LED breakthroughs

I was sort of bummed to read:

The nano-crystalline coating is very expensive and difficult to produce, and, so far, there aren't a lot of ideas as to how to mass produce these things.

But still, that's good news. I'm still looking for a good, energy efficient  lighting source that produces fairly white  (or at least natural) light.

300 lumens per watt seems a bit high given that most sources I've seen say that for white light, 242.5 lumens is the most possible.  Now if you restrict the frequency of the light to that which human eyes are most sensitive (yellowish-green), 683 lumens per watt is apparently possible.

A quick search revealed this one source that corroborates the above numbers: http://members.misty.com/don/bulb1.html

For now I am waiting for the C. Crane GeoBulb, due out later this month, that produces 800 lumens from less than 8 watts, the first LED lamp I've seen that exceeds 100 lumens/watt.

Hopefully I'll be able to report my findings here in a couple of weeks when I get my hands on one or two of them.

I just checked the C. Crane website and the GeoBulb is now out of stock until 6/30, moved from 5/14.  So I'll be waiting another month and a half for this bulb to be available.

I just checked the C. Crane website again and was disappointed to see that the availability of the GeoBulb has slipped another month to 7/31.

I guess this is giving the competition a chance to catch up to them . . .

Anyone know what chip they are using? I know Cree has a chip in development that's around 300 lumens per watt.

Check my first post in this thread; as far as I can tell the upper limit for the efficacy of generating white light from electricity is 242 lumens/watt, I've seen that number claimed in various sources.  If you have information to the contrary I'd love to read about it.

As far as what emitters are being used in the GeoBulb, that I do not know.

Apparently that 242 lumens/watt figure is for more full-spectrum white light as you would get from a theoretical black-body radiator; white LEDs usually produce so-called dichromatic white light (blue plus yellow typically) and I see now that there is more potential for high efficacy in that situation, at least according to this source:

http://members.misty.com/don/lfunfac2.html#ole

Maximum theoretical efficacy numbers approaching 500 lumens/watt are claimed near the bottom of that page.

Regarding the backorders - is there some type of production problem / delay, or is this really due to heavy order levels?

Here is the reply I got from C. Crane Customer Service when I asked about the slip from 5/14 to 6/30:

"Thank you for taking the time to contact our company.  The bulbs are in the process of being manufactured in small batches.  We are very confident about the date of June 30th however it is still tentative.

If you have additional questions or we can be of further assistance please email or call our toll free number."

and here is the reply about the slip from 6/30 to 7/31:

"We have missed our scheduled delivery date of the GeoBulb. We appreciate your patience and hope it has not caused unnecessary frustration.

We have made excellent progress in the last few weeks on the last few items to be resolved. UL and FCC listings should be complete in a few weeks and allow production to proceed. We hope backorders will be filled by 7/31/08 but it is possible it might extend a few weeks after that.

We will be posting additional information on our website as well as sending out an email to our customers with GeoBulb updates as soon as we receive them.  To sign up for the updates please sign up https://www.ccrane.com/ssl/subscribe.aspx

If we can be of further assistance please either email or give us a call."

So the problem would appear to be production-holdup related.

Ouch...that's a bit frustrating. Keep us posted for sure. Thanks Bob!

Sounds frustrating indeed.  ...I just saw red flags regarding production when it slipped as far as it did - as your specific order should never of pushed that far just for ordering levels as it would catch up to you sooner than that, with orders after you slipping. 

Hope you get the bulb(s) soon... can't wait to hear how they perform!

Actually I'm waiting until it says they are in stock before I place an order.  I suppose that might delay my getting one to some degree.  But I saw no indication that they would not charge me until it shipped, and I'd rather they didn't have my money while they slipped the availability by weeks at a time.

Well after further digging around on their website I now see that it does say that I will not be charged for backordered items until they ship:

http://www.ccrane.com/backorders2.aspx

But I have had problems in the past with companies disregarding their own policies and you usually have to push pretty hard to get the charges reversed in such a situation.  So for now anyway, I'll keep waiting for the GeoBulb to be in stock before I place an order.

Right up until yesterday the GeoBulb still had the 7/31 delivery date, now today I checked it and the date has slipped (for a third time) to 9/1.  I'm finding this situation difficult to understand.

Here is their reply to my contacting them about this latest slip:

Odd that you should write now, I was thinking of dropping you a line this morning.  We just received word yesterday evening that our latest samples in testing have passed FCC regulations.  As far as I know we are the first company to do so for a LED light bulb of our style. This should be our last big hurdle before mass production starts. We should have some completed units here near the end of August. Production quantities should follow soon after.  Keep up the faith, the future is here.

If you have additional questions or we can be of further assistance please email or call our toll free number.

I just checked again and the date has now slipped to 9/30.

You are a patient man, Bobkart... wow.

I recently ordered 2 CL-5 bulbs from EarthLED.  Will let you know how they are when I recieve them... 

Thanks for the update, stins.  That picture is helpful.

Their website still says 9/30, just a couple of days away.  If they know it won't happen by then they really should put a more accurate date up there.  Last-minute slips are far less understandable than ones disclosed well in advance of the missed date.

A while back I checked the specifications of the GeoBulb and they had changed from ~800 lumens to something less clear: "381 - 0° to 90° vs 337 for incandescent".  I initiated another exchange with the Customer Support contact I have been dealing with (John Wilder).  I expressed my concern that if this new LED lamp only emits 381 lumens from the top half of the lamp then there is no chance of it emitting 800 lumens in total, more like 500-600 lumens, a drastic setback from the earlier claims of 100 lumens per watt (down to 60-80).  Unfortunately he doesn't create the specifications but only reports them (in fact he had yet to see the revision until I pointed it out), so there's not a lot he can do but pass along my concerns.  The GeoBulb does still have good omnidirectionality going for it, that was the second thing that interested me in it, after the 100 lumens per watt.

But that "vs 337 for incandescent" part makes no sense, any standard-shaped bulb putting out 800 lumens will have at least half of that coming from the top half (maybe 500 from the top and 300 from the bottom?).  I tried to get to the bottom of what that part was all about but his response did not address it.

In response to all of the "fudged" lumen specifications I have been running into over the last several months, I've undertaken the construction of an Integrating Sphere.  For anyone that doesn't know what an Integrating Sphere is, it can be used to measure the total luminous output of a light source:

en.wikipedia.org/wiki/Integrating_sphere

Mine won't compete in terms of accuracy with those used in laboratories and costing tens of thousands of dollars, but my hope is that it can be used successfully to compare incandescent/fluorescent/LED lamps in terms of total light output.  I'm just waiting to hear back from the source of the highly-reflective white paint I will apply to the interior of the sphere, so I can place an order.  Here is a picture of it without the paint:

bobkart.gt3times.com/Integrating-Sphere.jpg

Wow Bobkart - very cool stuff!!!  Keep us posted as that device, even tho not a laboratory type - will likely give some reasonable number results for lumen output and would be incredibly helpful as a neutral third-party test... I can't help but think how the users on this site could benefit from it :-) 

LEDs are the most efficient when your power source is DC, ie. a battery, like on a car, truck, boat, or flashlight, but they are not competitive for efficiency when your power source is AC.   With an AC power source, fluorescents or CFLs are still more efficient.  It will take some major breakthroughs to change this.

In my experience the most efficient (AC powered) LED lamps are currently somewhat more efficient than the most efficient CFLs.  I'm seeing efficacy numbers of 70-75 lumens per watt for CFLs and 90-100 lumens per watt for LED lamps.  These are high-end numbers and admittedly it can be hard to find an LED lamp this efficacious, most are more like 50-75 lumens per watt.  Tubular fluorescent lamps are much closer and can have efficacies as high as 90-100 lumens per watt, but that doesn't consider ballast overhead, which brings it back to more like that of CFLs, depending on the type of ballast (magnetic ballasts add about 25% overhead, electronic ballasts add about 10%).  For this reason I won't be changing out any tubular fluorescent lamps for LED lamps any time soon, except perhaps in one fixture in my garage that appears to have a bad ballast.

Back to the GeoBulb, they have updated their webpage(s):

www.geobulb.com/products/light-bulbs/

Luminous output specifications have dropped drastically, to 446 lumens; it was 800 initially (almost a year ago now).  Availability has been pushed out to February 1.  Not much more to say about that other than "disappointing".

 Am I reading the website right?  For 8 units it is $119.95?

Each 'light bulb' is $120.  You may be looking at the specifications where it says that it has 8 LEDs.  There are 8 LEDs in each 'bulb' according to that specification.

$15/bulb would be an awesome price though!

(Hey seattlite, I'm across the lake in Bothell.)