2000lumem+ per side automobile fog lights.

VanIsleDSM

Enlightened
Joined
Oct 16, 2007
Messages
649
Location
Victoria BC, Canada.
I've been planning this construction for a while.. I have all the heatsinks and power sources figured.. I've been experimenting with a Q5 WC CREE, but for these fog lights I plan to use something around 3000k. I'm looking for a very yellow 3000k, and I'm not sure how to read the chromaticity graphs with the coordinates to understand the colour differences. Will any of these CREE be a very yellow colour or is there a better high power LED choice for a yellow ~3000k white? The rebels?

I was planning to use 12 CREE per side running at 700mA+ depending on heat.. so far the white CREE I have is only reaching 32C with the heatsink and fan @ 1000mA.

http://www.cutter.com.au/products.php?cat=12&pg=2 -This is where I was going to get them, as it's the only place I've found where I can get the colour bins I want..

Are there better LEDs for my purpose? cheaper? ..The rebels are very cheap, but from futerCB you can't get a specific bin.. maybe a reseller sells them binned I dunno... help and pointers much appreciated.. I just need to decide which LEDs I want to use now...
 
Thanks, I'll check that out, hopefully his prices are good.

Can anyone guide me though as how to pick out the most yellow coloured white LED?

Does anyone know where to get the rebels in bins? they don't seem to be quite as good as the CREE, but they they can run another 300mA (although I can keep the CREE really cool, I was thinking of just taking the warm white right past 700 up to 1000mA anyway) ..but if I can buy from a select bin for less than the CREE I'd probably go for it.
 
I'm not an expert in chromaticity diagrams, but I'll give it a shot. I'm more familiar with the Luxeons so I'll use that as an example. If you look at the warm white rebels datasheet page 30, you will see that the bins give you the CCT color temperature as well as the four corners of the parallelogram that defines the spectrum as an equivalent a black body, so per that page the MP bin would give you the lowest color temperature at 2760 degrees C and therefore the most "yellow". The table below it gives you the X and Y coordinates which you can match against the full color diagram here. so if you plot the coordinates you should be able to figure out how close to the yellow region it comes.

That being said, I would raise a caution flag that the CCT (or Correlated Color Temperature) is a very rough estimate. The rebel will never look close to say a piece of iron raised to that temperature. In fact if you look at the spectrum of a warm while rebel on page 13 of the datasheet above, you'll notice, specifically in the yellow range (approximately 570-590 nm) that there is almost no yellow and that there is even less red, whereas the piece of iron at that tempreature will have plenty of yellows and reds in it, so I don't know if/how this will affect you application.

An option would be to use amber LEDs, of whom I believe the rebel to be one of if not the most efficient, but yellow and amber emitters are absolutely pizz poor in efficienty. If you compare for example a red and a white rebel side by side, you'll see what I mean.

Another option I can think of is to make a composite yellow made of green and red emiters. (See related post here). Again, I don't know if this will work for you situation, e.g. penetration in fog, etc. Just throwing out some ideas here.
 
Last edited:
Thanks, penetration in fog is exactly what I'm going for.

down around 2600k I'd expect it to start getting kind of orange... either way I don't think it'll be easy to get what I want.. I'm pretty sure all these warm white LEDs will have too much orange/red and not enough yellow... I'll experiment with some amber and white LEDs to see what I get.

but I guess the bins with the higher Y axis are more yellow, and the bins with less Y are more red/orange going by those colour charts.
 
I think you meant to post a reply to this thread:

https://www.candlepowerforums.com/threads/184391



I can see how you confused the titles... I'll wait until you do that to reply so it stays neat.. but I will quickly say that I did not use the ebay specs.. I used 75lm/w as an estimate.. figuring that would be conservative for a low power LED as I expected them to be more efficient because of the very low drive currents..

Most of the LEDs on that page are old news.. and no test of the 5chip version... There are already some heat calculations in the proper thread as well...

There is no tiny foil, quite a decent chunk of copper under the LED, the leads also help conduct heat away.. you'll see heat hasn't been a problem when you check out the replies I figure you must have missed in the right thread..
 
Top