thesassyindian
Supporting Member
DUDE!! This is insane!
Let’s PCB’fy this thing!!
Melted AI prime
- Thread starter rygh
- Start date
Let’s PCB’fy this thing!!
rygh
Guest
Possibly.
I have done DIY PCBs, so no problem there.
The question is how to mount the current drivers.
On a PCB is nice and clean. But thermal-glued to the main heat sink would be a big bonus.
RandyC
Supporting Member
never used these, but have you looked into Philips coral care? Fanless design.
CoralCare Gen2 LED Reef Light - Black Fixture
An exceptional LED light, worthy of your slice of the reef. The Philips CoralCare Gen2 rethinks reef lighting, combining the advantages of LEDs with the aesthetic light qualities of T5s to create a light with tunable spectrum and intensity for vibrant, healthy corals, while emulating the even...
www.bulkreefsupply.com
rygh
Guest
Definitely interesting.never used these, but have you looked into Philips coral care? Fanless design.
CoralCare Gen2 LED Reef Light - Black Fixture
An exceptional LED light, worthy of your slice of the reef. The Philips CoralCare Gen2 rethinks reef lighting, combining the advantages of LEDs with the aesthetic light qualities of T5s to create a light with tunable spectrum and intensity for vibrant, healthy corals, while emulating the even...
ATI has a fanless design that is pretty similar.
Concerns:
About 1 square foot of heat sink for a 160W fixture is a bit sketchy.
Diode spread makes me worry about disco effect.
But I think I forgot one key item on my wish list
5) Really even overall coverage.
With a single 12" fixture on a 28" wide tank, the center will get a whole lot more light than the edges.
BleepBloopMunchMunch
Supporting Member
What is the disco effect people talk about?
henergizer
Supporting Member
I'm on the AI Prime/Hydra facebook group and this is common when the UVs are set past 100%. There's a person that has been building and beta testing replacement LED pucks with similar spectrums as ATI Blue T5HO. Not sure if those have burned through the lens as well.
Is AI wiling to replace the lens?
Is AI wiling to replace the lens?
sfsuphysics
Supporting Member
planned obsolescence combined with the desire to bring manufacturing costs means you're never going to get 20 years out of a product, plus there's the limitations of the product itself I mean we as reefers run LEDs WELL outside their specs just to crank more power out of them in a smaller fixture.
That said, I'm still willing to bet it's just the violet LED simply degraded the plastic to a point where it started to brown up, and as that happens then it's a runaway effect as the plastic then absorbs even more of the LED energy and leads with that melted look. Does the fixture not even turn on anymore? I assume you double checked all your other lights too?
I'd still send a complaint at AI over this, yeah they don't have to do anything, but who knows they might surprise you and actually do something.
That said, I'm still willing to bet it's just the violet LED simply degraded the plastic to a point where it started to brown up, and as that happens then it's a runaway effect as the plastic then absorbs even more of the LED energy and leads with that melted look. Does the fixture not even turn on anymore? I assume you double checked all your other lights too?
I'd still send a complaint at AI over this, yeah they don't have to do anything, but who knows they might surprise you and actually do something.
Patio
Supporting Member
How I understand it, in simple terms- Seeing all the individual led colors/temperatures in your tank that flicker or shimmer with water movement. individual colors showing up on the sand bed.
rygh
Guest
Ripples on the water can act like lenses. Focusing point lights above to point lights on the sand.
If the light is white, you get your usual shimmer lines.
But if the light above is red, you get a red spot. If the light above is blue, you get a blue spot.
As the ripples move, those virtual lens moves, and the spots move and change color.
Kind of like dots reflecting on a disc ball, but colored.
You can also get a static color band with shadows.
A red light might be shadowed from a certain spot on the sand, but a blue light is not.
So that regions ends up with a funny color.
So people are picky about it, some do not care.
There are two fixes:
1) Cluster the different color LEDs very close together.
Creates close to a single while point.
2) Have a large number of LEDs, with diffusion plates to spread the light.
Eliminates the single light point.
rygh
Guest
It is not just the plastic. The LED itself is looking fried.
It does turn on, yes.
It is not the only light, no. I guess I missed that rather key detail.
Which is why I am not sure I care what AI does. It might actually be a mild fire risk.
Attachments
sfsuphysics
Supporting Member
So something you can do in the mean time, go into the spectral profiles for your lights and just turn those LED channels to 0% should be ok.
But I do think it's a UV thing with plastic, your picture I can completely see a "glaucoma" look to the primary lens, and the secondary lenses look frosted I'm guessing as a diffuser. It definitely looks exactly like what happened with my DIY bulbs though but only on the "UV" LEDs, none of the other LEDs went wonky.
But I do think it's a UV thing with plastic, your picture I can completely see a "glaucoma" look to the primary lens, and the secondary lenses look frosted I'm guessing as a diffuser. It definitely looks exactly like what happened with my DIY bulbs though but only on the "UV" LEDs, none of the other LEDs went wonky.
rygh
Guest
Quite possible.
As you can see from below, depending on the type of acrylic, UV transmission can really drop off.
Which yes, means it then gets nice and hot.
rygh
Guest
A rather interesting study on UV LEDs.
1) Degradation was pretty dismal.
Our lights run maybe 8 hours/day, so about 3,000 hours per year.
2) Cycling them on/off made a huge difference.
Cycle time was about 30 seconds. Keeps junction temp cool.
Since it is UV, we would not see that cycling.
1) Degradation was pretty dismal.
Our lights run maybe 8 hours/day, so about 3,000 hours per year.
2) Cycling them on/off made a huge difference.
Cycle time was about 30 seconds. Keeps junction temp cool.
Since it is UV, we would not see that cycling.
thesassyindian
Supporting Member
Aah yes, I’ve seen some tanks with straight up “Windex” water. It does make the colors pop though
RandyC
Supporting Member
There’s also this light that I forgot about. ATI Straton
www.atinorthamerica.com
ATI Straton: Connection Methods - Direct vs Network - ATI North America
There are two ATI Straton connection methods for programming. This article will share the benefits and drawbacks of each of them.
www.atinorthamerica.com
rygh
Guest
Hmm, interesting.There’s also this light that I forgot about. ATI Straton
ATI Straton: Connection Methods - Direct vs Network - ATI North America
There are two ATI Straton connection methods for programming. This article will share the benefits and drawbacks of each of them.
At 15" light spread, it is approaching wide enough for a 28" wide tank.
Nice and easy to clean, no fan surface.
It looks like both the LED current drivers and the main power supply are separate from fixture. Very smart.
Decent clustering of diodes for disco reduction.
Might be a bit of a pain to mount in my system.
Reviews on fixture heat levels are pretty good.
Afropenguin
Guest
As far as im aware, leds are rated to last for about 50,000 hours so LED lights should last around 10 ish years. The old AIs mentioned they would last over 13 with a 10 hour photoperiod https://support.aquaillumination.co...-Sol-lifespan-and-how-much-power-does-it-use-.
I have the same problems with AI prime diffuser lens getting fried but im not sure if other LED lights have the same problem.
sfsuphysics
Supporting Member
50k hours may be under ideal conditions, meaning not overdriven as most LEDs in our hobby are most definitely overdriven.
Also the LEDs themselves may last 10 years, but they do degrade over time, they'll still work, and more often than not it's some other aspect that breaks down in your fixture other than the LED itself, which could come from cooling failure which only indirectly makes the LED fail.
That said that ATI Straton, talk about sleek looking design, but having seen a review on it I give up trying to be an armchair engineer, I wouldn't think passively cooling that running at 215 watts would be doable. It ain't cheap though, $1100 or so per fixture, ouch.
ATI Straton LED Light - Comments, Review, PAR, Coverage, Discuss...
Hi All, I purchased two pre-embargo ATI Straton units for testing and feedback. The two units were replacing an 8-Bulb 48" ATI Powermodule over half of my new ~340g SPS reef. I wanted to document my findings and comparison on PAR, heat, sound (none!), look, build quality, coverage, and so on...
www.reef2reef.com
rygh
Guest
Cooling is a good question:
Power supply is probably 94% efficient. So about 11W of heat there.
But power supply is separate from fixture, and metal.
LED current drivers are also probably 95% efficient, but again, separate from fixture.
The LEDs themselves are a big unknown. I do not know brand current levels.
But let's make it simple and guess at 50% efficiency.
Meaning, 50% of the power is light, and 50% is heat.
That means roughly dealing with 90W of heat dissipation.
At 18x18, two sides, presumably aluminum, that is about 4.5 sq ft of surface area.
So 90/4.5 = 20 Watts per square foot.
That is a lot, but possible.
If the LEDs are a lot more efficient, that comes way down.
And the "215 Watts" may be an over statement.
Could be as low as 10 Watts per square foot.
But still --- it is going to be hot inside. Could be an issue for longevity.