Wednesday, March 14, 2018

Review, GM Lighting S4 4" LED Disk Downlight, Driver On Board, Sold At Amazon

Don't go here!

GM Lighting - 4" Round LED Surface Mount 3000K 90 CRI Fixture White 

This looked to me, like Glimpse and Nicor DLS10, a reversion to junction box mating with screws revealed under a pretty access ring. I should have known better, where it is described as "driverless", then some cheap stamped-pan thing. 

As a dutiful reporter, I did show respect for yet another Amazon offering of a seemingly pretty LED disk downlight. The Amazon asked price is $24.95. I got a tempting better deal at Bees Lighting , $19.80 and free shipping. I should now be grateful and uncritical.

This is not offered with hardware to clip into a can light.

See the many slots that will be unprotected against intrusion of light-seeking-always bugs, to die in a pile of litter upon a loosely-attached obscuring lens. An informed installer might tape over the screw keyways. The large openings of lens keyways CAN NOT BE BLOCKED. Seal the junction box air tight, as I do, or don't install such cheap lights.

Keyways for box screws are at small 2 3/4" pitch.  Boxes with the small screw pitch are the most likely to be badly installed, with large visibility of spilled light.

Take this side by side comparison as confirmation of GM S4 claimed brightness and color temperature.

If glare numbers were required in advertising  and on packaging, the mindless degradation since Sylvania 70732 v1,  might have been avoided.

Monday, February 19, 2018

Review, Commercial Electric 74203 and 74207 Color Changing Edge Lights Almost Surface-Mount

There is big change in LED downlight offerings at Home Depot. A new two-member family of very promising ultra-thin LED edge lights offers easy mounting anywhere. First review the big brother of the family:
Commercial Electric 74207 (the identifier in the bar code) 

5/6 in./J-Box 12-Watt Dimmable White Integrated LED Round Flat Panel Ceiling Flushmount Light with Color Changing CCT  

Let us not encourage a naming as "Color Changing." I dispute any virtue in user control of the illuminating color temperature, achieved by phosphor coating of diodes.

Find an identifying number in the bar code at bottom of box. Find a box that is compact and fully recyclable. Commercial Electric (17801), product ID 74207. Unboxing, find preset of mounting as retrofit of an A-type bulb in a can light. Simply detach the plug at the quick-connect  and free the harp springs by compressing to remove nylon ties. You know the rest of the simple mounting that will probably air seal a decrepit can.

I am interested in the mounting upon a junction box, and pry off the mounting bracket to detach the harp springs. Look closer at the color temperature selector switch that you will not want to leave in the SWITCH position. In the default SWITCH position, the color is initially perhaps 2800°K, not the box-claimed 3000°K.

On my comparison stand with Nicor DLS10, 3000°K, 750 lumens at left, and CE 74027 at right. See about-equal brightness confirming the box claim of 800 lumens output.

Cycle the wall switch, to BW, Bright White, claimed to be 4000°K.

Cycle the wall switch again to bring up setting DL, Daylight, claimed to be 5000°K. 

Daylight? We all know sunlight does not shift perceived color of a beige wall, to blue.

In fact blue illumination is in absence of daylight. This photo is found among "color temperature" discussion at Wikipedia. A cream-colored house is blue and purple at sunrise.

Here is a spectrum of sunlight correlated color temperature from Wikipedia . The Wikipedia discussion states: Daylight has a spectrum similar to that of a black body with a correlated color temperature of 6500 K (D65 viewing standard) or 5500 K (daylight-balanced photographic film standard).

Blue colors are thousands of degrees Kelvin, above that of luminaires called 5000K. Something is wrong in our marketing of LED color achievement with painted-on phosphors.

See the test stand setup with 74027 at SW, Soft White setting.

74027 box mounting is simple and secure.

Get curious now, how "color changing" works. The white rim of the luminaire is resilient plastic, easily deformed to successively release keys that retain the metal pan.

One of the keys has easily-separated gluing to the pan and between the rim and the obscuring lens. Leave the rim/ lens bond alone.

Drop out the lens and backing paper to reveal three conductors between the diode strip and the converter block.

At full power in SW setting, find light of diodes with an orange phosphor coating.

In the BW setting, all diodes are powered, and light that will be projected through the obscuring lens, seems a natural color.

In the DL setting, only the diodes with a white phosphor coating, are powered.

I hoped to find the luminaire desirable, in the BW setting. I do not. There is unacceptable bluing of illuminated objects. I imagine virtue in display of multiple color temperatures to better represent natural sunlight. Perhaps many soft colors will be the emerging best design. The two colors equally blended here are of too extreme a range.

See remarkably small lens luminance. There are many wonderful things about this 

Here is Commercial Electric 74207 among Nicor DLS10 lights upon a kitchen ceiling. The 74207 color temperature setting is BW, Bright White. It is too blue, to my eyes and to sensitivity of my iPhone 6S camera. The DLS10 are pretty jewels especially with some dimming. The 74207 is harsh, and I must reject them, making returns at Home Depot.

I persist in supportive curiosity about the new Home Depot offerings of surface-mount ultra-thin LED edge lights.. I want to buy and try the 500 lumens little brother of 74207 . Not in my store. Nor sold online.

There is this, too, not really on offer. 
Commercial Electric 4/5/6 in./J-Box 12-Watt Dimmable White Integrated LED Energy Star Recessed Trim Disk Light with Color Changing CCT 

I have dutifully purchase the little brother of the new Home Depot edge light family, called 74203 in the bar code. Despite awful reception of all "color changing" luminaires, at March 2018 it is  prominently on store shelves among 4" can light luminaires.

Out of the box, 74203 is ready only for setting in a cylindrical 4" can light. I won't bother to remove the spring clips, sticking this onto my illumination comparison stand.

Original 4" Glimpse 3000°K 500 lumens at left. 74203 at right, at full power, first setting SW, delivered in SWITCH position. 4" Glimpse, 3000°K, correctly illuminates wall color.

Original 4" Glimpse at left. 74203 at right, at full power, second setting BW, delivered in SWITCH position. 4" Glimpse, 3000°K, correctly illuminates wall color, but my eyes and my Canon Digital Rebel now find yuck in Glimpse color.

Original 4" Glimpse at left, 74203 at right, at full power, third setting DL, delivered in SWITCH position. 4" Glimpse, 3000°K, correctly illuminates wall color, but my eyes and my Canon Digital Rebel find yuck in Glimpse color. The 74203 color illumination is GLOOMY.

Look from further back at comparison in the DL setting. Everything is wrong.

The 74203 does not fully dim. Here at about 20%, see very gloomy night lighting.

Now give another three-star review at, disliking push of "color changing" as virtue.

Done, and with regret. This light will be superior to many other luminaires that might replace dead bulbs in 4" can lights, especially as architectural lighting or security lighting upon home soffits. For this and with many installation virtues, this deserves at least four stars, and I will find means to fix my review. I do wish to make a strong vote against the blue diodes and luminaire switch setting confusion.

Friday, January 12, 2018

The Evolving Relationship Of Lighting Brightness, To Power Consumed

With informed good choices, lighting is already a resource available to all, at little cost. Light, clean air, good potable water and internet broadband are no-cost essentials of a just society in which the potential of every person may be achieved. Of these essentials, "free" light is the nearest to hand, perhaps not corruptible through stupidity and greed. With lighting we are on the verge of considering power draw only as it relates to potential of generation off-grid. Reduced in cost by more than 90% vs. that from dumb incandescent bulbs, we may choose to use light more abundantly, for safety, productivity and better quality of life. The potential is deserving of a global campaign to do away with light bulbs, just as we did with awful CRT computer and television screens. Again, the solution is in durable, flat things. And, let those flat things for awhile include BR30 LED flood bulbs directed from cans, functionally the same as any LED disk downlight.

This post will evolve over several weeks. Please watch it grow. And, know conclusions have already been expressed here over the four or five years largely mis-spent since excellent LED disk downlights became available for home installation. LED lighting will improve global quality of life and productivity, while hugely contributing to saving of grid electrical power used for residential lighting. Only about 10% of the potential has been achieved so far in the USA. Ref: Adoption Report, USDOE, 2017 .

Here are discussions in this blog:

Focus please on the word Brightness of the post title. Go with me to the linked Wikipedia article, and find disastrous confusion sowed by the US Federal Trade Commission in the matter of Lamps, the topic here:
The United States Federal Trade Commission (FTC) has assigned an unconventional meaning to brightness when applied to lamps. When appearing on light bulb packages, brightness means luminous flux, while in other contexts it means luminance. Luminous flux is the total amount of light coming from a source, such as a lighting device. Luminance, the original meaning of brightness, is the amount of light per solid angle coming from an area, such as the sky. The table below shows the standard ways of indicating the amount of light.

Today, has
lamp (noun)

1. any of various devices furnishing artificial light, as by electricity or gas.

There can not be a justification by FTC, to imagine and apply only light bulb point sources as lamps. And, of course an LED light bulb is not a point source. An LED bulb hemispherical lens glows to the side with task lighting efficiency of less than 30% of total lumens. About 70% of typical A-type LED bulb illumination is sent from a circuit board parallel with the base; a nasty and dishonest trick when upright in a table lamp or horizontal in a ceiling fixture. A large fraction of emitted light is lost in the thick diffuser. A wasteful device must not be promoted as other than decoration, without offer of purchase incentives.

A cheap and short-lived A-type LED bulb has sideways illumination like that of the old 100 watt incandescent, and is 40% brighter end-on. If used in a table lamp for reading, 40% of power is wasted. 

Here is proof from test stand comparisons .  At left "100 watt" incandescent bulb. At right EcoSmart ICES-003/B LED, 800 lumens, 10 watts, 3000°K. At full power. The EcoSmart is turned 90° to illuminate the screen with cast from glow of the lens hemisphere as it might in a table lamp or horizontal ceiling fixture. The EcoSmart side illumination is about 500 lumens, equal to that of a "100 watt bulb."

At left "100 watt" incandescent bulb. At right EcoSmart ICES-003/B LED, 10 watts, 3000°K. At full power. The EcoSmart is now straight-ahead to the screen and quite superior to a 100 watt bulb, the effect of 800 lumens, not the useful 500 lumens of a 100 watt bulb. Judge that the EcoSmart if placed upright in a table lamp or horizontal in a ceiling fixture is with usefulness ratio 500/800, about 40% energy waste.

We must go back to the correct usage of brightness as luminance, to inform our decisions, as we develop and select luminaires that ever-better, make light a cost-free resource to enrich human activity. Luminance is first a measure of the illumination to our eyes, of tasks at hand. Nearly as important, it is the test of glare as we gaze upon the luminaire. Glare as a nuisance or hazard, is not regulated and not listed in competition for purchases. Lights continually cheapened and glaring, discredit the switch to LEDs for energy conservation. Where best LED lights will serve essentially forever, we can afford that they be built without weak electronics, easy to look at.

Following the FTC mis-direction, we are taught that lumens count, luminous flux in all directions,  is the modern way to compare lighting choices.

Here is an example:
C/Net, Smart Home
Watts vs. lumens: How to choose the right LED light bulb 

Article by Alina Bradford, May 9, 2017

These comparisons are meaningful only for light in antique candle form, as point source illuminating a beam angle of about 320°. Achievable brightness of such old-fashioned bulbs actually can't exceed that of the well-known "100 watt bulb." Wiring that fed a 150 watt bulb was usually destroyed by heat. The table perhaps consciously under-values LED lighting, where LED tricked to look like point-source incandescent and CFL bulbs, has not been offered at more than about 15 watts. All LED luminaires of higher wattage are of other form, as directional flat plates, the natural display from diodes on circuit boards; and do not belong in this table. Disregard as ridiculous, "corn" LED bulbs at the high end of table wattage.

About half of point-source light is fully wasted where the purpose is task illumination as in reading. The comparison prevails where something is sought to fill  A-Series light bulbs with their global confusion of sockets, in USA and other countries with 110 volt mains, E26 threaded

We have long had lighting that is not dissipated as point-source, to confuse a relationship of watts to potential task illumination.

Again,  before LED lighting became practical, interim CFL point-source and flood or spot bulbs, muddied watts as a measure of brightness. 

The wrong teaching is in failing to distinguish lumens count from a directional light such as an incandescent flood, from wastefully-thrown light of point source bulbs. 

Hereafter let Luminance be a controlling factor in luminaire planning and design. 

Look at luminance numbers for some existing edge light LED examples:
Cooper SLD6: 0.064 lumens/ mm2
Conturrent 4": 0.074
Lithonia WF3: 0.228
Lithonia WF4: 0.124
EiKO 5": 0.059

Try imposing a uniform lens luminance limit of 0.05 lumens/ mm2

Tabulate lumens and lens diameter at this limit:

Let there be greater lens luminance, to get about 500 lumens from a 4", 100 mm lens diameter.

These are candidate lumens targets for a range of lens and called size-of-light, in inch,  25 mm increments.Let suggestion of luminaire description by a "brightness number", instead be a series of round-number lens sizes. And, let those size numbers not be those of the lens rim outside diameter.

Please look at last, for discipline in luminaire offerings, those volunteered by China factories without global leadership, and by specifiers everywhere, for competition upon reasonable glare numbers. For greatest number of directional lumens with typically 120° beam angle, we will choose edge light luminaires with proven effective 90° downward turn of light. 

Where edge light lumens might be proportional to strip diameter and circumference and luminance is then related to square of diameter, LED edge light strips will be intelligently related to square of diameter, rather than to diameter.

Efficacy based on task illumination will easily be with more than 90% reduction from energy cost with LEDs, vs. cost from A-type point source incandescent bulbs. Improvement will not be the 75% reduction imagined with A-type LED bulbs that look like A-type incandescents and fit in any old luminaire. Where edge lights become the production standard, I hope the diode strips will have a salting of phosphor colors, to better resemble sunlight effects in human vision.

This just in 2-7-2018:

Offered at Gordon Electric . But, I can't find these except through email links, among the vast forest of distributor offerings.

Do can light retrofits compete with an edge light surface mount simple future?
All of these Sylvania offerings are with circles of uniform brightness, here presumed 4" diameter.
All of these are too-aggressive cramming of light, unmindful of hurtful glare. Can lights must not be used for more than "65 watt incandescent equivalent." I bet that 16-watt offering runs quite hot, and I won't buy to try. For residential installation, there is no virtue in such large light concentration. In recessing deep within a can to not see the glare, expect that about 20% of the light energy is thrown away.

In support of this post, and for decision upon lighting selection in my own kitchen remodel, I have purchased several candidate LED edge lights.I want to find a pretty 500 lumens LED edge light with 4". 100 mm active lens area. Here is that remodel scheme:

Now in March, 2018, I have planned to use fewer lights of larger size. I thought to use available Nicor DLS10, but they are configured only for individual constant current drivers. Choose edge lights that, like all tape LED lighting, are configured for a constant voltage driver. To employ a common driver, luminaires must connect in parallel. I believe that light engine circuit boards are generally wired to want a fixed current.