DIY LEDs - The write-up
- Thread starter JohnL
- Start date
- Status
n0rk
More is better-er!
Theoretically no, it shouldn't... but I know personally I'd feel more comfortable if they were at the end of the string if I was doing it. But yes, it should drop the current to roughly half per emitter if you do it in parallel 
I suppose you could test it with a pair if the 1000mA-rated ones and test with a multimeter if you wanted to make absolutely sure.
I suppose you could test it with a pair if the 1000mA-rated ones and test with a multimeter if you wanted to make absolutely sure.
Iwishihadgills
New member
What do you guys think about this 9x3watt cree lamp?Maybe put a few on track lighting with 60 degree optics?
http://www.globalmarket.com/bestlamp/productdetail-12851147-0-95-1-200.html
http://www.globalmarket.com/bestlamp/productdetail-12851147-0-95-1-200.html
n0rk
More is better-er!
My two questions would be, what current do they run at and how effective is the cooling on them? Also, do we know what kind of Cree chips are in them?
Now a question of my own, something that's really been getting on my nerves. Cutter has optics designed for XP-G and XP-E series Cree emitters, for a price of $6 per unit. Now, RapidLed has optics designed for XR-E emitters for $1.25, which even including postage works out to less than 1/4 the cost of the ones from Cutter. I realise optics are designed to work specifically with a certain type of emitter... but is the introduced inefficiency by using an incorrect type optic really worth me spending more than 4x the cost? I saw the pictures earlier in this thread comparing an XP-G beam with optics designed specifically for them and ones for the XR-E and the difference didn't seem to justify the extra cost to me.
Thoughts?
widmer
Drug Enthusiast
Theoretically no, it shouldn't... but I know personally I'd feel more comfortable if they were at the end of the string if I was doing it. But yes, it should drop the current to roughly half per emitter if you do it in parallel
Assuming you're referring to the potential of one of them shorting and losing any resistance - the extra voltage would be applied immediately to the whole string regardless of their placement. But since the driver supplies a constant current, I would guess that this wouldn't be a problem..
But knowing almost nothing about electronics, I wonder if one would provide slightly less resistance than the other, so more electricity would flow through it, then it would tip the scales toward this one per se, and as more electricity flows through the one, resistance builds for the other until the one is overloaded and burns up, potentially subsequently causing the other to burn up as well. Or is this a totally irrational fear? :lol:
n0rk
More is better-er!
Oh no, I meant more from a current rectification point of view
only a minor concern, sure, but it's still a concern. The reason I say at the end of the string is, if it doesn't balance out perfectly going back to the driver it shouldn't matter, but going into another emitter you might run into dramas. Current spike will knock out both in the parallel and take the string down before anything can happen anyway, so that's not such an issue.Have you considered just running them on their own driver to eliminate any possible problems? Wouldn't need to be anything special, and you cover your *** well enough that way.
ReefEnabler
Premium Member
I just tested the voltage of a string of LEDs that had a mix of XR-E, XP-G and MC-E being run off a 700ma buckpuck.
Input voltage on the supply side measured at 25v.
Voltage across the XR-E was 3.4..... looking at the datasheet I'd guess that would be around 550-600ma
Voltage across the XP-G was 3.0.... guessing based on datasheet it's around 350ma. Hrmmm?
Voltage across the MC-E was ~12.7. Datasheet lists per-die graphs for the MC-E, but using those it looks like my MC-E is only running at only ~250ma????
Can anybody tell me why the voltage seems to be varying so much for the different LEDs? Are the buckpucks poor at driving different LEDs on the same string, or does this sound like a faulty setup?? I have 3 strings configured the same way and they all measured almost identical, so I'm pretty much ruling out bad solder jobs on the stars.
Oh and the total voltage drop SHOULD be 23.5v for each string if they were being driven at 700ma.
XP-G (3.2), XR-E (3.5), MC-E (13.6), XP-G (3.2)
So Maybe 25v is starving the pucks since I'm not quite meeting 2v headroom. Is it safe to crank the voltage on the 24v MPJA power supplies up to 25.5 or 26v??? The datasheet meantions a 10% adjustment range, so should be a 2.4v headroom, anybody object to running 2v over the 24v spec?
Input voltage on the supply side measured at 25v.
Voltage across the XR-E was 3.4..... looking at the datasheet I'd guess that would be around 550-600ma
Voltage across the XP-G was 3.0.... guessing based on datasheet it's around 350ma. Hrmmm?
Voltage across the MC-E was ~12.7. Datasheet lists per-die graphs for the MC-E, but using those it looks like my MC-E is only running at only ~250ma????
Can anybody tell me why the voltage seems to be varying so much for the different LEDs? Are the buckpucks poor at driving different LEDs on the same string, or does this sound like a faulty setup?? I have 3 strings configured the same way and they all measured almost identical, so I'm pretty much ruling out bad solder jobs on the stars.
Oh and the total voltage drop SHOULD be 23.5v for each string if they were being driven at 700ma.
XP-G (3.2), XR-E (3.5), MC-E (13.6), XP-G (3.2)
So Maybe 25v is starving the pucks since I'm not quite meeting 2v headroom. Is it safe to crank the voltage on the 24v MPJA power supplies up to 25.5 or 26v??? The datasheet meantions a 10% adjustment range, so should be a 2.4v headroom, anybody object to running 2v over the 24v spec?
Last edited:
kcress
New member
If they are in series then BY DEFINITION: They all have the same current running thru them.
Your calcs and looking at the data sheet are dishing up wrong results.
You cannot assume any particular current from a bunch of voltage drops, only the voltage drop distribution and/or the percentages of resistance distribution.
You must measure the current directly or put a fixed resistor in series and measure the voltage drop across it - which will then give the actual current in the string.
Your calcs and looking at the data sheet are dishing up wrong results.
You cannot assume any particular current from a bunch of voltage drops, only the voltage drop distribution and/or the percentages of resistance distribution.
You must measure the current directly or put a fixed resistor in series and measure the voltage drop across it - which will then give the actual current in the string.
perikaruppan
Member
another heat sink adhesive question. can we use cyanoacrylate to glue the Star to the Heat sink? we dont get the thermal adhesives here in India.
the other option is using screws on the star: but the problem that i am facing with screwing them is that the 3mm screws touch one of the + or the -ve leads on the Star base so i feel that will create a short circuit through the heat sink.
the other option is using screws on the star: but the problem that i am facing with screwing them is that the 3mm screws touch one of the + or the -ve leads on the Star base so i feel that will create a short circuit through the heat sink.
Neogenocide
New member
CA glue is not a good heat conductor, but it is more than capable of holding it to your heatsink. I use it all the time building R/C planes. As for screwing, have you looked into using nylon screws? I purchased some 6-32 nylon screws this weekend that fit snugly into the star, but is still usable.
Scott
Scott
der_wille_zur_macht
Team RC
I agree on the "no" answer for CA. Really, there are three valid options (IMHO):
1) Use thermal adhesive, i.e. the Arctic Silver brand thermal epoxies or similar products
2) Use thermal double-sided tape, including those pre-cut star pads some people are using
3) Screws with thermal paste under the star.
If you're going to use screws, either use nylon bolts (which obviously aren't conductive) or get nylon washers to put under the heads of the screws - that way, they won't create a short.
1) Use thermal adhesive, i.e. the Arctic Silver brand thermal epoxies or similar products
2) Use thermal double-sided tape, including those pre-cut star pads some people are using
3) Screws with thermal paste under the star.
If you're going to use screws, either use nylon bolts (which obviously aren't conductive) or get nylon washers to put under the heads of the screws - that way, they won't create a short.
Names Brucey
New member
Anyone have any experience or thoughts on the Luxeon Prolight 3w led's?
Wow, I don't think I've blinked in the last few days after reading most of this and the NR Ultimate LED guide.
I've currently got 4x54w T5 and 2x250w MH over my standard 75G. I keep a mixed tank, with SPS and LPS - no clams. I am looking to replace the MH with LED. The canopy is approx. 14" tall, so I have room to build 2 adjustable height LED racks I think (one over each section separated by the glass brace).
My questions:
1. Is the 24 LED kit from LED DIY kit enough to do what I want? Is it a good value?
2. What size optics (if any) should I get? LEDs will be in the middle, the T5s are 2 front and 2 back).
3. I have an AC Jr. controller - anyone know if there's a driver that's compatible?
Thanks all! I'm going to build a fuge light first as my test, this thread has been incredibly inspiring and helpful!
I've currently got 4x54w T5 and 2x250w MH over my standard 75G. I keep a mixed tank, with SPS and LPS - no clams. I am looking to replace the MH with LED. The canopy is approx. 14" tall, so I have room to build 2 adjustable height LED racks I think (one over each section separated by the glass brace).
My questions:
1. Is the 24 LED kit from LED DIY kit enough to do what I want? Is it a good value?
2. What size optics (if any) should I get? LEDs will be in the middle, the T5s are 2 front and 2 back).
3. I have an AC Jr. controller - anyone know if there's a driver that's compatible?
Thanks all! I'm going to build a fuge light first as my test, this thread has been incredibly inspiring and helpful!
der_wille_zur_macht
Team RC
Have a link to a product page or datasheet? I searched for "prolight" on Luxeon's website and nothing came up.
Have a link? I'm not familiar with a vendor called "LED DIY".
A 75 is not incredibly deep and your max. mounting height isn't hugely tall. I'd stick with medium-wide optics, depending on the mounting height you choose.
What outputs does your controller come with? Some reef controllers have "dimming" outputs or add-ons, and most are 0-10v analog output, which means the D-type ELN driver is the best choice from the commonly used drivers.
And I feel I have to ask - why keep the T5's?
Names Brucey
New member
Sorry, not a luxeon led but a company called prolight. I tried attaching the data sheet but couldn't due to size. I just did a search for prolight 3w led and you will find info. Can find them for around $3/each and seem to have pretty good specs but I am far from an expert or qualified to say.
https://secure530.sectorlink.com/led-bulbs/eShop/10Expand.asp?ProductCode=PG1N-3Lxx
https://secure530.sectorlink.com/led-bulbs/eShop/10Expand.asp?ProductCode=PG1N-3Lxx
Ooops, typing too fast - meant from RapidLED
Don't know, I think there's a separate dimmer module for their own LEDs. Wondered if anyone else knew of one.
Just got new bulbs, only a month old. I really like the color I get with T5s, and have the option of a few different colors of bulb to match my inhabitants.
If I'm sold on LEDs, then I might expand and add another bank of 24 to replace the T5s in a year or so.
That is 3.75 British Pounds. With the exchange rate it is about the same as the other LEDs being discussed.
der_wille_zur_macht
Team RC
The datasheet shows the white color to have a flux of 128 lumens "typical" at 700mA and a Vf of 3.8. That's 2.66w, so 48 lumens/watt, or about half what the typical LEDs we use will do. So they're pretty inefficient.
I've never heard of "prolight" so I wouldn't be surprised if they were just repackaging someone else's emitters - it kind of looks like a Luxeon III and has similar specs.
- Status
