DIY LEDs - The write-up

Status
Not open for further replies.
Here is my weekend project:
24 XPG whites, 24 XPE Royals and 48 individual heatsinks:
normal_40B_044.jpg


To save thermal epoxy I put a small dab on one heatsink from part A, same amount from part B, mix well, use what's on the mixing spoon for a new heatsink, scrape most from the mixed solution for another heatsink, use whatever is left for the heatsink used for mixing (so 3 heatsinks with one mixing):
normal_40B_046.jpg


U aluminum for a frame using #4-40 machine screws (the heatsinks tap very easily for these screws)
normal_40B_054.jpg


First of two fixtures ready (12/12):
normal_40B_057.jpg



3/4" ply used for a frame around the aluminum, mounted on tv articulating arms (ignore the mess around the tank :) )
normal_40B_060.jpg



EDIT: almost forgot, 25 degree lenses on the outside, 40 degrees on the inside mounted with hot glue :)
 
I like that terahz! How warm do the heatsinks get? Are they going to need fans?

Nice clean build powdr_dayz! What does the fan in the middle do for you?
 
I like that terahz! How warm do the heatsinks get? Are they going to need fans?

No fans for now. I've only had the lights on for about 10 mins with the heatsinks on the carpet. The max temperature I got was 90F (with 79F ambient).

My 10G nano has 12 XREs at .7A and .9A closed in a hood and they don't get above 110F with the fans off. Fans on they stay at ~ 85F.+



For the record, each of the fixtures (wood and alu included) is less than 3lbs.
 
Nice clean build powdr_dayz! What does the fan in the middle do for you?

I was having a reoccurring problem of blowning the 1Amp Buckpucks. I designed this fixture so that the Buckpucks were in the same box as the 2 x 24V power supplies (for simplicity reasons) and suspected that the heat generated by the supplies was cause for the failure. I added the fan to the box and now the power supplies and Buckpucks remain at a constant temperature of 33 deg. C. It's been working well for over a month now...
 
CRee XPG 7up from cutter

CRee XPG 7up from cutter

I ordered two of these to go over a frag tank with 40 degree optics. Should I use one driver like the lpc 60-1050 and parallel them or put each on its own driver????

PL112xx.jpg
 
If you parallel them they will each only get about 500 ma. You could do that if that is all you want to drive them at. If one string did go out the other string would just brighten and still be in a legal current range (make sure the heat sink can take the heat)
 
Skippy, for the record, can you share some specs on those?

Where you got them from?
What color and bin of LEDs?
How much they cost?

I'm guessing they came from Cutter but I'm still curious to know details since I don't know of anyone else who has used their larger multi-emitter MCPCBs.
 
I suspected it was from Cutter. But they don't provide bin information specifically for the 7-up MCPCBs so I was curious if the other poster had received that info. The 3 x XP-E RB/4 x XP-G CW is an attractive price, but kind of the wrong ratio (I'd rather have four blue and three white) and without bin information, it's not quite attractive enough to risk it, for me at least. Plus, the optics work out to around $2/LED which is a bit pricey, and the 7-up config makes for difficult math when trying to match them with our typical drivers.
 
I have RB XRE left over from the last build that I am going to use on this as well. I think I can even out the math by doing a 7up + 5 RB per driver. I think I will use the the constant current 700ma meanwells for this little project.
 
heya Mike!

Wouldn't a frag tank be okay with the color ratio presented there? I mean.. the point is for them to grow, not look pretty right?
 
Regarding the dimmable meanwell, can I adjust the current by using say a 10 ohm resistor and measure the voltage drop and use that to calculate and adjust the current? If so, I can avoid measuring the current in line with the LEDS string. Besides, my mutlimeter can only measure DC current up to 200mA.
 
You can, but you need more like a 1 ohm. 10 ohms at 700 ma would drop 7 volts which might kick the driver into a voltage limiting mode. Measure the resistor first so the calculations are correct.
 
You can, but you need more like a 1 ohm. 10 ohms at 700 ma would drop 7 volts which might kick the driver into a voltage limiting mode. Measure the resistor first so the calculations are correct.

I don't understand. 1 ohm even with 1 A will only give me a 1V drop and the minimal voltage from the ELN-60-48 is 24V. Shouldn't I be using a higer resistor at least a 25 ohm resistor and that will give me a voltage drop of 25V.
The resistor is going to be only thing in the test circuit without the leds.
 
swannking, you'd need one heck of a resistor to give you enough voltage drop to be in the "official" spec'd range of the driver with nothing else in the circuit! IMHO you'd be better off with the approach FishMan assumed, which is using a smaller resistor with (at least a few) LEDs in the circuit. Start by turning the current limit pot all the way down to ensure your measurements begin at a safe current level.

Or, just go spend $15 on a multimeter that can handle current measurement in the range you need! By the time you buy a power resistor with a high enough rating you'll be halfway there!
 
Yes as dwzm says I thought you were using as a sense resistor with the LEDs still in series with it. If you do use a resistor only make sure it is over 25 watts and it will probably still get hot.
 
I thought I directed this to Powdr_dayz, hence the edit. I'm a noob.
Cool!
I'm in Helena we need to get together! I have followed this post since Soundwave entered his project. It took me forever to learn how to subscribe to the thread- i just kept bookmarking new entries daily so i could find the latest post.
This is my first post, as I have just been lurking in the shadows learning from the boys!
I'm about to build an array for a 500 gallon aquarium in my restaurant, and i have a 120 gallon aquarium at home for testing.
I'll pm you.
 
Last edited:
For those of you using the dimmable meanwell, can you measure the current draw by the dimming circuit at different dimming voltage? I got answer from the co. stating 20mA is the minimal requirement but some stated otherwise. It will be helpful for those interested in using either the aquasurf or Apex dimming function to control multiple meanwells using a single channel. (max. of 20mA per channel according to Curt) I am hoping that I can control at least 2 meanwells per channel running them in parallel.
 
Status
Not open for further replies.
Back
Top