DIY LEDs - The write-up

Status
Not open for further replies.
widmer said:
Question:

Does anyone forsee any problem with having ~15 ft of cord between my 24V DC power supply and the buck pucks? I know that on the buck puck data sheet it says that there should be a capacitor on the input side if it's greater than 18" which I can do. But is there any problem with going way beyond 18"? I was going to use standard 14 awg lamp cord...

This is pretty exciting, just got the confirmation email that LEDSupply is sending the Cree 8 degree optics, which should cast a ~12" diameter beam at 5 or so feet, which is exactly what I need at long last...
Click to expand...

Another way of asking this is if I were to connect up the 24v supply to my 15 foot cord and at the tips if I read a 24V difference on my multimeter is everything cool? I have a 6.5A supply which will only be pushing 3 amps at the most if my LEDs are on full blast.
 
Interesting, Stu. Have you measured current and confirmed it's at a reasonable value?

I'm referencing the datasheet on meanwell's website:

http://www.meanwell.com/search/eln-60/ELN-60-spec.pdf

It's interesting that the one you're looking at has Powergate logos all over it, and actually references the one I'm looking at in a note at the bottom of one of the pages, but they don't agree. :D If it's working for you, it sounds like the Powergate version is "more" correct.
 
Widmer,

Yes that will work, measure the voltage at the BuckPuck (isn't that what you are using). But measure it at the maximum current you will be using. There is a voltage drop from the long leads of the power supply.

The Drop is equal to V (voltage) = I (current * R (resistance). So if you do not have to LEDs on the current and there for drop will be zero. If I read Wikipedia right the resistance for 14 AWG is 2.5 ohm/ 1000 feet. So you should be fine. But a bad solder joint/screw connection or anything else could change thate.
 
I always set my current directly through a DMM.

The 2 , 3, & 6 LED strings were the ones that someone else blew up in a few seconds.

Hook up the 2 LED string - POOF! HMMM.... I guess I'll try that with the 3 LED string....POOF!

You guessed it 6 LEDs string POOF! for a total of 11 LEDs in about 3 minutes.

Youd have thought that after the first POOF, hed pickup the phone & call me ;-)

Anyway, I left my DMM & current measuring harness for him to adjust, he just forgot.

Once I replaced them all & set them to 700mA, they have all been running just fine for months at the LFS.

Stu
 
Widmer, this is a little beyond my comfort zone to actually answer your question, but my interpretation is that they're more worried about noise and parasitic impedance than plain old static voltage drop. Maybe one of the EE's can advise us. I'm interested too, since my tank is just so bloody big, there's going to be l o oooo ng leads all over the place.

edit - I got db errors for a good 15 minutes trying to make this post, and I see now others have chimed in and made it a little out of date. So pretend this was right after widmer's last post. :D
 
L98-Z said:
I really doubt there is anything to worry about with only 15'.
Click to expand...

Do you have a supporting argument? The manufacturer is "worried" enough to suggest a cap on the input if it's more than 18". Suggesting a cap makes me think they are worried about dynamics, not a static voltage drop due to resistance in the wire.

But as I said above, I'm not an EE. :D
 
A capacitor will not help with a static drop in the wire, so yes the capacitor is for noise. The problem I was alluding to is if there is too much drop (probably not, but if there is) you will not have enough voltage to drive the buckpuck. Is there a reason you don't want the capacitor. Just follow their directions.
 
Will I have problems using xp-g and xr-e royal blue on the same string?

The reason i'm thinking about mixing them is I was originally going to try to mimic AI leds and do 4 banks of 24 but i decided that was too much and downsized my plan to 72 total, 2 banks of 36. Thing is, I will have 18 of each LED on each side and I'm planning on using meanwell drivers. All that means is i'll have a string of 12 blue another of 12 white and a mixed string 6 and 6 of each. The Vfs add up to within range so i shouldn't have problems right? Just want to be sure.

ALSO, I don't yet have a multimeter. My small starter projects have all been with buckpucks so I have been able to get by without it but since I'm going with meanwells for the big one, I'll need a multimeter to set them up to not fry the LEDs. Can someone recommend a good one?

Edit: sorry i'm sure this was asked before...
 
Skeptic,

You may already know, but also DO NOT exceed the current of the lowest rated LED. It used to be some the of LEDs were rated differently. I don't know if that is still true.
 
sbleile said:
I was wondering if anyone has used the XP-E in royal blue?
Click to expand...

I haven't personally but know that several other folks have. Similar absolute output and efficiency to XR-E but a WIDE viewing angle (30% more width than RB XR-E) so the light will be more spread out. I wouldn't want to use them in a fixture mounted more than a few inches off the surface without optics.
 
TheFishMan65 said:
Widmer,

Yes that will work, measure the voltage at the BuckPuck (isn't that what you are using). But measure it at the maximum current you will be using. There is a voltage drop from the long leads of the power supply.

The Drop is equal to V (voltage) = I (current * R (resistance). So if you do not have to LEDs on the current and there for drop will be zero. If I read Wikipedia right the resistance for 14 AWG is 2.5 ohm/ 1000 feet. So you should be fine. But a bad solder joint/screw connection or anything else could change thate.
Click to expand...

Excellent response. That's very reassuring. While I've got your attention, do you have any idea why the buck puck data sheet specifies that a capacitor must be used across the buck put power-input terminals if the power supply is more than 18" from the buck puck?

PS so if I understand you correctly, I'll be running ~3 amps through the cord so is it (3 amps) * (15 feet * 0.0025 ohms/ft) = ~0.11 V drop? Or in the case of a single buck puck it would be (1 amp) * (15 feet * 0.0025 ohms/ft) = ~0.0375 V drop?

I should have more than voltage to spare if I'm running 6 LEDs per string at 3.7V/LED = 22.2 V necessary to push them vs a 24V supply correct?
 
Last edited:
NO

But I will give you my opinions. Some electronics are sensitve. Every IC on a circuit used to have (and may still) have a by pass capitor between power and ground as close to the chip oas possible. Yes every chip and some more than one. So it could be that there inside circuitry is sensitive.

Option B. They are trying to regulate the current. If there is inconsistencies on the inoput it makes it harder to regulate to output. Imagine trying to keep a hose constant why someone is trunig it up down. They may need the extra insurance to stay within specification.

Every wire is a transmission line and an antennae. We just never know what they will pick up.
 
Aluminum bar as heat sinks

Aluminum bar as heat sinks

First, thanks to all the pioneers on this and the other LED DIY threads for all of the ideas!

I'm planning to switch the VHO (6 X 160 watts) system on my 135 (72" X 18" X 24") to and LED system. I keep mostly SPS along with a few LPS and softies. I plan to use around 48 XRE cool whites and about 48 royal blues powered by dimmable Meanwells. I have a nice oak canopy I wish to keep. The down side is it only gives me only 4" clearance above the tank and about 4.5-5" max above the waterline. A stacked configuration of fans, heat sink, LEDs, optics (maybe) and splash guard will be doable but a tight fit.

Instead of the traditional finned heat sinks, I'm thinking of using 72" long aluminum bar or channel stock. I'll likely use three or four 72" sections. Also, rather than dedicating individual fan(s) to a heat sink, I will use a number of fans in the back of the canopy to generate "œgood" airflow over the tops and bottoms of the aluminum bar or channel "œheat sinks".

I also plan to be able to move the individual bars forward as needed along with the capability to tilt the bars forward or back allowing me to fine tune light distribution. I'll place optics on the front and side LEDs.

The bar stock is available from a number of vendors in a large number of sizes. I ordered a some 3 foot sections from onlinemetals (great selection and inexpensive). I now have a 2" X 0.25" and a 2" X 0.375" bar along with a 1" X 1" X 0.125" thick channel.

I'm a bit worried that I could have some heat issues. So I'm going to make a small prototype. I have 12 XREs along with a Meanwell and a selection of optics and I plan to put together a 3 foot section and do some temperature measurements at different drive currents and at different airflows. I'll also place it over a section of my tank just to get an idea of the "œlooks". I will drill a tiny hole through the bar where the center of one star will be. I will then insert a 0.5mm temperature probe along with some arctic silver into the hole and in contact with the back of the star. While this will not be a true junction temperature, I hope it will give me some idea of the heat generated. I'll post results in a few weeks.
Has anyone seen any "œback of star" temperatures with setups using finned heat sinks with fans?

Any comments or suggestions would be appreciated.
 
Status
Not open for further replies.

Similar threads

R
DIY AI-coded ESP32 based aquarium controller
Replies
0
Views
309
ronnyking
R
AquaticReefTV
DIY Cable management time lapse
Replies
1
Views
531
S|\/|S
S
AZBigJohn
New Build... I'm back
Replies
15
Views
447
Dustin07
D
B
Beam Deflection - 15 years and I finally broke down and wrote a how to.
Replies
16
Views
916
BeanAnimal
B
A
Fleece filter bypass problems (Innovitech X filter)
Replies
1
Views
42
Vinny Kreyling
Vinny Kreyling
Back
Top