Almost Shopping Time.... [DIY LED]

Hi guys

As a few of you might have seen around the place, i'm gearing up to put together my next panel. Sourced a hefty 200mm x 200mm x 75mm alu heatsink for next to nothing. It'll be turned into a ~80W panel, maybe less than that.

It will be active cooled, probably by a 120mm fan running as slow as possible (which will be slow) for less noise.

I'll be running mostly WW with some red, deep red supplement.

The WW will be Cree XTE - XTEAWT-00-0000-00000BEE7 . the 3000k, 70 CRI bin. i considered oslons but decided i wanted to make the most of the cooling i had available, while keeping the number of modules I had to buy down (they are run at ~700ma). The XTE are rated for 1.5A, Though i will run them at 1.05A most likely. I also really like the 85C binning.

Sooo....

The view angle on the XTE is 110 degrees.

They will be running at ~1 - 1.1 amps.

Now onto the reds and deep reds... my main puzzle here has been finding coloured diodes with similar properties.

My first thought was to use up some of the 20 or so Ledengin deep reds i have. They run at the same amperage, but only have a 90 deg viewing angle. If anything, i'd want the reds and deep reds to have a wider viewing angle than the WWs.

So what i need help with, is some suggestions for coloured diodes that would match the XTEs nicely . by match up, i specifically refer to viewing angle (110 degrees or even more) and amperage (around 1A). What would you guys recommend?

I wouldn't know of a red driven at 1A! The ledengin you have?

Couldn't you have a separate driver for the reds (between 500 and 600 mA or 700 if you like)
In that case you have good rebels or osrams for the hyper reds and cree xpe for the reds.

But you prob know this?

Widest viewing angle is golden dragon plus 170°
Oslons 150°
Xpe 130°

interested to see what you come up with. are you gonna get dimmable drivers?

The best hyper/deep red to date is rebel ES 360mW @350! (Highest bin) At least on paper...
Its The One I would possibly drive harder that 600mA .. Look at data sheets there's a few different points compared to other makes. View angle would be 100-120

The Ledengin deep reds are rated for uhh.... 1.2A i think?

LZ1's. Gaming ATM but when i have a moment ill dig up the data sheet

im driving a bunch at 1.05A right now.

http://uk.mouser.com/new/ledengin/ledenginLZ1emitters/

the 660nm datasheet. http://www.ledengin.com/files/products/LZ1/LZ1-00R200.pdf

edit: i forgot to mention that one thing that i like about the LZ1, which is shared with the xte (makes it even more of a shame that i dont like the viewing angle... ) - they wont blow up if a fan fails and the heatsink reaches boiling point. (rated for 1.2A at 100C )

jubiare said:

Couldn't you have a separate driver for the reds (between 500 and 600 mA or 700 if you like)
In that case you have good rebels or osrams for the hyper reds and cree xpe for the reds.

Click to expand...

i really want to avoid any significant discrepancy in the power of the diodes, as it would annoy me having to back off or put the panel closer for the benefit of one colour at the detriment of others.

Kill a Watt said:

interested to see what you come up with. are you gonna get dimmable drivers?

Click to expand...

I haven't decided yet. If i did, it would be with a discrete scale.. ("clicks" rather than a smooth curve) This is for repeatability and side by side comparison. That said, i would probably be just as happy using the inverse square law as my dimmer

Still looking for suggestions guys, right now i think i may need to suck it up and use the ledengins...

You are not going to experience what you think you are with driving at different currents, what you say it's more to do with lenses degree angle.

Also, driving the xte hard is making them loose greater efficiency than one thinks from the datasheets. They are still drivable hard but maybe you want to reconsider and:
Drive the whole lot at 700mA, I can see you want to employ loads of WW, you could skip the reds which is all pretty much covered by the WW and add the deep reds, perhaps those rebels that looks like are all good at 700mA.

At that point I don't know about fewest blue too, or some cool/neutrals ..
Just coz I don't know how much blue you'd have from the WW (from the graphs, for what an indication is, looks like there's little blue)

Maybe!
Good luck anyhow!

This is all a bit theoretical, but not without any foundation. People are currently experimenting with this skipping the reds for the WW

You are not going to experience what you think you are with driving at different currents, what you say it's more to do with lenses degree angle.

Also, driving the xte hard is making them loose greater efficiency than one thinks from the datasheets. They are still drivable hard but maybe you want to reconsider and:
Drive the whole lot at 700mA, I can see you want to employ loads of WW, you could skip the reds which is all pretty much covered by the WW and add the deep reds, perhaps those rebels that looks like are all good at 700mA.

At that point I don't know about fewest blue too, or some cool/neutrals ..
Just coz I don't know how much blue you'd have from the WW (from the graphs, for what an indication is, looks like there's little blue)

Maybe!
Good luck anyhow!

This is all a bit theoretical, but not without any foundation. People are currently experimenting with this skipping the reds for the WW

I totally agree about the reds being made a little redundant by WW. I will still include a few, but it won't be many.

Im also undecided about NW/CW. In my experience you really do not need much blue at all, and if all the WW are giving out a little, this is probably enough.

If i run at 700ma I'll probably use all oslons (WW and coloured) for the beam angle and for general uniformity (not that it really matters) I'll certainly have a good look at the Rebels before i buy anything, though.

A lot of my concern about beam angle is because of the small size of the heatsink (20x20cm) - I want the panel footprint to cover double that and as uniformly as is practical. Obviously because of the relation between intensity and beam angle this is not too easy, and i imagine will be harder to get good blending and coverage if the diodes are different (have different beam angles i mean)