Not normally, no. The forward voltage drop of the LEDs is slightly different for each one, so one string will hog more current than the other. You may find one string will get most of the current and the other, virtually none.
DIY LEDs - The write-up
- Thread starter JohnL
- Start date
- Status
Not normally, no. The forward voltage drop of the LEDs is slightly different for each one, so one string will hog more current than the other. You may find one string will get most of the current and the other, virtually none.
CJO
New member
You can do it, but you have to bin the LEDs (measure the forward voltages for each one) and then plan your strings accordingly so that you don't run into the problem described by jusmee. Look at page 200 for a summary. Also, look at using a higher-powered power supply so that you can run both strings at 700mA.
CJ
der_wille_zur_macht
Team RC
Keep in mind that you need to consider current and voltage, not just wattage, when determining how many LEDs a particular driver can handle. It's common in our hobby applications that drivers are under-utilized because of the "difficulty" in exactly matching a certain number of LEDs to a driver's capacities.
Also it's important to note that buckpucks, despite having an input voltage limit of 32v, have an unofficial limit of 6 LEDs each (referenced in a comment in the datasheet, and confirmed by a few people who have tried to run 7 or 8 LEDs on one). The nice part is that the 6 LED capacity falls perfectly inline with cheaply available 24v power supplies, so you're maximizing cost/driven LED even though you're not technically maximizing the driver's capacities.
der_wille_zur_macht
Team RC
I think you found an exceptionally poor 60 degree optic!
It's also important to note (as we've discussed before) that taking a single PAR reading directly under a single LED is pretty much meaningless in terms of overall system efficiency, because it only tells you intensity at that single point. To get a better judge of efficiency, you'd have to determine the entire spectral distribution of light from the LED, which would require at least a few dozen carefully taken measurements (see Sanjay Joshi's writing about his light testing experiments for an example of good light testing methodologies.) It's entirely possible for a "wide" optic to have a high-intensity hotspot directly under the LED, compared to a "medium" optic, even though the overall distribution is bigger for the "wide" optic - and unless you've mapped that whole distribution, you really can't draw conclusions about efficiency. This subject has come up a few times in this thread, but the FWHM naming convention used for optics gives, at best, a very general concept of the light's distribution, NOT of it's intensity gradient across that distribution.
It's also important to note that any measurement scheme comparing different optics would pretty much HAVE to take place over a tank filled with water! One often-not-mentioned advantage of a narrow optic is that the light strikes the air/water interface at a smaller angle (to vertical). This means less light will get reflected. Hence, given all else is equal, a narrower optic should actually be expected to have a higher apparent efficiency on a real tank compared to measurements taken in air.
In general, given the choices of optics we do have, efficiency is definitely not linked to fixture height above the water. In the end what it boils down to is that the golden rule of LED builds definitely applies here - if people are going outside the norms, they should get a few different optics and experiment to determine what works in their situation. Widmer's thread is a great example of that - comparing different optics at a height not normally considered, with rather positive results. Another user did a comparison, taking measurements in his tank, using 40 and 60 degree optics, but the username is slipping my mind right now.
I was under the impression that they would mix the colors. I would email them or ask them in their forum to see if they will.
Reefobsession
New member
i have been looking into led's more and have come to conclusion that the led's are missing UV and cyan in thecolor spectrum. i now feel in order to compensate for his i am planning on adding two 36" t5's with purple fiji bulbs. i will have the two t5's running diagonally over my tabk. on alternating sides i will have 24 led's for a total of 48 3w led's. what do you guys think?
der_wille_zur_macht
Team RC
I think you should get some LEDs and try them out, and see what you think it missing, rather than trying to work it out via math. Some people think certain combinations "wash out" a tank, other people think differently. At any rate, there are cyan and UV LEDs out there, so you could incorporate those instead of T5s if you really felt there was a need - but I'd caution you that when dealing with those two colors, a little tiny tiny bit goes a very long way.
PurpleMonster
New member
Thank you all for all this info! with it i have gotten to understand all this a lot better, and well now after a few weeks from my initial post i would like to see what you guys think, themoney has been sitting in the bank waiting on me to decide on what i want to do lol.
36X30X18 right now with an upgrade to 48X30X20-22
Emitters i want:
12 XPG whites and 24 XRE Royal blues per fixture so a total of 72 LED's
My current system is all SPS and just a few LPS. Soon as i run out of room from all the SPS growing im going to get my new tank. I would like to make sure that my light will be good for when i upgrade. Im sure its fine i just want to click on the pay now button lol.
I was going to go with a Kit from Reefledlights but I want to use U "c" channel instead, this way if 36LED's are not enough I can alway add another length of u channel to add more LED's. Per fixture i'd like to do 4 channels of 9 LED's or if you guys think i will be ok with 3 channels of 12 then i will do that, again 36X30 for now and eventually 48X30.
I would like to drive the 12 XPG with 1 Meanwell D driver to control with my APEX, and run the other 24 in parallel with another Meanwell D. Or i can do 3 Meanlwells I guess it all depends on if running the blues at a lower current will be ok with my SPS.
U-Channel, from the links posted on here and other threads is all 3/4" but i see a lot of the suggestions are for 1" surface area being key! now i am not 100% on how big these stars are but is the 3/4" channel ok?
Thanks again!!
36X30X18 right now with an upgrade to 48X30X20-22
Emitters i want:
12 XPG whites and 24 XRE Royal blues per fixture so a total of 72 LED's
My current system is all SPS and just a few LPS. Soon as i run out of room from all the SPS growing im going to get my new tank. I would like to make sure that my light will be good for when i upgrade. Im sure its fine i just want to click on the pay now button lol.
I was going to go with a Kit from Reefledlights but I want to use U "c" channel instead, this way if 36LED's are not enough I can alway add another length of u channel to add more LED's. Per fixture i'd like to do 4 channels of 9 LED's or if you guys think i will be ok with 3 channels of 12 then i will do that, again 36X30 for now and eventually 48X30.
I would like to drive the 12 XPG with 1 Meanwell D driver to control with my APEX, and run the other 24 in parallel with another Meanwell D. Or i can do 3 Meanlwells I guess it all depends on if running the blues at a lower current will be ok with my SPS.
U-Channel, from the links posted on here and other threads is all 3/4" but i see a lot of the suggestions are for 1" surface area being key! now i am not 100% on how big these stars are but is the 3/4" channel ok?
Thanks again!!
wld1783
Premium Member
DWZW
The PAR measurements I have taken are from completed 48-60 LED fixtures and averaged over several points. Here is a RC link to a 60 LED Optics evaluation. True not to Sanjays work but close enough to post...
http://www.reefcentral.com/forums/showthread.php?t=1902716
DWZW you might be familiar with this.
"It's also important to note (as we've discussed before) that taking a single PAR reading directly under a single LED is pretty much meaningless in terms of overall system efficiency, because it only tells you intensity at that single point."
....Who would do this????
Anyone capable if a DIY LED setup would measure PAR across a given range of their build to compare it with the MH or T5 they are replacing.
"Another user did a comparison, taking measurements in his tank, using 40 and 60 degree optics, but the username is slipping my mind right now."
...Please post it as the only online optics evaluation I could find was done with 60, 70 and 80 degree optics commonly found online.
"It's also important to note that any measurement scheme comparing different optics would pretty much HAVE to take place over a tank filled with water!"
Why would you even consider that... Water clarity varies greatly between home reefs. One who uses extra carbon will have clearer water, and one with waves will have a different shimmer and readings than one without. Toss that into the mathematical equation...
I really dont want to argue with you but thought a different perspective on what has been done is needed...Hands On...
The collective knowledge of this forum makes it great. Hope others will post their results, and experience...not simply speculation.
Bill
TheFishMan65
New member
PrupleMonster - I would recommend the 1 inch channel.
Saraiva, you can mix LEDs in a string as long as the total voltage of the string does not exceed the driver. Also if you run parallel string on the same driver then the voltages must be matched at the desired current.
Saraiva, you can mix LEDs in a string as long as the total voltage of the string does not exceed the driver. Also if you run parallel string on the same driver then the voltages must be matched at the desired current.
bigdaddybo
New member
Good morning.
Im wiring up 12 Royal blue Cree leds. I went from Meanwell Driver ELN-60-48D Red to pos of the first led, from that leds neg to the next. So it went + - +- +- for all 12. I ended with neg on the last and connected it to the neg of driver. I turned down the volts and current POT. so not to blow anything. I power up and get nothing. If i un plug the power all 12 led light up for 1 second. its discharging?But not sure. I dont have neither Blue or White connected. I thought i my have a bad led, in the string but didnt think they all would light up and discharge if there was a bad led. Any ideas on what could be wrong or what i should try????
Thanks Bo
Im wiring up 12 Royal blue Cree leds. I went from Meanwell Driver ELN-60-48D Red to pos of the first led, from that leds neg to the next. So it went + - +- +- for all 12. I ended with neg on the last and connected it to the neg of driver. I turned down the volts and current POT. so not to blow anything. I power up and get nothing. If i un plug the power all 12 led light up for 1 second. its discharging?But not sure. I dont have neither Blue or White connected. I thought i my have a bad led, in the string but didnt think they all would light up and discharge if there was a bad led. Any ideas on what could be wrong or what i should try????
Thanks Bo
der_wille_zur_macht
Team RC
Bo, it's going to be hard for us to troubleshoot over the internet. Your configuration sounds fine, except you shouldn't ever need to touch the voltage trimpot in a typical installation (in fact, touching it can cause unexpected results, such as the driver no longer operating in a constant current mode). I don't think that's your problem, but something to keep in mind.
I would start by examining every single connection for unintentional shorts. It's very common when people say "I plugged it in and all (or some) of the LEDs didn't light up" for there to be multiple shorts to the heatsink from solder joints or sloppy wiring. Another common problem is polarity errors - people tell us they've got all the LEDs in correctly but then they come back the next day and say "oh, one of them was backwards!" In one thread we even had an instance of LEDs being put on the stars backwards by a manufacturer, so check the LEDs with a diode tester or multimeter to ensure the polarity is correct.
If you don't find any construction or polarity errors, I would first disconnect the driver from the LED string, and (carefully!) run it with no load, i.e. with nothing connected to the output. Measure the voltage across the output with a multimeter. Adjust the voltage trimpot until the voltage reads 48v. Insert the driver back into the circuit and measure again.
Yep, now I remember that thread. Given your above statement I had assumed you were testing a single LED. Sorry if my post seemed argumentative, I was just trying to make it clear why that would be a bad idea in case others were reading along, because it's commonly asked "how much PAR do you get from a single LED of model X?" which is a pretty meaningless question.
It's been referenced in this thread several times. I don't recall the username or I'd search and repost it, but I'm sure it could be found if you cared that much. The thread was originally posted in the main reef forum, so maybe try searching there.
Why would I consider that? Because I keep my corals underwater. If you do too, you MUST consider the effect that the air/water interface has on distribution and intensity underwater, at least if you're comparing different optics and different fixture heights. LEDs are inherently very directional, which is an advantage, but it also means that there's a very sharp relationship between effective PAR underwater and optic selection/fixture height. This relationship plays out in a way that when comparing higher-mounted fixtures with tighter optics to otherwise-similar low-mounted fixtures with wide optics, the results will NOT be the same as if you test in thin air. You were suggesting that a lower-mounted fixture with wider optics might be more efficient, based on tests in air - hence why I thought it was important to note this. Again though the point is somewhat moot, since I think it's pretty clear that your results were based on properties of those single optics and cannot be extrapolated to "wide vs narrow" in general.
I agree.
der_wille_zur_macht
Team RC
Well stated. I would only add that if you're mixing LEDs in a single string, then you must run the string at a current below the lowest-rated LEDs' max current rating.
In other words, if you're mixing 8 LEDs rated at 1.5A with 4 LEDs rated at 1A, you cannot run at 1.5A!
You need to have the white and blue connected to a dimming signal otherwise you will get no output.
CJO
New member
You beat me to it. I would, however, like to clarify that the dimming signal would be a 1-10V DC signal from a power supply or other voltage source.
CJ
Reef_Matrix29
Water Lover
biddaddybo, you shouldn't have to turn down the voltage pot at all, just the current pot like you have done. I take it you are testing all of this with a volt meter as well to ensure things are running within range?
+1 to wfournier and CJO, you need to have those dimming leads hooked up to your 0 -10vdc source and that source (Apex, etc) needs to be ON so that current starts to flow through your meanwells and you can then adjust your internal max current for your LED strings.
+1 to wfournier and CJO, you need to have those dimming leads hooked up to your 0 -10vdc source and that source (Apex, etc) needs to be ON so that current starts to flow through your meanwells and you can then adjust your internal max current for your LED strings.
der_wille_zur_macht
Team RC
S2minute
New member
Leds
Leds
RIGHT,so in Laymen`s Laymen`s terms, for the sake of not looking noob, which i will anyways:
2 parallel strings mixed color LEDs. EX: 1 color is an xp-g CW 3.3 VF @ 1000MA , 1500MA max. The other RB xr-e VF 3.5 @ 700MA..1000MA max.
I can do those 2 in a string as long as 1: the total VF for each led in the string does not exceed the drivers rating and...
if say the RBs max is 1000MA and the whites 1500MA, i can`t drive the strings more than 1000MA?
Or, do the different colors on the string have to be the same VF rating?,(which i don`t think is possible).
Then there`s swapen out the leds on the high n low end or just putting in resistors inline.
See, i`m getten there. I`m just taken the scenic route to do it :spin1:.
Leds
RIGHT,so in Laymen`s Laymen`s terms, for the sake of not looking noob, which i will anyways:
2 parallel strings mixed color LEDs. EX: 1 color is an xp-g CW 3.3 VF @ 1000MA , 1500MA max. The other RB xr-e VF 3.5 @ 700MA..1000MA max.
I can do those 2 in a string as long as 1: the total VF for each led in the string does not exceed the drivers rating and...
if say the RBs max is 1000MA and the whites 1500MA, i can`t drive the strings more than 1000MA?
Or, do the different colors on the string have to be the same VF rating?,(which i don`t think is possible).
Then there`s swapen out the leds on the high n low end or just putting in resistors inline.
See, i`m getten there. I`m just taken the scenic route to do it :spin1:.
- Status
