Comparing LED flashlights using lumens

Like any mass market product, there is a tendency to dumb down the specifications to make product selection less painful for the GUM (great unwashed masses). This is unlikely to change as long as everyone considers shopping, consumerism, and product differentiation to be non-technical exercises.

However, that's no reason you should accept a dumbed down method of specifying lamp brightness. If you want to know what 1000 lumens looks like, you can measure the total light output of an LED or incandescent light bulb. The official way is with an integrating sphere, which will probably cause you some difficulties when you drag one into the store. More useful is a piece of cardboard with a circular hole, tape measure, lux meter, and calculator. With a few measurements, you can calculate the light output in lumens of the display lighting. My math sucks, but I think I can work out and post the method if you suspect it might help.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann
Loading thread data ...

To expect anything in life to be "technical" is folly. Marketing

*has* to be "dumbed down" to the common man. I see no reason that "lumens" won't enter the lexicon, replacing "watts", given time. The GUM never understood "watts", either, but could make marketing use of the term. I can deal with lumens (though have to calibrate myself against the "standard 60W bulb every time I buy bulbs) but have a problem with temperature. A 5000K lamp is a *lot* brighter than 2400K with the same luminous flux.

"That's no reason you should accept a dumbed down", then followed by "you can measure the total light output..." and "the official way...integrating sphere" and "calculate the light output...".

Yeah, right. That's how everyone should buy light bulbs. Good grief!

Reply to
krw

I once had a vague illusion that product evaluation would be somewhat technical in this newsgroup. I would expect people to evaluate their needs, read some of the available literature, skim the product reviews, compare available products, calculate cost effectiveness, estimate cost of ownership, determine the value of the warranty, and factor in any other factors that can be reduced to dollars. This should be nothing new as this is roughly what we all do when determining which components to use in out designs. You can buy your LED bulbs by looking at one number (equivalent incandescent watts), you can extract something more useful in reviews that actually test the bulbs, or you can do your own measurements. I prefer rolling my own when possible.

Product marketing doesn't need to be "dumbed down" if one is selling to an educated market. However, when sales drop due to the lack of a sufficient quantity of educated buyers, the inevitable result is a general trend toward selling to the lazy and clueless. An odd example can be found reading Home Power Magazine. In the distant past, the magazine was full of do it thyself construction projects, how it works articles, and even schematics. In other word, teaching the readers how solar power works. Advertisers were mostly showing system components. In the last 10 or so years, the solar market has matured, and totally non-technical people are buying solar power system. The magazine is now full of articles on ready to use solar systems, professional installers, and solar systems sales. What's left of the technical parts that deal with solar are now articles on maintenance, system financing, and calculating how big a system is required using some really odd attempts at simplification. However, small hydro and wind power has not really taken off as much as solar. So, the articles on these areas remain similar to the original construction projects, etc.

Incidentally, some of the more spectacular failures were during the early days of home computers, where teaching totally non-technical buyers how to program was considered the best way to market a technical product. That didn't last very long, and neither did the companies promoting the idea.

Agreed. However, I also see no reason why it couldn't have been done with the introductions of LED lighting.

I think you mean "color temperature". Yeah, that's a problem. Unfortunately, it has become a political issue when selecting the color temperature of LED street lighting. Various advocacy groups claim the LED lighting in the blue range is harmful to health and the environment. The list of associated maladies is truly impressive. The push is to get municipal utilities to switch to warmer 2700K LED's. When discussing this will the proponents, I soon found that they didn't know anything about lighting or color temperature.

Yes, that's a bit extreme. However, Jasen Betts did mention that he didn't know what 1000 lumens looks like, so I thought it might be useful to provide a really simple testing method. If it works, I could probably create a smartphone app that would do the measuring and calculations. Then, you could easily run around testing everyone's LED light bulbs, for a nominal charge of course.

Drivel: It's rather difficult to see differences in white light intensity. However, differences in color are easier. I had that problem evaluating differences in bicycle headlights. So, I used ImageJ and ImageJ2 to produce a pseudo color intensity map. Note that the ground reflectivity needs to be uniform for this to work. If anyone wants to know how this was done, ask and I'll post something.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

If you do something repetitively and often enough, it becomes instinctive. That probably applies to incandescent lighting, but not to LED lighting, which is a more recent innovations. All Edison screw base style bulbs have roughly the same hemispherical light pattern (except for the 30 degree wide cone shaped shadow produced by the screw base). You can interchange 60 watt incandescent bulbs from any other manufacturer and get fairly close to the same light pattern and intensity.

Not so with LED lighting. The bulbs have different light patterns, different color temperatures, and different beam widths. Grab a random 60 "incandescent equivalent watts" LED bulb off the shelf and it's anyone's guess what it's going to look like. Notice the bulb in the test fixture. It has about a 90 degree conical area around the base with no light. That allows the bulb to concentrate more light in other areas resulting in a brighter and more "efficient" light with no increase in power consumption.

There is also an LED flicker problem:

Sure. We also tend to do arithmetic in the language in which we learned it. Getting computah users to switch from their initial word processor is very difficult. When I'm not paying attention, I tend to use Wordstar style keyboard sequences. Brand loyalty is some markets is amazing. People will defend their choice of overpriced and buggy smartphones even when there are better and more economical choices available. We're a conservative culture and hate change in any form. For supermarket shopping, I use imperial. For science and engineering, I use metric. I grew up using imperial, and therefore tend to think in terms of imperial units, and then convert to metric as required.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

They always did. In our house we had many small halogen spot lights, for example above the kitchen counter. They illuminated only the counter an hardly anything else. Still, I compare the CFL in there now to "equivalent incandescent watts". I know that is iffy but its had tro change such habits. Plus it usually works.

Oh yeah, and also every little glitch on the line is immediately responded to by the lamp over our living room table now. Not nice but we tolerate it considering that we dropped it from 150W to 25W or so.

Old saying: "If it works, don't try to fix it". All my biz databases are still generated in MS-Works format, something I started when first going self-employed in 1989. The software has been discontinued eight years ago but not in this here office. I found that all other databases I tried were either overly complicated or not very good. So wy change?

With software it is often like with wine, older is better.

Worse here. For aerospace clients it is always metric, has to be. For a few (very few) others it's metric. Occasionally I have to switch more than once a day and sometimes that also includes switching the language.

I grew up using metric but I am now mostly "transformed" to imperial. Except when working on bicycles. Ok, my mountain bike is American and has some imperial stuff.

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

One of our best customers, an aircraft engine manufacturer, does all their stuff in english units. That's astonishing.

We do mechanical design and PCB layout in inches, but all engineering math is SI.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Huh?

Do you really spend three or four hours to buy a frappin' light bulb?

Since when are light bulb manufacturers selling to a technically educated market?

Nontechnical people buy light bulbs, too.

The original buyers were technical, or at least educated do-it-yourselfers. How is this any different than what you describe as the early adopters of solar power?

No way! People want what they're used to. They *know* what a 60W light bulb is. That's what they want to replace.

Same-same.

Not surprising. Cell phones. Power lines. Alar. Global Warming.

But he, nor I, will remember it next time we go to buy a stinkin' light bulb. As, I think it was JimT said earlier, I know I want a 60W lightbulb in my reading lamp and 100W in the garage. Until I can remember the "lumens" I bought last time, I'm going to want the "watts-equivalent" number printed on the box. Basically, I don't want to remember the conversion. Let the manufacturer print it for me. Eventually I'll remember.

Reply to
krw

Once when I was in line for an oil rig helicopter flight near Aberdeen (huge heliport) there was a guy with a note pad in hand. When it was my turn he asked "How many stones do you weigh?" ... "HUH?" ... I was holding up the line line .. ahm .. queue until we could find someone to translate that into kilograms.

There ain't much left in imperial in EE. The big difference is mech. That's mostly imperial in my case. Now I may have one of those rare occurences, a large project where the data presented by the customer is all metric. So on our next call I'll have to ask them which they want for the design since it's a US company. My layouter probably isn't going to be enthused if metric.

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

That means that if you discuss a product, it should be in the form of measurements, test results, data sheet references, and numerical specifics, rather than vague feelings and impressions. For example, if the light isn't bright enough for your application, your evaluation should be liberally laced with numbers such as the part number, lumens, measured lux, distance, illumination angle, LED specs URL, etc.

Nope, but I seem to spending that long replying to a Usenet thread on light bulbs. I certainly don't spend 3-4 hrs in the store sifting through the available light bulbs. If I'm after something specific, I usually spend 15 to 30 minutes online looking at data sheets, reviews, and test results. If I find what I like, I either buy online, or if I'm in a hurry, find a local store that stocks the particular LED bulb. If I had a cheap and easy way to measure lumens in the store, such as a portable integrating tube: I might try to use it in the store, especially if I suspect that the numbers on the package are inflated. Also, I don't try to match LED lights to any existing incandescent lights. They are sufficiently different that such an effort would be futile. Instead, I remove all the incandescent bulbs and replace them with LED.

For friends and customers, I don't have any problems dealing with either lumens or equivalent incandescent watts. The problem is the color temperature. As I previously mumbled, the human eye is very good at differentiating colors, but not so good at brightness. Subtle differences in brightness are easily ignored, while variations in color temperature can easily be seen. I bought a case of no-name Chinese LED light bulbs that apparently had not been "binned" or sorted by color temp. Most were "warm" wide variations. The customer was not happy, so they were unloaded to a thrift shop.

The target market for LED lighting is certainly not the technically educated segment of the light bulb buying public. I had some vague illusion that most readers of an electronic design newsgroup would tend to fall into the group of technically educated light bulb buyers. Considering the gratuitous insults in off topic political discussions in S.E.D., perhaps that would be an overly optimistic assumption.

Of course. Marketing always targets the largest and most profitable market segment, which usually is the lowest part of the bell curve. Nontechnical people buy whatever the store happens to have in stock. They have little or no idea what the numbers mean, and probably don't have a clue how to verify that they're not getting cheated by inflated or obscured claims. I'm attempting to remedy the situation.

It's not different. Targeting intelligent buyers always works, until you run out of intelligent buyers. At that point, both the product design and method of selling needs to change. In solar power, computahs and smartphones, we haven't quite scraped bottom yet. In LED light bulbs, we may have hit bottom, especially when I see retail LED light bulbs sold more by the packaging than the contents.

I beg to differ. The purpose of marketing is to confuse the buyer. The purpose of sales is to offer a solution to the confusion. If the light bulb buying public had been forced to learn a new technical term (lumens), at least some confusion would have been inevitable. This would have provided a golden opportunity for the practitioners of merchandising and salesmanship. A few years after the initial shock, buyers would have been accustomed to the situation and accepted the change.

Yep, they all have similarities and I was surprised by the discussions. I've recently become accustomed to arguing with fairly unemotional individuals with technical backgrounds. In other words, I was spoiled and found it difficult to believe that people were debating technical issues using emotional terminology. The unanswered question for me was should I switch to emotional and almost irrational methods of discussion, should I try to educate the opposition in lighting terms and measurements, or should I try something else.

I have the same problem. So, I save identifying pieces of the packaging from a product that I like, toss these into a plastic Ziploc bag, and keep it in my car. When I go shopping, and can't recall some detail, the package acts as a reminder. If I need to ask the salesperson where they were hiding my favorite product, the package serves as a suitable reminder.

That's the same argument against switching to the metric system. People would not remember metric, therefore it's bad. While it's difficult to imagine that a survivor of the American educational system can learn one new technical term, it might be possible to convince congress to allocate billions of dollars for the Dept of Energy to educate the GUM (great unwashed masses) in a new and improved way to measure and buy light bulbs. If the billions of dollars appears to be an impediment, perhaps a free smartphone app instead.

"Lumens and the Lighting Facts Label"

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

The "color temperature" would be a useful measurement, _if_ the light source would be a true black body radiator (CRI=100). Unfortunately the "white" LEDs are far from black body radiators.

The LED semiconductor chip generates a single blue spectral line and the phosphor then convert some of that blue spectral line to a more or less continuous spectrum of red, yellow and green light. To generate blue light, the blue spectral line is allowed to penetrate the phosphor. By adjusting how much blue spectral line gets through and how much is converted to longer wavelengths, you can generate different color temperatures.

Regarding which is the "correct" color temperature for street lights, take a look at the Kruithof curve which tells what color temperature people find pleasant, depends on the illumination level,

formatting link

For low illumination levels (as in street lights) the range is around

2000 to 3000 K, for medium levels (1000 lux) any color temperature above 4000 K is considered pleasant.

For street illumination, it would be important to have a more or less continuous spectrum from red to blue. The worst example is the LPS lamp with only two spectral lines around 589 nm (yellow). It is impossible to determine the color of cars and the color of the cloths people are wearing. Don't expect that eyewitness reports would be accurate in such illumination.

Reply to
upsidedown

That's great for technical subjects but we're talking about a frappin' light bulb. It's not, and shouldn't be, rocket surgery.

But that's what this thread is about.

The question is what LED? "Equivalent watts" is the handy, if not perfect, gauge to use to replace the incandescent. That's why it's not going away soon.

Color variations can be seen but not remembered well, either.

Not the subject of the thread but, no, people here aren't going to spend hours researching a light bulb.

Good luck with that.

Which is why we have "equivalent watts". People have a feeling for "watts". Good enough, they don't need/want more. I think they get the idea of "warm white" and "cool white", maybe even "daylight", but color temperature is too much information.

Utter nonsense.

Complete bullocks. The seller is going to force the buyer to become technical? You're nuts!

Yet you think sellers can force buyers to become technically savvy?

Do you cart around a file of all of these hints? Really?

Precisely! That's worked out well, hasn't it?

It *is* bad. There is no need to switch. It's wasted work. If I need to convert, I *have* the smartphone app. It's called a "calculator".

Reply to
krw

How about one of these? etc. Kinda pricy. Maybe next Christmas.

Yep. Good explanation.

According to LED lighting is evil advocates, anything above 4000K is dangerous and uncomfortable. Personally, I find bluish light to be rather irritating. When bright blue LEDs first became available, some equipment manufacturers incorporated them in their equipment. I have a few of these (Asus RT-N66u router, HP m9077c computah, etc) that have extremely bright blue LEDs on their front panels which I find very irritating. My reactions do not seem to follow the Kruitof Curve.

So that's how the Klingon cloaking device works.

Some interesting reading:

"Streetlights for local roads" Notice the rather narrow horizontal illumination angles for LED luminairs.

"The brightness you see versus the brightness you read about: on light levels, LEDs, and the shortcomings of the lumen"

"AMA Adopts Guidance to Reduce Harm from High Intensity Street Lights" "Industry Responds to AMA LED Streetlight Warning"

"Will tunable street light breakthrough silence LED critics?"

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

Yes, it does work, and works well for me. I carry mine every day. It's great for looking at small parts, shining up from underneath a PCB so that you can see the traces from the top side, looking underneath the hot water heater - just about anything that has a relatively small dark space you need to see. For me, the bright setting is too bright, the dim setting is too dim, and the medium setting is exactly right. It will not light up the world, like some of the huge lumen fans seem to want. :-)

Ed

Reply to
ehsjr

These at least claim to measure Correlated Color Temperature (CCT), but how well does this match with the visual experience, when a spikey spectrum is used ?

That is strange, since there is good 6000 K black body radiator available, the Sun. Is daylight dangerous ?

The problem with traditional street lights was that they spewed light all over the places, up to at least the horizontal plane. Thus, when walking on a street, the bright light sources from the next few poles, will shine directly into your field of view.

A bright light source in your field of view will destroy your night vision for several minutes and the ground level illumination looks too dark. People demand more street lights and if such bad luminaries are added, there is just a vicious circle.

The correct way of designing street lights is to limit the radiation strictly downwards (OK illuminating the ground halfway to the next pole). In this way, the bright light source will not inadvertently hit the field of view, so dark adaption is maintained. The ground looks bright enough all the time, sometimes making it possible to _reduce_ the total illumination. With LED luminaries, it is easy to design he illumination pattern as required.

Reply to
upsidedown

It's probably not great for your eyes. Short-wave radiations like ordinary solar UV-B (400nm) wear out your retina over time, degrading night-vision with age.

Exposure to blue light has the same effect, albeit with greatly reduced potency.

Both promote cataracts, too.

Cheers, James Arthur

Reply to
dagmargoodboat

You mean other than causing cancer?

--
This email has not been checked by half-arsed antivirus software
Reply to
Jasen Betts

My sony vaio has a wireless keyboard and mouse and both have multiple LEDs and both use a single 1.5 alkaline cell.

And amazingly a single cell lasts for months, even if I constantly forget to turn it off. Sony came up with an excellent sleep mode on these things.

Reply to
DecadentLinuxUserNumeroUno

Was it on sale? The usual price for a Cree 100 watt equivalent is typically about $13. The cheapest Home Despot Cree 100w equivalent LED I can find online is $11.

That's odd. The package and data sheet says 16.5 watts for the soft-white version and 15 watts for the daylight version: CRI on both is 85. Some other CREE models are 90+ but I couldn't find any at 95. No specs on power factor. Is this the same as what you bought?

I have three kill-a-watt meters. One was reading somewhat high. To test, I wired them in series, and plugged the string into a known load. One was definitely reading too high. I took it apart and found the soldering around shunts R17 and R20 looked suspicious. I resolder everything and the readings returned to normal. You might check your kill-a-watt meter: The lower photo is the P4460 with the soldering problem. The upper is a P4400.

If it were that easy for everyone, LED lighting manufacturers would offer only one model and buyers would all rate that model as magnificent. Instead, Cree and others are constantly introducing new models and the buyer feedback sections of web site are often highly critical of the product. In my case, I bought an assortment of color temperature A21 size lights and retrofitted them into various fixtures around the house. I'm still juggling bulbs to see what works best. My eyes have trouble focusing on red and find blue rather harsh and irritating. Most of the survivors of my testing are "daylight".

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

Made this a long time ago:

formatting link
One low-Vce(sat) transistor for about a watt at 67% efficiency. Custom transformer, but an off the shelf 2:1 inductor would do very nicely, I think.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Nope. I've never considered any form of mental health treatment, unless you consider unloading my frustrations in Usenet newsgroups to be theraputic. Is there something that I wrote which you find offensive? Is there some reason that the you're avoiding my question(s)?

100 watt daylight equivalent is about 1600 lumens[1]. If the bulb which you purchased draws 20 watts, that's: 1600 / 20 = 80 lumens/watt Using the numbers for a Cree 100 watt equivalent bulb from: I would expect: 1650 / 16.5 = 100 lumens/watt (soft-white) 1700 / 15 = 113 lumens/watt (daylight) 80 lumens/watt seems rather inefficient, which made me suspect that either you purchased a counterfeit LED light, or your Kill-a-watt has a problem. [1] The luminous flux and power draw for a 100 watt equivalent LED varies considerably depending on vendor and source. I picked 1600 lumens for 100 watts as the most common number appearing in this mess:
--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.