Gavin on August 5th, 2010

Your government at work producing good standards to keep the industry honest.

http://www.lightingfacts.com/Downloads/Performance_Scale.pdf

All lighting designers should be paying close attention to these standards

Tags: , , ,

Gavin on August 24th, 2011

Seminar for US Green Building Council – St Louis chapter went very well yesterday. Guess I should have posted a link to it before the event. But it was a full house anyway. Thanks Hope for organizing the event and Color Arts for hosting.

Gavin on July 4th, 2011

Why do they call it offline? You are getting power off the power line? I would think this would be on-line.

Anyway, the problem is ow to get good efficiency, power factor,and low cost in a driver that can take as input 90VAC-300VAC so it will work world wide, and output DC power, current controlled, voltage limited, able to withstand transients, overheating etc.

As with any design, what started out as a big box of parts is heading towards a single chips solution, with some passives (LRC) around it. So how few and how small are the passive components in your design? Minimize inductors first, they are the most expensive. How small can you get a 5-20W supply? It needs to fit in a lamp base until we get the last hundred years of Edison fixtures replaced.

Oh, and what about dimming? Does your PS-CC solution allow for triac controlled dimming?  Those triacs are nasty to power factor, but I have an answer! Send me an e-mail if you are interested. —  gavin at this site.

Gavin on April 13th, 2011

I’ve been working on LED lighting a lot lately. The field is still moving fast. Here’s a link for understanding CQS, my latest TLA (Three Letter Acronym)

http://www.nist.gov/pml/div685/grp05/vision_color.cfm

“A Color Quality Scale (CQS) is being developed at NIST with input from the lighting industry and the CIE (International Commission on Illumination) to address the problems of the CIE Color Rendering Index (CRI) for solid-state light sources… It will then be proposed as a new international standard.”

Tags:

Gavin on February 15th, 2011

Can’t believe it has been almost 5 months since I posted. I got samples from several flex strip manufacturers, data on hundreds more. Here’s the scoop on flex LED strips:

Quality varies considerably not all strips are the same, not even close. Price is no indicator of quality unfortunately. Some were OK, many poor quality (“waterproofing” epoxy not fully cured). Decided I need to design my own. Here’s what you need to consider when designing an LED lighting system.

1. Which LEDs are you going to use, don’t just say Cree or SSC or any “brand” there are hundreds of models of LED. Look at thermal performance as well as optical. If you don’t want to mess with lensing or much more optics than simple diffusers, choose carefully. Luminit has a really cool diffuser, holographic based design, nano-technology well applied.  Check it out if you can afford the good stuff.

2. HB LEDs (High brightness, also HP or High Power) are not usually suitable for flex boards. Too much heat to remove. Common are 0.2W SMD5050 and less than 0.1W SMD 3528. Newer SMD3535 that run 0.5 watt may be possible but you will need a good thermal analysis. Anyone out there have access to a thermal simulator? I can’t afford ComSol but it would do it. Meanwhile stick with 1/4 watt or less on flex.

3. Flex board layout for good thermal design is not difficult, but it looks nothing like the strips you get today from China. If you plan on sticking the flex strip to some metal surface, be sure it is clean… really clean, and large enough to dissipate the heat at a low temperature. Even the 3M glue will let go if it gets too warm for too long. Consider alternative backup adhesive.

4. Consider how you are going to drive the LEDs: Constant current is the best choice for a series string, but if you get too long the flex board can’t withstand the voltage across it. Beyond 7 LEDs in series (~24V) you probably need to think about rigid boards. Once you realize you need parallel circuits you have to balance current between the parallel sets. Voltage variation between LEDs and the response to temperature will lead to “current hogging” by one leg and thermal runaway unless you design in some appropriate feedback.

5. Once you’ve picked a series/parallel design and the number of LEDs to get the output you need (use a few more and don’t drive them so hard, you will get more light per watt at lower drive, look at the data sheets, that lumens per watt curve is a curve not a line. It’s called droop, but that’s too much detail for here.

6. OK so now you need a power supply capable of xx watts, see if you can find one that has appropriate approvals, ratings and protection for less than $0.50/watt. Maybe need to do your own supply too? Well that’s another expedition for another day. Stay tuned to learn about CC CV PFC and more lighting power acronyms.

Tags: , ,

Gavin on September 24th, 2010

Amazing response to my request for bids on FlexStrip LEDs. We had over 100 companies respond to initial request through Global Sources. Here is the final RFQ that I am posting so all companies can have a fair chance to bid. Please do not reply if you do not meet all requirements.

Request for quote modified Sept 24 in response to questions received from bidders.

Please reply to:  gavin@gptsllc.com   Subject: LED strips

First order will be samples to verify, then 90 or more reels (full cartons) per order

Each reel: 5 meter of flexstrip Warm white SMD5050 LEDS either 56 or 60 LEDs per meter (see below)
Waterproof not needed, indoor application – IP20 is OK, or IP65 if not more than $0.50 to $1 additional per meter
Width 10mm or 12 mm wide strips with 3M adhesive backing. Height 2-3mm is not important.
120 degree viewing angle, smooth illumination curve, no hot spots or color variation with angle.
Input voltage 24V
SMD5050 LED Warm White (3000K-4000K)  long life high brightness please specify brand and model number
Which company is applying phosphor and doing chip packaging? Yours or another – specify if not your company.
Specify nominal and maximum operating current per SMD package.
Specify range of forward voltage at given nominal current: e.g. 2.95V – 3.2V
Specify maximum operating temperature for >50,000 hours life at 70% or better remaining brightness. >45C
Specify total light in lumens per LED and lumens per meter initial and after 1000 hours
Specify brightness in mcd at 1 meter 0 degrees and 1 meter 45deg minimum, more angle data welcome
(If you have different grades of LED brightness you may submit multiple options and prices)

Samples may be 60 LED per meter  with 6 LED in series standard product (cut at 100mm)
Final order to be 56 LED per meter (8 strings of 7 LEDs each)  Cut unit length 125mm

Please price per meter for quantity 100 and 500 meter (or appropriate price breaks)
Also state sample price and shipping costs

Please note:
If you have Lighting Facts data and certification, or are working towards this goal it will be considered a good reason to work with your company on this and future projects.

http://www.lightingfacts.com/

http://www.lightingfacts.com/default.aspx?cp=content/manufacturers

Gavin on August 5th, 2010

Cephalogics has ordered an upgrade to the source boxes (LEDs) of one of the older DOT machines.

Thanks to this order, I’ll be busy for a while longer.  Thanks David!

Tags: ,

Gavin on June 24th, 2010

Show Me Solar, in cooperation with The Green Center (host site) is presenting Dan Chiras of the Evergreen Institute for two weekend workshops on PV solar power. For more details go here for the one day workshop on July 31 or here for the more advanced two day workshop on August 14-15.

The one day workshop was a great success, thank you Dan! We are looking forward to the two day intermediate workshop coming up in a week and a half. Sign up by Friday Aug 6 to get 2 free lunches.

Tags: ,

The Rochester DOT system was finished and shipped the first week of June. Delivery was promised by mid June.

Tags: ,

Gavin on May 22nd, 2010

The baby DOT system for Rochester is nearing completion.

Tags: ,