BEST DIY LED fixture ever?

Chad and Sean,

just throwing this idea out there, it may not be original but folloing this thread made me think of it, It's way beyond my skill set at present, but maybe someday soon.

What if you were to mount your heatsinks on a spindle and have a gear on one end of each with a chain atached to a motor that is controlled by your controller so that not only could you vary the intensty of each string of LEDs but also adjust the angle of the whole array thoughout the day!

Seems doable to me, but I'm no engineer, you all seem way smarter than I when it comes to this stuff so maybe you've already thought of that and dismissed it for some techical reason.

Just a thought........

Chad your build is absolutely amaizing! Props!
 
LOL, it seems like I am going to have someone's significant other after me!! Although I suspect you may be right about future improvements being limited by material access and time.

Thank you very much for the comments, y'all!

Zachts: I think it is a case of "pick your poison"... What you are describing could be accomplished, I think, with mounting the light on a travelling track and having a some simple, position-based mechanism that "pushes" into a stopper and would cause the fixture to tilt a bit at both ends.

I didn't think about doing it that way, but I could certainly see it's use... especially on a larger (longer?) setup. I may just put that idea in my back pocket for future reference. Thanks!!

I say "pick your poison" because the effect can be accomplished via programming and positioning or by mechanical means... whatever you are more comfortable with! (I voted for the "harder" way for my with the positioning and programming...)

Update coming soon... I have been supoer busy with work this week and haven't had many chances to accomplish much.
 
Any verdict on how the XP emitters work with the XR-E optics? I'm having a heck of a time finding decent XP optics in the ~20-25 degree range, but XR-E optics are still available.
 
^I still haven't done the energized checks with them. I think it will be fine, although they might not be 80* at FWHM, but I don't really care about that anyway. I am only using the optics for looks and protection. I will let you know when I know for sure, though.

OK, so a bit of an update. Wow, I have been busy which means that I haven't spent a whole lot of time working on this... However, I did make some progress worth an update (IMO, it's my thread anyway!)

6912483421_fa9459b360_b.jpg


I spent a bunch of time polishing up my edges and making the fixture look much better. For those interested, I wet sanded with 600 grit, then 1200 grit, then buffed with my power drill using Ultra Cut and Ultra Finishing compounds. The finish is almost mirrored, especially on small surfaces. Larger surfaces have a little bit of distortion, but the overall effect was well worth it - look at the reflection on the top surface :)

6912484017_0e95bcfecf_b.jpg


Once the fixture was all polished up, I got back to wiring. In this picture my driver just need to be connected to the power supplies and the strings.

6912484535_6767c1b32a_b.jpg


One of the things that I enjoyed about doing this project is that I can jump from thing to thing when one gets frustrating. In this picture, I have pretty much finished up all of the non-controller wiring, but have not cleaned it up yet (I still want to be able to very easily access everything until after I do energized checks). I also started polishing up the perforated stainless steel using some Mother's mag polish and some buffing wheels... these went from raw machined SS to quite shiny... It looks nice

6912484535_6767c1b32a_b.jpg


A close-up of the shininess (doesn't come through terribly well, but I think you get the idea):

6912485249_70c4f0228b_b.jpg


Test fitting the trim pieces

6912485793_6d531e7f16_b.jpg


And installed (and dirty!! I'll clean it up really nicely when it is complete).

6912486729_12040108df_b.jpg


Well, that's that for now. I expect I am going to remain really busy for the next couple of months, but as I get some work done I will update.
 
Thank you for the compliments all! I greatly appreciate it!

Aww man, I thought you were going to let there be light.

Me too!! I wish this was progressing a little faster, but work has had me swamped recently and I have been getting about an hour a week to work on it... Not sure when it will let up, but I am steadily chipping away this and making progress on a random night or weekends. I also redesigned the layout on the protoshield for my arduino, which was a little bit of a setback... the next step is to get that all soldered up with the display, after that (and any trouble shooting I need to do) the whole shebang will be pretty much done.

that is damn impressive. a little be beyond my diy capabilities though.

It was beyond mine too, that one of the reasons I took it on. A large or complex project divided into smaller pieces makes any project conquerable... You just cannot lose the status of the big picture or how the individual portions need to progress to fit into that picture. In other words, it takes planning before doing. Also, keeping the goals you have set for yourself at the forefront. Work toward them and refer back to them frequently scope growth will kill a project faster than many other things, as it tends to cause things to spiral overwhelmingly out of control.

When I started this project, I first decided what goals I wanted to accomplish with the fixture (I think I posted them above somewhere) and then divided the design into four chunks. Each chunk being different enough that when my frustration would rise with one piece, I could move to another and not have the same problems. My chunks were:

1) Solid models and drawings: I used AutoCad, but something like google sketchup would have worked as well. These were originally conceptual, but became more refined as parts (and thus interfaces) were chosen.

2) Research and component choosing (also included sourcing): I have quite a few pages of excel spreadsheets that compared cost, functionality, and various tidbits about different components I considered.

3) Program writing: By far my weak point. I am capable at this, but by no means adept... basically if I bang my head against it long enough I can make it do what I want, but it is a long, long tedious process for me (which also made it the most frustrating).

4) Bench testing components: For some things, I just wasn't sure if it was going to work, so would order a few things and test it out. This also gave me something physical on which to test my programming.

I guess long story short, while I would consider this an advanced project, anyone who can maintain organization of a multi-faceted project and keep focus on the big picture over the length of the project (this is almost a year's worth of effort to this point) could do it.
 
Last edited:
I just purchased my first MH set but I really wanted LED's. This is really pushing me to start my year long journey into building one for myself. Just afraid :worried: of what the wife will do to me after spending as much as I did on the MH's. I will definitely be checking back here.
 
Aswome job I am going to be making an led fixture for my tank. what were the heat sinks from? looks like shelf brackets?
 
I have to agree with the OP...

I have to agree with the OP...

breaking up a project like he described is the key to successful completion of nearly any project, and I have to give him credit for stating it so clearly and eloquently. I also would love to DIY an LED fixture, and this has been an inspiration. I might drop the Gaussian curve and in it's place put a more simplistic morning, day, evening, night cycle on the tank. but either way. Impressive.
 
Back
Top