DIY LEDs - The write-up - (split again)

I don't know as I don't have a lot of experience before LEDs. However it is the amount of RB that cause the major shift. I don't think going from CW to NW would change the overall look much. It just adds a little more red, I thnk people have saind that
2:1 XP-E Royal blue XP-G white is about 16-20k
 
LEDs

LEDs

My fixture is still "under construction" but, myself, i'm using a good mix of whites with XP-E RB. My ratio allows for a 2:1 XP White vs RB XP and a 3:1 RB vs XM-L .
As follows: 144 LEDs
24 White XP-Gs (12 Cool W, 12 NW)
18 XM-Ls (9 T6 Cold W), (9 T3 WW) from what i've been told the cold white is more on the blue side.
Lastly 108 XP-E RB

Now i'm sure the "looks" will be fine. Besides, i can aim each led in just about any angle i need to BUT, i'm doing it more for the health of the corals themselves. :crazy1:
 
Fishman,

this is what I had in mind, what do you think? gives me 2:1:1 ratio

C R R, C, R R , C , R, R, C, R, R, C, R R C R R C R R C R R

R R N R R N R R N R R N R R N R R N R R N R R N

R R, C, R R , C , R, R, C, R, R, C, R R C R R C R R C R R C

N R R N R R N R R N R R N R R N R R N R R N R R

the only reason for sticking with some cool whites is because I had 30 already

How do I run these now? Can I do the CW with 2 strings of 8 on one 48d driver and the 2 strings of NW on another 48d.

Is it possible to run 8 strings of 8 for the RB on the Mean Well HLG-185-42B
 
that mix would work. Let us know how you like the color.

I don't remember the exact forward voltage for all the LEDs, but 8 LEDs with a Vf of 3.5 would be 28 volts. The drivers are usually specified in a number of watts. Watts is voltage times current. You could use the drivers listed above (assuming they go low enough), but you would only be using 28 of the 42/48 volts possible. If you can find the same driver at a lower voltage you will get more current out of them.

The HLG-185-42 outputs about 4.5 amps (142/42) so each string would get about 0.5 amps. If they make and you can find a 28 volt version the output would be 6.6 amps (182/28). So you could reach 0.8 amps per string.
 
What do you think about using the one I mentioned and run 6 strings of 11. That would give me around 4.4A/6 = 733mA per string.
 
And some more info
I'm expecting them in the first half of January, but I don't have an exact date yet and things always seem to be bogged down a bit by the holidays.

For pricing, I'd expect XM-L lenses to be about $1.25-$1.50 and the XM-L NW's to be the same price as our CW's, $9.
Click to expand...
Which is a better price than cutter in small quantity.
 
Is anyone using the grid wiring suggested by MeanWell for larger parallel builds (ie where you also wire LEDs across each parallel string)? All I can find are references to Evil66, who states that it is recommended, but doesn't actually use it.

I'm looking to replace the drivers on my current setup: 5-6 drivers running strings in series with a single dimmable driver. However, I have several colors and bins mixed on different strings so I'm wondering if the grid wiring will circumvent the concern over matching the voltage of different LEDs?
 
IIMHO it circumvents the voltge difference, but you need to watch the current. Say you have 4 string at 3 amps or 750 each. Half way down an LED opens. Since all the LEDs are tied back together 3 amps is not split across three LEDs or 1 amps each. If 1 amp is the limit of the LEDs and these three are not well balanced or say you started with 3.5 amps you are goin got loose that row of LEDs. Fuses don't help unlesss you have a fuse for every LEDs since the currents keeps going back to 3 amps.

You need some sort of a rule like the current is (number of strings) - 2 times mas LED current. This allows the loss of 2 LEDs in a row before you reach the maximum current. You might want 3 if you run a whole lot of strings.
 
With the "Grid" you sort of lose your control of what's happening and you also lose any ability to measure anything. Since even an under-driven LED is too bright for a human to look at you can't assess any form of current balance that way either. Plus, it's twice the wiring hassle.

The "Grid" is best for fully matched systems going into inaccessible locations, like street lights and traffic lights.

As for your case you can run different LED types in different strings, you just have to have the same number of types in each string.
 
Fish: have you guys looked into using a PLED in parallel with each led, as a protection against a failed-open led wiping out the string. I saw this first proposed by nuclearheli in his "higher" voltage build. ( see: 210 upgrade with LED build- page 35 #846).
It (the PLED) seems to have been widely adopted by sign and lighting manufacturers for improved reliabilty. Nuke has taken a fair amount of grief over his use of "higher" DC voltages, but his gear is well thought out, and flawlessly executed. the PLED6S made by Littlefuse costs only about 50 cents each. Integrating them takes a little planning but the protection provided is surely more useful than fuses and opens up full use of parallel strings without fear of expensive failures. If you have covered this already I apologize, but it seems like the magic bullet we are looking for to protect our precious and $$$ little lites.
sorry for my ignorance of how to cite someone else's post.
 
the really sweet fix will be when Cutter or someone integrates the PLED6S right on to the MCPCB star. I bet we see this soon. Of course then they might sell fewer replacement LEDs.
 
Hi Dirrk.

Several reasons we avoid the PLEDs.

We build our own strings. Most of us don't make our builds so its a hard problem to replace a burned out LED. If a LED burns out we fix it we don't ignore it, happy there are others still pulling for us. If you know a LED is out would you just "live with it"? I don't know of any builders who would unless it's a major hardship to replace the burn-out.

PLEDs make some sense on shipped products if it can prevent expensive returns. They make much more sense if the product is very expensive to repair. Streetlights would be an example. Just getting a person up to fix a streetlight might cost many hundreds of dollars. A rational response would be to leave running - just dimmer.

Spending 50 cents a LED in our builds to add yet another component that has to somehow be soldered in, makes little sense to me.

Besides 50 cents represents the price of optics!
 
You must log in or register to reply here.

Similar threads

S
DIY canopy lift with linear actuator - complete guide
Replies
1
Views
4K
Zephrant
Zephrant
Zatoichi
FS: tons of stuff!!
Replies
8
Views
2K
Zatoichi
Zatoichi
R
LED and MH FS
Replies
7
Views
1K
robthorn
R
WmTasker
Need help with DIY full spectrum LEDs
Replies
2
Views
698
Bpb
Bpb
laverda
FS/FT Deltec AP600, Iwaki MD-70RLT, MRC 36" reef sump
2
Replies
25
Views
8K
laverda
laverda
Back
Top