Translate

Thursday, December 27, 2012

Review, Home Depot/ CREE T67 LED Disk Light

At 7/5/2016, retract the warning below. Better T67 and T47 LED disk lights have been in Home Depot stores for at least one month. Improvements include elimination of buzzing with placement in a dimmer circuit, better efficiency in diodes no longer recessed, surely larger beam angle. The five tightly-placed diodes, identical in T67 and T47, are increased in size, from 3 mm square, to 5 mm square. Diode luminance is reduced, which should make lights more pleasant in ones' gaze. As measured with my Kill-A-Watt meter, T67 power draw at full brightness is reduced by improvements, from 12.5 watts, to 11 watts, as claimed on the package. I measure identical 11 watts draw for T67 and T47, as it should be, and in conflict with new T47 package claim of 9 watts.

Please see supporting web album:
Improved LED Downlights at Home Depot

I ponder my role as a self-appointed reporter. Obligation to be a responsible observer is negated by manufacturer action to offer improvements, and sometimes a degraded product, with no announcement. For T67, or for T47 , nothing on new packaging or at homedepot.com, informs the buyer of changes made. A reporter is needed when products are discontinued without explanation, as with wonderful and not-glaring Nicor DLS10 .
------------
At 6/29/2014, warn: don't buy Home Depot T67. Instead buy  Nicor DLS10  in online purchase. Nicor DLS behaves perfectly on any dimmer and has other advantages. At 7/14/2014, add this: Another good option is Sylvania 70732 . The Sylvania 70732, very nicely priced at Lowe's, has a major improvement in July 2014, now behaving as perfectly on a dimmer as the Nicor DLS. Manufacturers and their big-box outlets which don't respond as LED technology rapidly advances, deserve no sales.
------------

("New" was December, 2012, very long ago in the evolution of LED lighting.)
This post will present portions of a Picasa Web Albums review of Home Depot's new T67 LED Disk Light, Manufactured by CREE, and packaged as Commercial Electric. 

This is the promised replacement of T91 LED plate lights. There is a distinction between plate and disk, as developed in the previous post .  The Commercial Electric and Lighting Science Group packaging of T91/ Glimpse lights nowhere uses the word disk. I hope Glimpse packaging will adopt the label plate LED.

The T67 disk light  loses about 10% efficiency in presenting the diode phosphors within a cone above the ceiling plane. The light is thrown away for style and with misunderstanding. The volume of the diffuser cone could be employed to reduce the space in a junction box demanded by the power block.






Let us reserve the label plate LED, for presentation with negligible diffuser angle. And, editing this post at 3/10/2015, admit giving up on "LED plate light" as wordcraft. The phrase is securely related to license plate illumination. "LED downlight" may gain acceptance as the alternative for best-efficiency luminaires. In this, again exclude beam focusing in a cone.

Home Depot is again offering only one color temperature, this time 2700°K. I hope they will reconsider, and will offer instead a 3000°K product, or a hybrid where LED phosphors are of a sampling of colors, For now the block over the diodes will interfere with some found lighting junction boxes, where most boxes unfortunately, and perhaps needlessly, include posts for 2 3/4" oc light fixtures. Boxes knocked out of service are the good family of fiberglass boxes, by Allied Moulding. My preference is for RACO 175 steel boxes, with only the 3 1/2" oc screw positions, which work fine. 

Home Depot and CREE have put good thought into bug-proof simplified mounting.

I continue to develop my own experiments in comparative rating of brightness of LEDs and other light forms. Here in Google Docs is my latest effort, now including T67 lights. To view this, please download, and study on your own computer. Google Docs does a poor job of rendering. I now project lights indoors against a 4' square of drywall. The drywall has a satin finish application of very bright white paint, and yet photos capture general brown tones. I don't think brown tones are reflected from the OSB divider, as colors don't change much when a white towel is draped over the divider. The study is subjective, but I think it is consistent in comparisons, especially where two lights have the same color temperature. Whatever the flaws or naivete of my effort, it is better than accepting package lumens claims as statement of relative ability to produce useful illumination.

Here is the comparison of T67 and T91 brightness and color:
Album Caption:
Employing Brightness Numbers I advocate.

By lumens ratio, the T91 should be brighter by 15%. I call the difference 25%. If the T91 is B6.7, the T67 is B5. B5 is equivalence to a 125 watt incandescent light. It is interesting that the T67 package makes no statement of incandescent equivalence.




Here is the comparison of a new Philips more-efficient "100w" incandescent bulb, with the T67:


If T67 is B5, and the incandescent is B4, the T67 should be judged 25% brighter here. 











Here is a fresh comparison of that 100w incandescent bulb, with T91:

If T91 is B6.7, and the incandescent is B4, the T91 should be judged 67% brighter here. 












Unbearable Buzz With T67 On Dimmer
At 12/28/2016, I add a very important forgotten observation. There is a loud hum, from the T67 luminaire,  when connected to any dimmer. The direct observation is with two of the most-useful dimmers, Cooper Wiring Devices D106P, and Lutron C-L Dimmer, CTL-153PDH.This alone renders the T67 unacceptable in its initial form. I'm sure CREE will find a way to fix this. LSGC did an excellent job of ridding Glimpse lights of a hum or buzz some complained against, also in the luminaire. I found that hum barely discernible, and acceptable; but was glad to see it gone. I would not tolerate the hum now with T67 lights. 

Readers, please comment. I think there are very important shifts of concept here. Powerful consensus will speed action to bring suitable lights to contractors, where we might eliminate all future installation of can lights, and pursue a strong weatherization campaign to eliminate existing attic-floor can lights. For now, only the 6" Glimpse is ready for deployment in this campaign.

At 1/19/2013, add note: Find further review in post LED Lighting In RACO 175 Junction Boxes. Also, follow all discussion of LED plate lights by label selection .

At 2/23/2013 add observation of dimming with many LED disk lights in parallel. A 600 watt dimmer will control more lights than you would ever want at one switch, easily forty. The T67s come on instantly. Tests are reported in comparison to T91 lights, at this new post

No More T91's
At 1/27/2013, add another note: Although the T67 is an unacceptable alternative to T91,  Home Depot has ceased entirely to offer the T91 online. The listing at homedepot.com reports this today as Out Of Stock Online. Telephone inquiry reveals the status is in fact inactive, synonymous with discontinued. I'm not surprised. It can not be profitable to continue to assemble this product in Commercial Electric boxes, for very small sales volume. A diligent shopper might yet buy from Home Depot in Lighting Science Group packaging as product Glimpse, 4" or 6", by in-store special ordering .


New in 2014
At 1-24-2014, add link to additional T67 review, along with a new LED disk light from Cooper Industries, now sold at Home Depot. The review is:
http://energyconservationhowto.blogspot.com/2014/01/review-cooper-lighting-halo-led.html 

Please note this review finds a family of new lights no better than T67/ T47, with about-same brightness whether 4" or 6", and with the same old hum/ buzz, whatever you call it. Maddening. Home Depot learned nothing from criticism of T67/ T47, here.

Response to Commenter JoeBreeves, 2/15/2014: 
You suggest buzz might be controlled with Leviton 6615-P0W Electronic Low Voltage dimmer.  

Several reviewers at homedepot.com report success against buzzing, in using Electronic Low Voltage dimmers, so, maybe this works. 

This dimmer may be purchased for competitive $31.97 at Amazon, but I don't wish to buy and try it. I don't like emphasis of the slider in this dimmer, preferring big on/ off toggle. Such nicer ELV  dimmers typically cost more than $70. 

A better luminaire such as Sylvania 70732 is a better choice, getting better efficiency and no buzzing, with ordinary dimmers.

Monday, December 24, 2012

LED Plate Light, LED Disk Light, Can Light, Light Bulb

We need a dissertation on terminology, as two rival light forms emerge in our inevitable shift to all-LED lighting. The rival forms are those modeled on and used as retrofit for, light bulbs; and flat forms best employing the plate nature of LEDs. The basic terms are bulb (antique) and plate (the future). There are forms between these two terms, where LEDs are forced to fit the bulb notion. 

Let's state some terms, and see if they are supported at Wikipedia.

LED Plate Light does not exist.

LED Disk Light does not exist.

Can Light does not exist,

Light Bulb Hyperlinks remain active in this copy & paste of the first screen. One may advance to the further screens in my search, from this blog post.

  • An incandescent light bulb, incandescent lamp or incandescent light globe is an electric light which produces light with a filament wire ...
    69 KB (10,239 words) - 21:39, 22 December 2012
  • Brazil and Venezuela started the phase-out of incandescent light bulbs in 2005, and the European Union , Switzerland and Australia ...
    51 KB (7,531 words) - 03:32, 21 December 2012
  • A lamp is a replaceable component such as an incandescent light bulb , which is designed to produce light from electricity . ...
    2 KB (184 words) - 17:02, 30 October 2012
  • LED lamp (redirect from LED light bulb)
    An LED lamp (or LED light bulb) is a solid-state lamp that uses light-emitting diode s (LEDs) as the source of light. LED lamps offer ...
    34 KB (5,110 words) - 19:18, 22 December 2012
  • This is a list of the longest lasting incandescent light bulbs. Before the development of LED light bulbs , which have extremely long ...
    7 KB (1,071 words) - 10:16, 27 November 2012
  • A compact fluorescent lamp (CFL), also called compact fluorescent light, energy- ...20 million incandescent light bulbs with CFL from a total ...
    68 KB (9,875 words) - 19:11, 22 December 2012
  • light bulb is an incandescent light bulblight bulb may also refer to: Lightbulbs (album) (2008), the third album by Fujiya & Miyagi ...
    1 KB (149 words) - 15:43, 9 December 2012
  • Mazda was a trademark ed name registered by General Electric in 1909 for incandescent light bulb s. The name was used from 1909 through ...
    5 KB (688 words) - 22:16, 27 November 2012
  • The Floating Light Bulb is a 1981 Broadway play by Woody Allen . Semi- autobiographical , it focuses on a lower middle class family ...
    3 KB (421 words) - 02:26, 6 July 2011
  • The A-series light bulb is the "classic" type of light bulb that has been the most commonly used type for general-purpose lighting ...
    969 B (155 words) - 19:47, 18 April 2012
  • The excited mercury atoms produce short-wave ultraviolet light that ... The compact fluorescent light bulb (CFL) is becoming more popular. ...
    89 KB (13,388 words) - 06:30, 18 December 2012
  • A lightbulb joke is a joke that asks how many people of a certain group are needed to change, replace, or screw in a light bulb . ...
    6 KB (881 words) - 07:16, 28 October 2012
  • Like all incandescent light bulb s, a halogen lamp produces a continuous spectrum of light, from near ultraviolet to deep into the infrared ...
    17 KB (2,438 words) - 06:53, 22 December 2012
  • Work on the incandescent light bulb: a Canadian patent application on an electriclight bulb It was granted on August 3, 1874 as Canadian ...
    5 KB (821 words) - 07:01, 11 July 2012
  • Light bulb bayonet mounts: File:Ampoules. jpg | LED lamp s with GU10 bi-pin twist-lock mount File:Bulb-bayonet-male. png | Compact fluorescent ...
    10 KB (1,524 words) - 23:41, 21 November 2012
  • composed of three layers of moldable light-emitting polymer blended with a small ...mimicking conventional light bulbs to unusual forms such ...
    3 KB (370 words) - 20:07, 16 December 2012
  • influenced life around the world, including the phonograph , the motion picture camera , and a long-lasting, practical electric light bulb . ...
    79 KB (11,053 words) - 15:56, 22 December 2012
  • Full-spectrum light is light that covers the electromagnetic spectrum from infrared ...applied to an electrical light bulb but rather a ...
    7 KB (974 words) - 10:41, 7 August 2012
  • image:Gas Core light bulb... Fused silica is almost completely transparent to thislight, so it would be used to contain the uranium ...
    2 KB (284 words) - 23:23, 29 November 2012
  • A metal-halide lamp is an electric light that produces light by an electric arc ...moderately long bulb life (6,000 to 15,000 hours and ...
    27 KB (4,010 words) - 18:35, 18 December 2012
View (previous 20 | next 20) (20 | 50 | 100 | 250 | 500)

I ask agreement that we are missing a distinction between light formed at a point in a lamp-as-bulb-holder/ light fixture, and light directed from a flat array of LED's:
3000°K LSGC Glimpse 4" Disk Light

LEDs are a circular array of eighteen chips.
450 lumens per 9.5 watts is 47 lumens per watt. 
Each of the eighteen chips produces  25 lumens.





2700°K  CREE/ Commercial Electric T67, 6" Disk Light
More illumination must come from each diode  in this array of five chips, and this seems undesirable. Yet this light is not more blinding to look at. The chips are substantially recessed, and that may explain the difference.
If the T91 and T67 are both disk lights, then the difference might be expressed where the T91 is more of a Plate LED, than the T67.
650 lumens per 12.5 watts is 52 lumens per watt. Each of the five chips produces 130 lumens.


I offer  three significant illumination comparisons as demonstration of shape factor differences. The Plate Shape of T91 is the same as that of the Glimpse 4" in the photo above. The shape factor leading to illumination inefficiency is bad in an incandescent bulb. It is far worse in a CFL bulb.

100w incandescent, 1490 lumens vs. T67, 650 lumens. The T67 is brighter by about 20% and more-yellow color. By lumens ratio, the incandescent should be brighter by times 2.3. There is more than a times-two difference I attribute to shape.

100w incandescent, 1490 lumens vs. T91, 750 lumens. The T91 is brighter by about 50%, and same color.

Here, the incandescent bulb should be brighter than LED by the lumens ratio, times two. The LED is underrated times one third, as lumens ratio.












T91. 750 lumens vs. T67, 650 lumens.

I call the T91 about 25% brighter than T67. I think the T91 is brighter by more than the lumens ratio, 1.15, and the difference might be in chip placement vs, the diffuser or lens. 
Light here is projected on the pure white pitted fiberglass facing of GP Densarmor drywall.  I hoped for a very white background, expecting less-subjective brightness comparison. The brown tones were about the same with projection against off-white low-gloss latex-painted walls.










I have thought to edit my blog posts, changing "plate" to "disk" at each appearance. I will not. That would unfairly grant shape equality of the T67 and the T91/ Glimpse family. There are some wonderful things that are done right in Glimpse Plate Lights, that would exist  independent of lens being round, or square. A plate is better than any old disk.

Monday, December 17, 2012

Learning My Lighting Color Temperature Preference

With at last  a complete family of LSGC Glimpse lights, I am able to choose my preferred light color temperature. It's been a small issue for those experimenting with fairly-inexpensive CFL bulbs. Where a Glimpse luminaire costs from $38 to $50. the debate is more serious. 

Where the illumination is pure, from my ceiling, the choice is not complicated by a color-filtering nature of some lamp/ bulb holder. One 6" Glimpse at ceiling center gives very useful office illumination, beaming down onto my desktop, and bouncing off walls. I like brightest illumination, and do not consider a 4" Glimpse for this application. In a 9' by 10' room with 96" ceiling, one good light is quite enough.




I have dutifully tried 6" Glimpse lights of the three color temperatures, in the overhead light of my home office. I hoped my decent Canon digital camera, and professional computer software would allow comparisons to share, where I have photographed the light, then cropped zoom-in on light lens alone. My software fails in this. The three samples are found equal. Room illumination by the three lights is NOT equal. I get used to each, but prefer the illumination of highest color temperature. The 4000°K highest temperature also is most satisfying in rendering my understood color of walls. Surprisingly, walls at 2700°K illumination are yucky. One blog commenter wanting 2700°K lights gave me a bias toward "warm" coloring. I now do not think 4000°K coloring is cold. I fear the choice will be different for individual preference, perhaps flipping as one chooses new wall color.




JPEG photos of the three color temperatures are distinct at least in an aura about the lens.
I am used to lights of 3000°K in all other rooms, and upon leaving and return to my office, find the 4000°K light is harsh. I have swapped back to 3000°K light, ending the experiment. 3000° is comfortable. I don't notice walls as yellowed or glaring. I will not ponder the decision again.

For others too, I think 3000°K is a choice you will not regret. I think most of us will want consistent lighting, the same color temperature in all rooms. Just choose the size needed. 4" for spots. 6" for work areas. 6" with a dimmer in hallways, where you sometimes need brightness, and sometimes want a night light. Maybe 6" with dimmer over a kitchen sink needed as a night light.

I wonder who is responsible to aid this consumer choice. Does every product have to be offered with color choices? Can a generic store display inform the choice, such as these at my Home Depot:
































I take these displays to be of 2700°K, called Soft White, 3000°K, called Bright White, and 4000°K, called Day Light. Can't we choose consistent names? Or, might we avoid debate where with a plate light, a spectrum of diode phosphate coating color might be applied? In a 4" Glimpse, there are 18 diodes. Why not apply a sampling of the three colors. or a fuller spectrum, arranged appropriately?


Why must a fixture have only one diode color? This is Warm White 3000°K, and it is interesting that diode phosphor coating color is related intuitively to illumination color.

Thursday, December 13, 2012

My Comparative-Lighting Lab, Second Result

I received my first 4" Glimpse lights yesterday, in an order from 1000Bulbs.com. Here are two photos where a 3000°K 4" Glimpse is at the right. A dimmer, on the Glimpse light only, is set at full brightness. The LSGC package states this is 450 lumens. Much to my dismay, the package also states incandescent equivalence of only "up to 45 watts ." 

In the first photo, the light at left is an old 100 watt incandescent. I see some color difference even against a white background, and about-equal brightness. This is the result I expected, 4" Glimpse brightness B4, that of 100 watt incandescent, and 6" Glimpse yet-brighter, by 50%.













In the second photo, the light at left is a brand-new Utilitech #0338931 Warm White 3000°K LED bulb, rated 800 lumens, drawing 13.5 watts and with intended 60-watt  incandescent equivalence. This bulb is a licensed representation of the Philips bulb that was awarded an L-Prize. This bulb has expected adverse shape factor, formed to illuminate about a bulb centerpoint through a range of 300°. The Utilitech bulb is less bright than the 450 lumens 4" Glimpse, by perhaps 50%, and with equal color. Accept Brightness Number B2.4, the proper 60% of the illumination by a 100 watt incandescent, B4.

I am puzzled by the in-photo rainbow in both photos, about a light of bulb shape.













I am up early and engaged in this, to proceed in a challenge of brightness ratings by proven-wrong methods. I want illumination superiority of plate LEDs to be weighed in their favor, to speed demise of all bulb lighting, inferior for both quality of illumination, and, by large margin, in energy efficiency. I would kill all manufacturing of can lights. I will never buy another light bulb of any kind. I love the LED plates on all of my ceilings, and only want evolution to bring DC strings and artistic starry-sky effects.

Read more . .

Friday, December 7, 2012

My Comparative-Lighting Lab, First Result

I am inspired to know that measured lumen output of a light is not correct statement of its ability to provide useful illumination, by this photo:
The photo is posted in the Facebook pages of LED Source, in Wellington, Florida.

Caption:
Changing lamps at our new corporate headquarters. Look at the comparison between a 13W CFL and a 15W LED Glimpse. Can you see the difference?





A 13W CFL will be rated at about 930 lumens. The 15W Glimpse is rated at 750 lumens.


This is my apparatus attempting similar object illumination. Thinking outdoors despite wet Fall weather, I beam this at my freshly white-painted garage door. 



































On the left, a 100 watt incandescent bulb, 1750 lumens. On the right, a T91, 6" Glimpse 14.5 watt plate LED, 750 lumens. Compare brightness as that in each shadow of the divider. I call the 6" Glimpse  brighter by times 1.5, where the manufacturer, LSGC, has been reluctant to claim superiority to more than a 65 watt incandescent. And, I must take care now to not call the 6" Glimpse, the Home Depot abandoned T91. For a time, I must take care to apply both labels.

I think under-rating of brightness is not a defensible safe approach. Those first trying a 6" Glimpse will often find them too bright, wanting them dimmed. Many will prefer a 4" Glimpse, and have not yet been offered the choice. Knowing better, they will demand the choice.

Who ever said a 100 watt incandescent was a standard of brightness? It was only at, or beyond what a fixture could tolerate without damaging wiring insulation, or starting  a fire. A 6" Glimpse is never hot to the touch.

With new standards stated for LED, we can have whatever illumination we need, from fewer fixtures. Those new standards need not carry description of the now-inferior illumination from still-hot compact fluorescent bulbs. How about Brightness Numbers? Say, B6 for 6" Glimpse, B4 for 4" Glimpse. A 100 watt incandescent might be B4, 25 watt B1. Let's get that naming conversation underway.

Perhaps it is the outmoded notion of bulb construction that gives disadvantage. Surely a plate LED will greatly surpass an LED assembly in a bulb, recessed into a can. Some kind of shape factor will affect the relationship of LED power consumption, and its effective illumination. It is only the illumination, the projected brightness in zone of need, that matters. Incandescent and CFL lights can not benefit from shape factors, and are doomed in that way too. Good riddance to foolish contraptions holding bulbs, and surpassing bulbs in cost. 

An LED plate light is a round-the-clock available spot of sunshine on your ceiling or wall. Better light quality, and at night too, than a sun tunnel blocking passage in your attic, maybe leaking heat, and compromising your roof, at large installation cost. Sorry, good folks at Solatube. A little PV array can serve up the daytime electricity for lots of LED plate lights. Let them be, now, strung and more-readily dimmed, as low-voltage DC circuits; said despite concern to not confuse, where DC wiring has not yet been offered except in commercial, RV and boat applications.