LEDs more economical in the long run the government will have to subsidize the costs like they have done with solar power. It would be interesting to see who is really behind the wheel of some of these technology pushes.

At this point, I don't see any basis for tax payer subsidy since there is no energy advantage.

The best LED fixtures have higher efficacy than relatively common moderately priced high efficiency fluorescent fixture out of the box, but the higher output depreciation makes them about the same, or less efficacious than fluorescent over the course of useful time. CREE's own datasheet shows that the decay is most severe during the first 6,000 hours.
www.cree.com/LM80/

If there is to be consideration for mercury elimination, then government is expected to provide same kind of subsidy for replacement of lead acid batteries with lithium ion phosphate type battery which is available as drop-in replacement for lead-acid for elimination of lead.
 
At this point, I don't see any basis for tax payer subsidy since there is no energy advantage.

There may not be any "energy advantage" if you leave them on as much as florescents, but with some of the newer control systems you don't have to. How about one motion sensor per fixture, instead of per room? There are all kinds of interesting things you can do then. Add a light sensor and do some daylight harvesting. All of a sudden, the LED system is at least on par, and possible better, than any classic florescent installation. And IMHO the LEDs themselves will continue to get better and drop in price.

(Disclaimer: my employer makes such a control system. PM me if you're curious about it.)
 

Electric-Light

Senior Member
There may not be any "energy advantage" if you leave them on as much as florescents, but with some of the newer control systems you don't have to. How about one motion sensor per fixture, instead of per room?
Sure, but these controls are not exclusive to solid state source excited fluorescent luminaires, aka LED lighting. I'm not sure where you're getting that idea.

You can implement the same controls for traditional fluorescent system. Instant start electronic ballast is spec'd for applications where lights will be left on for over 3 hours at a time, but if you intend on cycling them often with sensors, all you've got to do is spec out programmed rapid start ballast which adds like $5-10 per fixture and 2-3W more per ballast.

There are all kinds of interesting things you can do then. Add a light sensor and do some daylight harvesting. All of a sudden, the LED system is at least on par, and possible better, than any classic florescent installation. And IMHO the LEDs themselves will continue to get better and drop in price.

(Disclaimer: my employer makes such a control system. PM me if you're curious about it.)
If so desired, dimming can be implemented too. 0-10v analog dimming control is readily available. They're used in conference rooms and class rooms for visual comfort while presentations are taking place, but just like LEDs, they're costly, so they're not widely used.
 

Little Bill

Moderator
Staff member
Location
Tennessee NEC:2017
Occupation
Semi-Retired Electrician
I am sure they are, that is how it always goes.
Actually LED is looking better and better every day and will surely take over as THE dominant light-source in the near future. Just a few months ago I felt less confident about his, that is how dynamic this market is.

To bad the induction light became such an orphan, but it seems that the newer generation of LED's will surpass the efficacy and are dimmable.

One major missing point in all comparative calculations is the actual heat rejection that requires additional air-conditioning - one system over the other - for commercial and institutional indoor installations.
 

Electric-Light

Senior Member
There may not be any "energy advantage" if you leave them on as much as florescents, but with some of the newer control systems you don't have to. How about one motion sensor per fixture, instead of per room? There are all kinds of interesting things you can do then. Add a light sensor and do some daylight harvesting. All of a sudden, the LED system is at least on par, and possible better, than any classic florescent installation. And IMHO the LEDs themselves will continue to get better and drop in price.

(Disclaimer: my employer makes such a control system. PM me if you're curious about it.)

The newest CREE troffers don't offer full range dimming at this time. Only two-step 100%/50%, but to be introduced is 0-10v dimming.

The shorter the on-time, the longer the calender year payback period.

All those motion sensor based controls and possible dimming option can be implemented in fluorescent system with the same amount of work.

Sensors and control wiring adds the same amount of cost whether its LED or fluorescent system.

Dimming ballast in each fixture adds about $50/unit. The efficiency penalty is 4.3% reduction. The power usage goes down with dimming, however how far into dimming it maintains proportional ratio, I'm not sure, either LED or fluorescent.

GE dimming two lamp T8. 56W BEF 1.57
GE instant start 53W BEF 1.64

It's possible to use a feedback loop to compensate for lost light due to lamp depreciation. LED requires initial oversizing by 1.43x(30% depreciation). Fluorescent by 1.064x (6% depreciation)

On the day it is installed, both fluorescent and LED provides the same output and same FC at the floor. Near the end of life, feedback loop ensures they're providing the same output, but you'll find the fluorescent increased in energy use by 6.4%. LED by 43%, because LEDs decay more, it needs greater amount of compensation.

To bad the induction light became such an orphan, but it seems that the newer generation of LED's will surpass the efficacy and are dimmable.

One major missing point in all comparative calculations is the actual heat rejection that requires additional air-conditioning - one system over the other - for commercial and institutional indoor installations.

Fluorescent and LEDs are both dimmable with paired with an appropriate dimming control. The CREE troffer doesn't offer full range dimming right now. 3-100% range dimming is already available with fluorescent using 0-10v control if so desired.

LED and fluorescent are about the same efficacy at the beginning, so the heat load into HVAC is about the same for LED and fluorescent. With over spec'd fixture to make sure same output is maintained for the life of fixture, fluorescent grows 6% in power use. LEDs grow 43%. (Where 1,000 lumen is needed, 1,430 lumen LED is spec'd and they operate at 70% output initially. As the LEDs decay to 70% of original rating, they'll be cranked up to maximum output, which would have produced 1,430 lumen brand new, but near the end of life, 1,430 x 0.7 = 1000lm)

I personally think that active feedback to compensate for lamp lumen depreciation is not needed for fluorescent where LLD factor is about 0.96 (96% maintained over lifetime) but it is necessary for LEDs which has an LLD factor of 0.70.

Feedback loop works like the cruise control. Near the end of life, LEDs are not putting out anymore light, yet they're using 43%more power, because the feedback is compensating for the lamp decay which caused them to become less efficacious by 30%. If feedback isn't used, it will be providing reduced output while using the same energy
 
Last edited:
Status
Not open for further replies.
Top